Electric Vehicles

ELECTRIC VEHICLES
Ready(ing) for Adoption
Citi GPS: Global Perspectives & Solutions

June 2018
Citi is one of the world’s largest financial institutions, operating in all major established and emerging markets. Across these world markets, our employees conduct
an ongoing multi-disciplinary conversation – accessing information, analyzing data, developing insights, and formulating advice. As our premier thought leadership
product, Citi GPS is designed to help our readers navigate the global economy’s most demanding challenges and to anticipate future themes and trends in a fast-changing and
interconnected world. Citi GPS accesses the best elements of our global conversation and harvests the thought leadership of a wide range of senior professionals
across our firm. This is not a research report and does not constitute advice on investments or a solicitations to buy or sell any financial instruments.
For more information on Citi GPS, please visit our website at www.citi.com/citigps.
Citi GPS: Global Perspectives & Solutions June 2018
Raghav Gupta-Chaudhary is currently the European Autos Analyst. He has been an Analyst for seven years
and joined Citi's London office in July 2016 to cover European Auto Parts. Raghav previously worked at
Nomura from 2011 to 2016, where he started off on the Food Retail team and later transitioned to cover the
Automotive sector. He has an honours degree in Mathematics with Management Studies from UCL and is a
qualified chartered accountant.
+44-20-7986-2358 | [email protected]
Gabriel M Adler is a Senior Associate in the Citi Research European Autos team. He is currently based in the
London office and started with Citi in October 2017. He is a qualified accountant and has a BA in Politics from
the University of Cambridge. Gabriel previously worked as an auditor at EY.
+44-20-7986-8562 | [email protected]
Dr. Menahem Anderman PhD. Jeff Chung

Total Battery Consulting, Inc. China Auto & Auto Parts Analyst
+852-2501-2787 | [email protected]
Itay Michaeli Arifumi Yoshida

U.S. Auto & Auto Parts Analyst Japan Auto & Auto Parts Analyst
+1-212-816-4557 | [email protected] +81-3-6776-4610 | [email protected]
Ethan Kim
Korea Auto & Auto Parts Analyst
+82-2-3705-0747 | [email protected]
Contributors Kota Ezawa Charlie Grieg

Mohammed Zaheer Ana Chikhikvacze
Christian Wetherbee Tim Thein
June 2018 Citi GPS: Global Perspectives & Solutions 3
ELECTRIC VEHICLES
Ready(ing) For Adoption
Kathleen Boyle, CFA What do you think of when you think of the ‘Car of the Future’? The first thing that
Managing Editor, Citi GPS springs to mind for me is from the cartoon The Jetsons where George gets into his
flying car and then folds it up and puts it in his briefcase before he gets to his desk.
While that might be a bit further out in the future than is reasonable, what you
consistently do not see in most futuristic stories is someone stopping on the side of
the road to fill up their gas tank.
Electric vehicles as a concept have been around since the very beginning of auto
invention. But it’s only in the past few years that the technology has caught up with
the concept. Cars that can drive for long distances without a charge are starting to
be produced and demand for electric vehicles has slowly started to increase but
why hasn’t demand for them taken off? In the pages that follow, the authors look at
where we are now in terms of battery electric vehicle adoption, where we’re likely
going, and how long they think it will take for us to get there.
There are four main barriers to adoption for consumers to fully embrace battery
electric vehicles — range, infrastructure, battery degradation, and cost. Basically
consumers want to know if the car will get them to where they want to go, can they
plug it in when they need to recharge it, will the battery last long enough and is it
cheaper than a regular gas/diesel passenger car. On most of these fronts, we find
that we’re getting closer to ‘yes’, but just aren’t quite there yet.
So why are people buying battery electric vehicles at all? When we look at current
demand, we find that we’re still in the early stages of EV adoption with just 1% of
battery electric vehicles and plug-in hybrids making up just 1% of new car sales
globally. Demand up until now has been driven primarily by ‘push’ factors such as
government support through the use of taxes and incentives. Regulations on
emissions are used to make electric vehicles cheaper and traditional internal
combustion engines more expensive to meet emission targets.
In order for penetration of battery electric vehicles to truly increase, demand ‘pull’
factors will need to take over. Consumers need to believe that the utility of a battery
electric vehicle is higher than for a conventional engine vehicle. For this to happen,
the price of batteries has to come down enough so that cost parity with traditional
vehicles is reached. Lower battery costs will also extend the range of battery electric
vehicles and newer technologies will solve the problem of battery degradation.
Finally, a network of charging stations will need to be installed to alleviate the ‘plug-
in’ fear.
The timing and scale of electric vehicle adoption is open to debate and the range of
estimates for penetration in 2030 is pretty wide. Our forecasts are based largely on
push demand in the sense that we make no assumption around consumers actively
pursuing battery electric vehicles over and above traditional vehicles. Our base
case scenario is for 10% battery electric vehicle penetration by 2030 with Europe
and China remaining the largest EV markets. Our bull case scenario lifts penetration
to 18% in 2030 and assumes both European emissions targets will be met and
China new energy vehicle targets will be reached. Our bear case at 5% penetration
sees a more significant miss on both targets.
We firmly believe the future is electric. The question today, as it was in the past,
remains when.
© 2018 Citigroup
The Future is Electric,
the Question is When
THERE IS A LOT OF TALK AND INTEREST AROUND ELECTRIC VEHICLES (EVs)
BUT EVs ARE STILL A MINOR PROPORTION OF AUTO SALES
...but battery 2017 2014

Battery Electric electric vehicles
Vehicles is now the still make up
#1 ISSUE
on Automobile <1%
Executives minds, of global new
up from #10 in 2014... car sales...
0.8% vs. 0.2%
(Source: KPMG Global Automobile Executive Survey) (Source: Citi Research, LMC)
CURRENT DEMAND FOR EVs IS DRIVEN PRIMARILY BY ‘PUSH’

(I.E. INCENTIVES) VERSUS ‘PULL FACTORS’ (I.E. CONSUMER DEMAND)
EV’s highly sensitive to changes to incentives: Push factors
2016 Sales impact from:
Incentive Decrease Incentive Increase
Denmark Netherlands Norway
(BEV & PHEV) (PHEV) (PHEV)
-60%
-50% 164%
What are the main reasons that you would NOT buy an EV?
Barriers to Entry:
35%
30%
Price Charging Infrastructure Range
25%
20%
15%
10%
5%
0%
YouGov Total EV UCS USA* UCS USA**
(>2,000) (1,000) (1,213) (1,213)
Source (Number of respondents)
© 2018 Citigroup
THERE ARE STILL SIGNIFICANT BARRIERS TO MASS ADOPTION FOR EVS
Range - Will it get me there? Total Cost of Operating – They cost more
(Source: Company data, Whatcar)
2024
Cost of Battery Average cost of equivalent
Claimed
8 Electric Vehicle Internal Combustion
201 (BEV) Engine (ICE) model
Real world
2012
100% vs.
63%
Infrastructure - It’s hard to find a place to plug in
Based on an average sample of 10 countries, there are 3.7 gas pumps per 100km of road
for each charging station (Source: IEA, CIA, Fuels Europe, Petrol Plaza, U.S. Census data,
Global News, Statista, Citi Research)
30 Charging Gas
25 Points Pumps
20
15
10
0
Netherlands Norway UK China Germany Japan France U.S.A Sweden Canada India
OUR FORECASTS FOR EV PENETRATION ARE 20% Citi Bull-Bear Case for
Bull
A function of both the production plans at Global BEV Penetration
18%
OEMs and what we believe carmakers will need
to achieve in terms of BEV and PHEV volumes 16%
to be compliant in a regulatory context. 14%
(Source: Citi Estimates)
12%
Base
Assumes consumers 10%
actively pursue EV’s over
8%
and above ICE vehicles
6% Bear
Demand driven by 4%
government support
2%
0%
Regulatory targets are
2014 2018 2022 2026 2030
missed in EU and China
6 Citi GPS: Global Perspectives & Solutions June 2018
Contents
Introduction 7
Where Are We Now? 10
What Does an EV Look Like Today? 11
Where Are We Going To? 12
What Will Drive Adoption? 13
Mass Adoption Requires the Stars to Align 14
Regulatory Factors 15
The Supply Outlook for EVs 16
Carmakers Will Dictate the Pace of Adoption 17
Current Barriers to Mass Adoption 21
1. Range: “Will It Get Me There?” 21
2. Infrastructure: “Where Do I Plug It In?” 25
3. Battery Longevity: “How Long Will the Battery Last?” 31
4. Residual values and TCO: “Why Should I Pay over 30%
Higher List Price for an EV?” 33
Deep Dive into Battery Cost 35
Demand Driven Factors 38
How Many EVs Are Required? 40
Model for European Powertrain Mix 40
Model for China Powertrain Mix 43
The Future of Electric Vehicles 46
Citi’s Powertrain Forecasts 52
The Outlook for Supply (from OEMs) 56
Conclusion 62
Appendix 1 63
Vehicle Types 63
Appendix 2 64
Top Selling EVs by Region and OEM 64
Appendix 3 65
Current Penetration by Region and OEM 65
Europe 65
China 68
U.S. 72
Japan 77
Appendix 4 79
Regulation by Region 79
Appendix 5 96
Product Pipeline by Manufacturer 96
Appendix 6 99
Navigating the Shift from NEDC to WLTP 99
© 2018 Citigroup
Introduction
Electric vehicles (EVs) are all the rage. Politicians, senior executives (from inside
and outside the automotive industry), investors, journalists, and Joe Public all have
something to say about them. It is no surprise that the number one topic on the
minds of automotive executives is battery electric vehicles (BEVs), having been just
10th on the list in 2014 (see Figure 1). In this report, we tackle the key issues
relevant in the decision-making process, from a car manufacturer’s perspective, for
electric vehicles, and we explore what needs to happen to break down the great
wall of barriers to adoption.
Figure 1. Battery Electric Vehicles Are at the Forefront of the Minds of Automotive Executives
2014 2015 2016 2017

0
1
2
3
Ranking (#)
4
5
6
7
8
9
10
BEVs (#1) Connectivity (#2) Fuel cell Evs (#3) HEVs (#4)
Source: KPMG Global Automotive Executive Survey (2017), Citi Research
While their rise has been gradual, global penetration of battery electric vehicles is
still at a low level. Until now, regulation has been the primary driver of supply, as it
has influenced the actions of car makers. Consumers are coming to grips with the
technology, and while generous incentives help, there are still several barriers that
need to be overcome (see Figure 2). Like with any new technology this will take
time. The most optimistic forecast we have seen for BEV penetration is 14% (from
BCG) in 2030, and in our base case we assume it will reach 10% (see Figure 3).
© 2018 Citigroup
Figure 2. EVs – The State of Play as It Stands Today Figure 3. Forecast EV Penetration (% New Car Sales) in 2030
35%
37m
30%
25% 30m 27m

25m 25m
20%
15%
10%
17m
5% 13m 15m
7m
0%
Clean Bloomberg Boston Citi McKinsey LMC* Wood
Energy New Consulting Energy Mackenzie
Ministerial Energy Group Insights
Finance
BEV PHEV HEV "EV"
“EV” is defined as BEV, PHEV, and Full Hybrid, *LMC forecasts for 2027
Note: Data labels show the number of units that are expected to be “EV” and BEV
Source: Citi Research Source: Citi Research, Company reports
The future is electric, but when has always been the question. The battle
between internal combustion engines (ICEs) and electric vehicles (EVs) is not new.
The car as we know it (gasoline/diesel internal combustion engine) was invented in
~1870, and, depending on which account of history you read, by either Siegfried
Marcus or Karl Benz. They were not, however, the first to make a motorized vehicle
– that honor goes to Frenchman François Isaac de Rivaz, who in 1808 invented a
hydrogen-powered vehicle. What’s perhaps more interesting is that electric vehicles
started appearing some 30 years prior to internal combustion vehicles, so arguably
the “disruption” that many believe is imminent has been a long time coming. We
think the reasons why one technology succeeded while the other remained niche
are the same today as they were almost 180 years ago: energy density (i.e.,
amount of energy stored in a given system or region of space per unit volume) and
its relationship to cost/convenience, but this looks set to change.
Figure 4. Energy Density by Chemistry: Gasoline Wins Easily Figure 5. Energy Transfer Efficiency Using Renewable Energy: EV Wins
400 Wh/kg 80%
Wh/kg 12,000
350
10,000
300
60%
250 8,000
200 6,000
150 40%
75%
4,000 73%
100
50 2,000
20% 40%
35%
- 0 26%
22%
Gasoline
Lead Acid
NiCd
(phosphate)
(manganese)
(cobalt)
NiMH
Li-ion
(rhs)
Li-ion
Li-ion
0%
Tank to wheel efficiency Well to wheel efficiency
BEV ICE* Hydrogen Fuel Cell
*Based on latest Toyota 2.0 litre gasoline engine

Source: Citi Research, epectec Source: Transport & Environment, Toyota, Citi Research
© 2018 Citigroup
Regulatory pressures are reducing EV’s cost disadvantage. Transportation is

ultimately about getting from one place to another, and therefore utility in the
context of transport is largely a function of speed and convenience. We think other
factors are secondary. Comparatively poor energy density (see Figure 4) has been
the Achilles heel of EVs in terms of range, refueling, and cost. It might be improving,
but it is unlikely to ever do so sufficiently so as to result in a threat to fossil fuels.
However, if you broaden the definition of cost to include air quality, then energy
density could improve. Authorities are acutely aware of the increasingly urban
nature of populations and are pushing for zero in-use tailpipe emissions as a result.
Using taxation and incentives rather than outright bans (thus far), authorities are
driving up the cost of conventional ICE vehicles, while growing EV volumes are
providing the scale for both higher levels of research and lower costs of production.
1
Cost equivalence between EVs and ICE is now within sight.
Vehicle Types
Electric vehicles can broadly be split into four categories. Hybrid Electric Vehicles
(HEVs), Plug In Hybrid Electric Vehicles (PHEV), and Electric Range Extended
Vehicles (E-REV) all include internal combustion engines that are supplemented
(usually at low speeds, in urban areas) by electric batteries. Battery Electric
Vehicles (BEVs), however, are entirely powered by electricity from the grid. The
distinction between HEVs and PHEVs is that the batteries for the HEVs are charged
from energy recuperation, while PHEVs, like BEVs, are charged from the mains.
1
When Will Electric Vehicles be Cheaper than Conventional Vehicles?, BNEF, 12-Apr-
17
© 2018 Citigroup
Where Are We Now?

EVs are still a minor proportion of auto production. Contrary to media hyperbole
and the large tomes of commentary written about the rise of EVs and the demise of
the internal combustion engine, in reality the volumes of electric vehicles (PHEVs
and BEVs) produced and sold globally are still relatively minor in the context of
global vehicle production (see Figure 7). We think there are good reasons for this,
not least because the incumbent carmakers have been somewhat reticent to put the
full weight of their R&D and marketing force behind EVs. Ultimately, however, we
think the weak demand for EV stems from the shortcomings of current battery
technology.
Figure 6. Google Search Interest for “Tesla” and “Electric Vehicles” Figure 7 Global New Car Sales Penetration for PHEV and BEV
120 0.8%
Tesla announce Model 3
100
0.6%
80
60 0.4%
40
0.2%
20
0 0.0%
1/13/2013 1/13/2014 1/13/2015 1/13/2016 1/13/2017 1/13/2018 2013 2014 2015 2016 2017
Tesla Electric Vehicles BEV PHEV
Source: Citi Research, Google Trends Source: Citi Research, LMC
The OEMs are readying themselves for a shift away from ICE. We believe that
the mass adoption of an alternative, cleaner powertrain is inevitable, and if the
commentary and intentions from the car manufacturers are to believed it would
appear they agree. In Figure 9 we show how the average research & development
(R&D) spend per vehicle sold among global original equipment manufacturers
(OEMs) has evolved over the past 12 years. It shows clearly that it has risen by
40%. There could be a number of factors driving this increase, but we would argue
R&D in relation to electrification is the primary contributor. In Figure 8 we show how
LMC’s BEV and PHEV forecasts have evolved for the past two years, it shows
rising forecasts in China and Europe, but falling in North America.
© 2018 Citigroup
Figure 8. Evolution of LMC’s BEV and PHEV Forecasts Figure 9. Average R&D per Vehicle for Global OEMs Trending Upwards
8 Units (Mn) 1,500
1H16 2H16 1H17 2H17
7
1,400
Average R&D per unit (€)

6
5 1,300
4
1,200
3
1,100
2
1 1,000
-
Europe N. America China Europe N. America China 900
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017
2020 Forecasts 2026 Forecasts
Source: Citi Research, LMC Source: Citi Research, Company Data (BMW, Daimler, FCA, Ford, GM, Honda,
Renault, PSA, Toyota, Volkswagen)
What Does an EV Look Like Today?

Smaller range, longer refueling, and higher price – not a very compelling
pitch. As Figure 10 and Figure 11 illustrate, the current generation of EVs (both
BEV and PHEV) compares poorly to conventional ICE vehicles. This might be
considered an unfair comparison as the fueling and maintenance costs of an EV are
considered to be substantially lower. For example, at prevailing energy prices a
Nissan Leaf costs €0.04 per mile whereas a 1.6L diesel VW Golf costs €0.08 per
mile. We challenge this analysis, however, as embedded in the overall running
costs of an EV is the depreciation of its battery, and, as we detail later, this adds a
significant burden to the running costs of an EV. Based on the metrics we have
chosen, it is hard to conclude that today’s EVs are well placed to displace ICEs in a
meaningful way.
Figure 10. Characteristics of an EV: Lower Emissions, but Lower Utility Figure 11. Characteristics of an ICE: Easier and Cheaper, but Dirtier
Model Range Rapid Standard Average Direct CO2 Model Range Re-fueling Average Price Direct CO2
(electric only) Charge Charge Price ($) Emissions (km) Time (mins) ($) Emissions
(km) (min) (hour) (g/km)
Tesla Model S 489 40 11 79,445 0 Toyota Corolla 989 3.7 23,666 168
Nissan Leaf 249 30 5 29,608 0
VW e-Golf 299 35 5 39,082 0 Ford Focus 1,377 3.7 21,351 120
Renault Zoe ZE 402 60 6 22,658 0 Volkswagen Golf 1,065 3.3 24,017 148
Chevrolet Bolt 238 60 6 37,707 0 Toyota RAV4 1,410 4.0 33,312 126
BMW i3 314 35 5 49,487 0 Honda Civic 1,175 3.3 28,330 129
Average BEV 332 43 6 42,998 0 Volkswagen Polo 1,006 3.0 16,458 137
Toyota Prius 1328 (39) N/A 3 27,445 91 Honda CR-V 1,220 3.9 27,897 148
Mitsubishi Outlander 873 (33) 25 4 36,064 96 BMW 3-series 891 3.4 40,159 178
Chevrolet Volt 676 (53) N/A 4 34,688 29
BMW 330e 882 (25) N/A 2 37,275 44 Renault Megane 1,255 3.1 25,791 112
Volkswagen Passat 1063 (31) N/A 3 36,165 40 Mercedes E-Class 1,728 4.4 60,501 122
Average PHEV 964 (36) N/A 3 33,868 65 Average ICE 1,212 3.6 30,148 139
Source: Citi Research, LMC, Journal of Industrial Ecology, Edmunds, AutoHaus, Source: Citi Research, LMC, Car-Emissions.com, CarAraC.com, AutoHaus, Voiture-
Voiture-Neuve, Broadspeed, Zap Map, Smart EV, Pod Point Neuve, Broadspeed
© 2018 Citigroup
Where Are We Going To?

The timing and scale of a potential EV disruption are open to debate. It may
well be that the future is electric, but the timing and scale of that epiphany is not
clear. As we detail below (Figure 12), courtesy of data collected by industry
consultant Ricardo, there is a broad array of EV penetration forecasts in the public
domain, and the further out we look the greater the variance in estimates. This is
problematic for both manufacturers and investors: Investing for 40% penetration in
2025 is clearly a very different prospect than sub-10% penetration in the same
timeframe. It is interesting to note that the more recent forecasts are less optimistic
than prior years. It may be that the initial hysteria around EVs is calming down.
Figure 12. New Car Market Penetration Predictions for BEV & PHEV
acea LOW 2017

70% acea HIGH 2017
ERTRAC Prediction assuming breakthrough in energy storage IEA B2DS 2017
cost reduction and major infrastructure investments for re-charging IEA 2DS 2017
60% Paris Declaration 2017
ERTRAC Prediction assuming CO2 targets achieved via IEA RTS 2017
improvements in ICE-based vehicles and hybridisation Roland Berger Scenario A 2016
Market penetration (new car sales)
Roland Berger Scenario B 2016

50%
Oliver Wyman "slight change" 2015
Oliver Wyman "Awareness" 2015
Oliver Wyman "Green world" 2015
40% McKinsey below 10 2014
McKinsey below 40 2014
McKinsey below 100 2014
30% Camecon tech 2 2013
Camecon tech 3 2013
ATKearney 2012
20% CE Delft "most realistic" 2011
CE Delft "ICE breakthrough" 2011
CE Delft "EV breakthrough" 2011
JRC EU15 low 2011
10%
JRC EU15 medium 2011
JRC EU15 high 2011
JRC EU12 low 2011
0% JRC EU12 medium 2011
2015 2020 2025 2030 JRC EU12 high 2011
Source: Published forecasts, Ricardo
We believe further penetration is a function of push vs. pull demand. Our own
forecasts for EV penetration (see section Citi’s Powertrain Forecasts) are a function
of both the production plans of the OEMs and what we believe the carmakers will
realistically achieve in BEV and PHEV volumes as they strive to be complaint with
fuel-efficiency targets. Arguably, our forecasts are based largely on push demand in
the sense that we make no assumption around consumers actively pursuing EVs
over and above ICE vehicles. We view this as the upside risk to our forecasts, but,
as we explain below, the timing of this swing is at least five years away and far from
being assured. For now, we think regulatory pressures will be the key driver of
adoption.
© 2018 Citigroup
What Will Drive Adoption?

In the absence of government support EVs need to be as good as, or better
than, an ICE. Car buying, like purchases of many consumer products, is an
emotive topic. The suggestion that one vehicle or brand is better than another is
almost guaranteed to generate significant debate. However, it is worth remembering
that more than half of the vehicles sold today, depending on the market, are to
business users. We argue their purchasing decisions are more rational, based on
utility vs. cost i.e., what is the most cost-effective vehicle that will complete the
required function. Furthermore, we argue utility vs. cost is still a key element in the
decision-making process among private buyers, especially where the main intended
use of the vehicle will be commuting. Mainstream adoption of EVs requires
equivalent, or better, utility versus conventional technology, which is not currently
the case due to limited range, higher total cost of ownership, and limited charging
infrastructure. This is clearly illustrated when contrasting comparable EV and ICE
model sales volumes in Figure 13 and Figure 14.
Figure 13. Toyota Prius Hybrid and Toyota Camry (ICE) Sales in the U.S. Figure 14. Nissan Leaf BEV and Nissan Versa (ICE) Sales in the U.S.
500 LTM Sales (k) 500 LTM Sales (k)
400 400
300 300
200 200
100 100
0 0
12/10 09/11 06/12 03/13 12/13 09/14 06/15 03/16 12/16 09/17 12/10 09/11 06/12 03/13 12/13 09/14 06/15 03/16 12/16 09/17
Toyota Prius Hybrid Toyota Camry Nissan Leaf Nissan Versa
Source: Citi Research, Autodata Source: Citi Research, Autodata
Electric vehicles, however, also suffer from a perception problem that makes
this gap in utility difficult to narrow. When drivers switch from ICE to EV, they
are overwhelmingly pleased with their decision. According to a survey of over 850
EV drivers in Europe and North America conducted by a group of car manufacturers
and EV organizations, 85% of electric vehicle drivers are happy having made the
2
switch to an electric car. This suggests that although there are clear functional
benefits of driving a conventional gasoline or diesel vehicle, there are also aspects
of driving electric that consumers prefer. The problem is that the former (range, total
cost of operating, infrastructure, battery longevity) are more prominent in the minds
of consumers than the latter (environmental benefits, quieter drive experience, fuel
costs). Both the actual and the perceived disadvantages of electric vehicles will
need to be overcome in order for mass adoption to occur.
2
85% of electric vehicle drivers are happy having switched to electric driving, EVBox,
Apr-17 link
© 2018 Citigroup
Mass Adoption Requires the Stars to Align

We are in the early stages of electric vehicle adoption, with some car manufacturers
still deciding on their respective strategies. Up until now, much of the supply has
been driven by regulation, with some countries enforcing minimum sales
requirements and/ or implementing stricter fuel efficiency targets that will only be
attainable with higher EV penetration. In the absence of regulation we believe EVs
would have remained a niche offering, as the costs to manufacture (and therefore
the price to buy) would have remained prohibitively high. It is regulation that has
forced the hands of OEMs, making them reallocate capital in order to provide an EV
offering, to avoid being fined or, worse still, being considered irrelevant (as an EV-
brand) by consumers. The bottom line is that in order to encourage adoption of EVs
at this stage, financial incentives have been necessary.
Figure 15. There Is Still a Long Road to Navigate Before We See Mass Adoption of Electric Vehicles
Source: Citi Research
We believe that true mass adoption will be reached when neither regulation nor
incentives are required to force OEMs to supply, or consumers to buy, EVs. In order
to reach this stage there are numerous barriers that need to be overcome. We take
each of these in turn in a later section of this report.
© 2018 Citigroup
Regulatory Factors
In the near term, we expect EV demand to be largely propelled by a
combination of regulation and government incentives. Thus far, the growth of
EVs has been distinctly supply- rather than demand-driven as EV enthusiasm
among consumers remains muted. Despite the fact that EVs are currently margin-
dilutive (Daimler estimates that, at least initially, EVs will be half as profitable as
their ICE equivalents), OEMs are working on changing the perception of EVs
among consumers. We believe this is largely due to the regulatory environment in
which OEMs operate requiring them to do so.
A combination of carrots and sticks from local, national, and supranational

governments is the real engine behind EV market growth. Carmakers are being
pushed towards increasing investment in and development of EVs as a result of the
regulatory environment. In China, the government is targeting 2 million New Energy
Vehicle (NEV) units by 2020 (vs. ~580k in 2017), and we estimate 1.3 million units
will need to be sold in Europe in 2020 (vs. 270k in 2017) to comply with emissions
targets. In order to help generate demand, subsidies and non-financial incentives
are being implemented, and we see a strong correlation between generous
subsidies and EV penetration.
We summarize the key areas of regulation that are driving demand in the world’s
three largest markets in Figure 16 and analyze the relationship between regulation
and EV penetration for each region in further detail in the section Regulation by
Region.
Figure 16. Summary of Regulatory Drivers for EV Adoption in Top 3 Markets

Region Market Size EV Driver Government Target/ Comment
(mn units) *Implied EV Volume
Dual credit NEV system:
New Energy Vehicles (NEV) is a collective term used
1) NEV credits must reach 10%/12% of 2019/20 annual volumes 2020: 2m units in China for BEV and PHEV models.
China 25
2) Corporate average fuel consumption (CAFC) credits must be at a 2025: 7m units
non-negative balance each year
California and 12 other states have historically been
CAFE fuel-economy targets originally set under the Obama allowed by an EPA waiver to set and opt in to stricter
U.S. 17 administration, which targeted average fuel efficiency of 41/50mpg by N/A standards, but this waiver is also likely to be
2021/25, are currently under review and are expected to be relaxed. challenged by the Trump administration.
Additional regulation is driving up ICE vehicle costs
CO 2 emissions targets of 95g/km in 2020/21 and 15%/30% reductions 2020: 1.56m BEVs*
Europe 16 by 2025/30 while local authorities are simultaneously pushing for
2025: 3.7m BEVs* low-emission zones in major cities.
Source: Citi Research, MIIT, NHTSA, European Commission
In China, the carmakers operate under a dual-credit system whereby credits are
received based on both NEV production and average fleet fuel consumption, which
the government hopes will help meet the target of 2 million NEV sales by 2020 and
7 million by 2025.
U.S. fuel-efficiency regulation is in a state of flux as the 2021/25 targets

implemented by the Obama administration are under review and likely to be relaxed
by the Environmental Protection Agency (EPA). California’s EPA waiver, which
historically has allowed it to set its own (stricter) standards and for other states to
opt in, is also likely to be challenged by the EPA. Of the three markets, regulatory
supply-driven factors are weakest in the U.S., and EV penetration is also set to
grow the slowest.
© 2018 Citigroup
In Europe, emissions targets are the most significant regulatory driver of EVs.
Manufacturers are required to reduce average fleet CO2 emissions to 95g/km by
2020/21, which in an environment of falling diesel sales and growing popularity of
SUVs is putting pressure on OEMs to improve sales of low- and zero-emission
vehicles. Local authorities are also aiming to reduce emissions in cities through the
implementation of Low Emission Zones and other measures that are driving up ICE
vehicle total cost of ownership.
The Supply Outlook for EVs

Let’s take a step back and think about the shift towards EVs from the perspective of
the auto manufacturing companies and their boards. Companies with a long-history
of manufacturing traditional vehicles tend to make decisions based on the
prospective investment return relative to their cost of capital. Structural threats, like
the shift towards EVs, can derail the traditional process, although this is rare. The
majority of decisions (particularly those of material size) require sign-off by the
company’s board, which can slow down the process. This is an important distinction
between incumbent operators and new entrants, where decision making is more
fluid.
This leads us on nicely to what we dub as ‘The CFO’s Conundrum.’ What do we

mean? Automotive OEM margins are thin, and while their balance sheets are more
robust than they were a decade ago, they need to be disciplined with their capital
investment decisions. The dilemma is how capital spend should be split between
improving existing technologies/ infrastructure and investment in new technology —
like EVs – where there is significant uncertainty about the demand outlook.
The CFO’s Conundrum Matrix
We think ‘The CFO’s Conundrum’ is best illustrated by the following matrix:
© 2018 Citigroup
Figure 17. The CFO’s Conundrum: Where Are We Most Comfortable?
High Follower Leader

Technological Evolution/ Progression
• Makes the decision to invest once there • New entrants and disruptors occupy this
is proven demand quadrant
• At the point in time when the decision to • For industry incumbents positioning the
invest is made the cost of development is company in this segment may impair
much lower, and therefore it takes less profitability in the short term, but enables
capital to manufacture a technologically association with the new developments
advanced product (no learning costs etc.) (gaining market share early-on)
Laggard Improver
• All incumbents start off in this quadrant • Investing to be the best in the old
• If demand for technologically advanced technology (generally seen as a
products takes time, then these prediction that there will be no structural
companies benefit change in customer preferences)
• Like with ‘laggards’, if demand for
technologically-advanced products takes
Low time, then ‘improvers’ benefit
Low High
Capital & Research Expenditure
Carmakers Will Dictate the Pace of Adoption

The automotive industry is vital for some of the world’s largest economies. It
employs 14% of industrial workers in Germany, 7% in France, and 5% in U.S. It also
accounts for 14% of gross domestic product (GDP) in Germany and 3% in U.S. For
that reason governments of the respective countries keep a watchful eye, setting
regulation, injecting money to stimulate demand (incentives, scrappage etc.), and
giving worker unions a degree of power that is rare in other industries. That does
not mean the car makers are given a free-ride; this is best highlighted by the
3
tightening of fuel efficiency regulation following the diesel crisis. It is no surprise
that battery electric vehicles have gone from being tenth on a list of key trends that
automotive executives are focused on in 2014 to first in 2017. This does not mean
that EVs make economic sense for car-makers right now, but we would argue that
the future is inevitably going to be electric. In the meantime, carmakers need to
decide where they want to sit in our ‘CFO’s Conundrum’ matrix, while at the same
time playing lip service to the rise of EVs if for nothing else but the sake of their
brands.
Who is Leading Today, and What Have the OEMs Announced?
As shown Figure 18, BAIC sold the most EVs globally in 2017, but with EVs
accounting for 16% of the group’s volumes. In 2017 it only sold 7 models but plans
to have 13 models available by 2020 and 2025. The picture in 2020/25 looks rather
different, as shown in Figure 19, with VW and SAIC leading in terms of absolute
volumes globally, although pure-play EV manufacturers aside (i.e., Tesla), BAIC and
Jianghuai are expected to have the highest EV penetration in 2025.
3
European Autos - Diesel outlook remains hazy.
© 2018 Citigroup
Figure 18. Automotive OEMs: BEV Volumes (Global; 2017) and % Figure 19. Automotive OEMs: BEV Volumes (Global; 2020 and 2025) and
Penetration % Penetration
120 1,600
2020 2025
1,400
100
1,200
80
Units (000's)
1,000
Units (000's)
60 800 12%
16%
600
40 4% 100% 1%
20%
400 6% 45% 14% 100%
20 12% 6%
5% 13% 2% 0.5%0.5% 200 4% 6% 10% 47%
1% 0.2% 6% 15%
1% 2% 7% 2% 29% 7% 100%
0 2% 1% 1% 3% 28% 1% 6%
0
Hyundai
Tesla
Jianghuai
GM
VW
BMW
BAIC
Geely
Daimler
BYD
Chery Gp
Changan
VW
Hyundai
GM
Tesla
Jianghuai
SAIC
BAIC
Geely
Toyota
Daimler
Honda
Changan
Renault
Renault
Source: LMC, Citi Research Source: LMC, Citi Research
© 2018 Citigroup
Cost Parity Is a Critical Milestone for an Inflection of Supply
Ultimately, the auto OEMs are faced with a choice: Either invest today in EVs when
the cost of the technology is highest and demand is uncertain, or wait for costs to
fall and demand to rise. The critical milestone for supply to inflect will be when cost
parity between EVs and traditional internal combustion engines is reached. With the
cost of batteries expected to decline, and with the cost of emissions compliance for
conventional engines (ICE) rising, we wouldn’t be surprised if cost parity is reached
before 2025 (see Figure 20). The challenge, until then, is that electric vehicles are
considerably more expensive to manufacture (see Figure 21), and as (most)
carmakers are unable to charge consumers to compensate for this gap, the sale of
electric vehicles, as it stands today, is a loss-making endeavor.
Figure 20. EVs vs ICE – Estimated Industry Costs Figure 21. Estimated Breakdown of Costs by Powertrain
USD Gasoline Diesel 48V PHEV BEV
Engine 2,000 2,200 1,800 1,200 -
Transmission 1,600 1,600 1,600 1,840 800
Fuel System 400 800 400 400 -
Axles 800 800 800 800 800
GPF* 100 100 100
SCR/ LNT - 500 - - -
Electric Motor - - 380 842 1,684
Additional Battery - - 400 2,838 11,350
DC/DC convertor - - 50 50
TOTAL 4,900 5,900 5,480 8,020 14,634
Cost vs ICE-gas +1,000 +580 +3,120 +9,734
Battery size (kWh) - - 0.2 12.5 50

Source: BorgWarner, Delphi, American Axle, Magna, IHS, Robert Bosch, Aumann,
Source: Daimler McKinsey. Citi Research estimates. * Gasoline Particulate Filter
© 2018 Citigroup
OEMs have been readying themselves for EV-adoption
While some argue electric vehicles are just hype, the actions of the carmakers
suggest that they believe EVs have a sustainable future. In Figure 22, we show how
capital expenditure (capex) and R&D (i.e., ‘investment’ by our definition) per unit
(i.e., vehicles sold) have evolved for global carmakers over the past decade. We
attribute the greater than 40% increase in investment since 2006 to EV-related
expenditure, and recent commentary from OEMs suggests the upward trajectory is
set to continue.
Figure 22. Automotive OEMs: Evolution of ‘Investment’ Per Unit Figure 23. Cross-Sector ‘Investment’ to EBIT Ratio and EBIT Margins
2,900 'Investment' : EBIT ratio Adj EBIT margin (RHS)
2.0 40%
2,700 1.8 1.7x
Average 'Invstment' per unit (€)
35%
1.6 1.5x
30%
2,500 1.4
1.2 25%
2,300 1.0 20%
0.8x
0.8x
0.8 15%
0.6x
2,100 0.6 0.4x 0.4x 10%
0.4
1,900 0.2 5%
0.0 0%
1,700
Pharma
Autos
Supplies
Components
Hardware
Software
Internet
Medical
Semis &
1,500
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017
Source: Citi Research, Company Data (BMW, Daimler, FCA, Ford, GM, Honda, ‘Investment’ = Capex + R&D expenditure
Renault, PSA, Toyota, Volkswagen) Source: Citi Research, Company data. 2017 data
4
Despite their thin EBIT margins we found it interesting to see that carmakers
5
featured six times in the list of top 20 R&D spenders globally — the most of any
industry. The six auto manufacturers in the list are VW, Toyota, GM, Ford, Daimler,
and Honda. This point is further accentuated in Figure 23, where we analyzed the
relationship between ‘investment’ and EBIT across the 20 companies, split by
industry group. This shows that on average the automotive OEMs spent 1.7 times
EBIT on capex and R&D (in 2017) compared to an average of 0.8 times in other
industries.
4
EBIT = earnings before interest and taxes, which is a measure of a firm’s profit.
5
https://www.strategyand.pwc.com/innovation1000#VisualTabs1.
© 2018 Citigroup
Current Barriers to Mass Adoption

As it stands today, we see four barriers to mainstream adoption of EVs. In the
section below, we consider the current situation on each issue and more
importantly, what we see happening around each concern over the coming years.
1. Range: “Will It Get Me There?”

Range anxiety is one of the greatest hurdles for BEV adoption. In Figure 24 we
show the results of various surveys which asked the question ‘What is the single
biggest barrier to you owning a BEV?’ While price and charging infrastructure
feature as barriers, range anxiety is consistently one of the top three reasons given.
We suspect this is a function of consumers wanting the option to drive long
6
distances, even though the average trip length is only 15 miles in the US and 7
7
miles in Europe – both well within the available range of current BEVs. A further
complication is that it takes significantly longer to recharge (refuel) a BEV than it
does a traditional vehicle equipped with a fuel tank (as shown in Figure 25). We
believe this places even greater emphasis on range as refueling is less convenient
in terms of time taken, but to the degree charging can take place when the vehicle
is not in use, the range challenge can be reduced.
Figure 24. Survey Querying ‘Main Barrier for Higher Adoption of EVs Figure 25. Average Refueling Time: ICE is the Clear Winner
35% Price Charging Infrastructure Range Hours
8
7 hrs 20 mins
30% 7
25% 6
20% 5
4
15%
2 hrs 52 mins
3
10%
2
5% 43 mins
1
4 mins
0% 0
YouGov Total EV UCS USA* UCS USA** ICE BEV * PHEV** BEV
(>2,000) (1,000) (1,213) (1,213)
Source (No. of respondents)
* respondents from California * Rapid charging to 80%

** respondents from 9 Northeast U.S. states ** Time taken to charge battery, refuelling combustion engine in ~4 mins also available
Source: Citi Research, YouGov, Total EV, USC USA Source: Citi Research, Zap Map, Pod Point, Smart EV
Successive BEV generations have seen an improvement in range. While BEV

penetration is rising, we would argue the niche adoption thus far has in large part
been a function of limited range. It might be argued that infrastructure and residuals
also pose a challenge, but as Figure 26 shows, the range hurdle appears to be at
the forefront of existing carmakers development plans for BEVs — each model
update has seen a meaningful jump in range. The car industry has made the news
in recent years for the gap between real world and claimed emissions, and there is
a similar story to tell for BEV driving range (Figure 27). This issue is exacerbated for
BEVs given the range deterioration at different ambient air temperatures (extremes
impact the battery performance and energy consumption), when interior functions
are used (e.g. cabin climate control) and depending route topography (flat is good,
hills are not).
6
National Household Travel Survey, U.S. Department of Transport (2001-2002).
7
Analysis of National Travel Statistics in Europe, European Commission (2013).
© 2018 Citigroup
Figure 26. Battery Range in Successive BEV Model Generations Figure 27. Average BEV Range Evolution: Claimed and Real-World
400 Range (km) 450 Range (km)
40 kWh
350
41 kwh
400
300
350
33 kwh
250
30 kWh
300
200 22 kwh
22 kwh
250
24kwh
150
100 200
50 150
2013
2016
2012
2016
2016
2018
2011
0 100
BMW i3 Renault Zoe Nissan Leaf 2012 2014 2016 2018 2020 2022 2024
Gen 1 Gen 2 Gen 3 Claimed Real World
Source: Citi Research, Clean technical, Autocar, Greencar reports, Inside EVs, Source: Citi Research, LMC
Company data
As battery costs come down. vehicle range is being extended. In Figure 28 we

show how the driving range (on a single charge) has changed for various BEV
models since 2011. All three models at the mass-market end of the spectrum
(Nissan Leaf, Ford Focus BEV, and VW e-Golf) have seen an improvement in
range, which is largely due to an increase in battery capacity. More interesting is the
recent launch of the Chevrolet Bolt, which is equipped with a larger battery and
offers a quoted combined range of 383km/238miles, but is priced around £30k. In
Figure 29 we plot range against vehicle price, and the Chevrolet Bolt is the outlier
on this scatterplot. It looks to us as though GM (the manufacturer of the Chevrolet
Bolt) is acting as a price-aggressor, i.e., offering a superior product (if range is the
key determinant) at a low price point.
Figure 28. Range (km) Evolution of BEVs, According to the U.S. EPA Figure 29. Range (km) vs. Price – For a Selection of BEVs
500 Range (km) 140,000 Price (£) Tesla X P100D
450
Tesla X 90D 120,000
400 Tesla X P90D
Chevy Bolt Tesla X 100D
350 Tesla Model S 100,000
(60 kW-hr)
300 BYD e6
80,000
Tesla S 75
250
VW e-Golf 60,000
200 BYD e6
Nissan Leaf Ford Focus
Ford Focus (BEV)
150 40,000 Chevrolet Bolt
BEV
VW e-Golf
100
20,000
Nissan Leaf
50
0 0
2011 2012 2013 2014 2015 2016 2017 2018 160 210 260 310 360 410 460
Range (km)
Source: Citi Research, U.S. Department of Energy Source: Citi Research, WhatCar, AutoCar, U.S.Department of Energy
© 2018 Citigroup
Why Not Just Add Bigger Batteries?
Battery size is a key determinant of driving range, but also vehicle price. The
interplay between battery size and efficiency, charge time, and cost is problematic,
and there is also the added complication of safety (see Figure 30). To keep weight
down, ideally, manufacturers would like to increase energy density, but at the same
time reduce charging times. Faster charging creates more heat, as does higher
energy density, which in turn presents a bigger challenge in terms of fire
propagation. It’s a tricky conundrum.
Figure 30. Citi’s Battery Challenges Diagram
Battery
density
Safety
Battery Charging
costs Decrease in charging time increases costs time
A larger battery leads to improved range, but is less efficient. As you would
expect (Figure 31), battery size and range show a strong positive correlation, with
2
an R of 0.865. If range is an issue, it would therefore makes sense to make the
battery bigger (i.e., higher kWh), but this would also make the vehicle heavier. As
you might expect, vehicle weight is a key factor in determining an EV’s potential
driving range, and it follows that the charge time for a larger battery would also be
longer. A study “Analysis of Parameters Influencing Electric Vehicle Range” by
Martin Mruzek et al. at the University of Zilina (Slovakia) found the larger battery to
be less efficient. We expect vehicle range to remain one of the foremost factors
when buying a BEV, although we are cognizant that as range anxiety lessens,
consumers may opt for the more efficient battery size.
© 2018 Citigroup
Figure 31. Battery Size vs Range (km) — For a Selection of BEVs Figure 32. Vehicle Weight vs Range (km) — By Battery Capacity
Range (km) Tesla X 100D 210 Range (km) 300Ah 260Ah

480 200Ah 100Ah
190
440
Tesla S 75 Tesla X P100D
Chevrolet Bolt 170
400
360 150
320 BYD e6 130
280 110
240 Ford Focus 90

(BEV) R² = 0.865
200 VW e-Golf 70
Nissan Leaf
160 50
20 40 60 80 100 800 850 900 950 1,000 1,050 1,100 1,150 1,200
Battery Size (kWh) Vehicle weight (kg)
Source: Citi Research, WhatCar, AutoCar Source: Citi Research, University of Zilina
Falling battery costs should alleviate range issues. Given the correlation
between battery size and range, and range and price (see section above), it follows
that battery size and vehicle price are well correlated. GM’s decision to offer a
vehicle (Chevrolet Bolt) equipped with a 60kWh battery pack is somewhat
aggressive compared with similarly priced vehicles (e.g., VW e-Golf, Nissan Leaf
and Ford Focus BEV), which all have a battery size below 40kWh. As the costs of
battery systems (cell + pack) fall, the potential driving range (on a single charge) will
rise, and this will help address the range anxiety. We look at this in our Battery Cost
section.
© 2018 Citigroup
2. Infrastructure: “Where Do I Plug It In?”

There is a chicken and egg quandary with EV charging networks. As we
pointed out in Figure 24 charging infrastructure is consistently seen as a barrier to
broader adoption of EVs. If we consider the density of charging networks compared
to gas stations (see Figure 33 and Figure 34) it is clear the availability of public
charge points for EVs still significantly lags the already well-established network of
gas pumps for ICE vehicles. But a comparison of EV and ICE public infrastructure is
somewhat academic given the different practices of refueling an electric vs. a
conventional car. The option to refuel (charge) an EV at home or work and the
longer charging times diminishes the need for public charging infrastructure, while
on the other hand range anxiety and a lack of visible public infrastructure act as
barriers to adoption. It’s not as simple as “if you build it, they will come.”
Figure 33. Charge Points per 100km of Road Figure 34. Gas Pumps per 100km of Road
25 Slow Chargers Fast Chargers 25
Charger points per 100km of road
Gas pumps per 100km of road

20 20
15 15
10 10
5 5
0 0
U.K.
China
France
U.S.
Sweden
Canada
Norway
Japan
China
U.K.
U.S.
Canada
France
Netherlands
Germany
India
Japan
Norway
Sweden
Netherlands
Germany
India
Source: Citi Research, IEA, CIA Source: Citi Research, IEA, Fuels Europe, Petrol Plaza, U.S. Census data, Global
News, Statista
Home charging is popular, but this is partly explained by the lack of charging
infrastructure elsewhere. A survey conducted by the EV website Zap Map in 2016
found that 81% of EV owners have access to chargers at home and only 15%
charge their cars at work (Figure 35). Yet it also revealed that only 18% of
respondents had the option to charge their car at their workplace, and of those who
did, over 80% chose to do so. Therefore, although home-charging is more common
than charging at work or at public stations, where workplace chargers are available
they are popular with drivers.
Secondly, the results suggest that public chargers are still required in combination
with home and workplace charging as almost half of respondents said they use
public facilities at least once a week (Figure 36). This figure may come down as
vehicle ranges improve, but, for now, it shows that regular charging using public
facilities remains a reality for many drivers, and it follows that improving the
infrastructure available to drivers will therefore be key to addressing range anxiety.
© 2018 Citigroup
Figure 35. Do You Have Access to Dedicated EV Charging Point at Your Figure 36. How Often Do You Use Public Charging Facilities?
Workplace?
60% 30%
50% 25%
40% 20%
30% 15%
20% 10%
10% 5%
0% 0%
Yes, I charge my Yes, but I don't No Does not apply More than Once a week Once a month Less than Not sure
EV at work charge once per week once a month
Source: Citi Research, Zap Map survey 2016 (1,463 respondents) Source: Citi Research, Zap Map survey 2016 (1,283 respondents)
Outlook for Charging Points by Region

Although China boasts the largest charging infrastructure in absolute terms,
the European nations have the best charging infrastructure per capita. In 2016
there were over 88,000 publicly available fast charging points in China, almost 15
times as many as in Japan, which had the second most. Interestingly, despite the
boom in fast charging points in China, which increased from 9,000 in 2014 to over
88,000 in 2016, the development of slow charging units has not kept pace. In fact,
China’s EV/EVSE (relationship between number of vehicles and charging stations)
ratio when considering only slow chargers has increased each year since 2014.
Figure 37. Publicly Available Charging Points per Country Figure 38. Publicly Available Charging Points per Million Population
160 Slow Chargers Fast Chargers 1,800 Slow Chargers Fast Chargers
Charger points per million population
1,600
No. of charger points ('000s)
140
120 1,400
1,200
100
1,000
80
800
60
600
40 400
20 200
- -
France
U.K.
U.S.
Canada
China
Norway
Sweden
Japan
Netherlands
Germany
China
U.S.
France
U.K.
Japan
Canada
Norway
Sweden
Netherlands
Germany
Korea
India
Source: Citi Research, IEA Source: Citi Research, IEA, World Bank
© 2018 Citigroup
When accounting for population size, Norway and the Netherlands provide
the best charging infrastructure. We have analyzed the number of publicly
accessible chargers per million people in order to assess how developed the
infrastructure is across key EV markets relative to population size. Our findings (see
Figure 38) suggest that the Netherlands and Norway, which have the highest EV
market shares (Figure 39), are also leaders in charging point infrastructure relative
to their population size. Norway’s slow charger electric vehicle (EV)/electric vehicle
supply equipment (EVSE) ratio (Electric, however, is the highest of the countries we
analyzed, suggesting that although it is one of the most developed relative to
population size, there is still room for improvement when it comes to servicing the
133,000 EVs currently on its roads.
Figure 39. EV Charging Infrastructure by Country Figure 40. Global Charging Points
EV Market Share EV/EVSE m units
(BEV & PHEV) (Public slow chargers) 14
13
Norway 28.8% 19
Netherlands 6.4% 4 12
Sweden 3.4% 13
France 1.5% 6 10
U.K. 1.4% 8 CAGR 51%
China 1.4% 12
U.S. 0.9% 16 8
Germany 0.7% 4
Canada 0.6% 10 6 2012-2014
Japan 0.6% 9 Conventional charging points >2x
Others 0.5% 4 Fast charging points 8x
Korea 0.3% 10 4
India 0.0% 15
Total* 1.1% 10 2 2
*Calculated based on total market size of all countries within IEA report
0
2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Source: Citi Research, IEA Source: Citi Research, Aumann presentation
At the national level, there appears to be little correlation between charging

infrastructure and market share. We have summarized the EV/EVSE ratio for
publicly-available slow chargers in Figure 39. The data do not suggest a strong
correlation exists between EV/EVSE and EV market share. Norway has the highest
market share and EV/EVSE ratio, but the Netherlands has the 2nd-highest market
share and among the lowest number of EVs per EVSE. India has the lowest market
share, but the 3rd-highest EV/EVSE ratio.
Figure 41. European Infrastructure Policy Initiatives and Incentives

Country Infrastructure Incentives
Denmark - Individuals receive a tax rebate of up to 18,000 DKK on the home installation of EV chargers
Ireland - Individuals installing the first 2,000 home chargers in Ireland would receive a grant for the total cost. This target was reached in 2017
- New €600 grant from the government towards installation of home charging points from 1 Jan 2018
Italy - Large non-residential building will be required to install EV charging points in order to meet building regulations
Malta - €2,000 grants are available for companies to purchase charging points (up to a maximum of 5 per company)
Norway - Government is providing public funding in order to reach its target of having at least one fast charging station per 50km of main road
Spain - Subsidies are available for both private and public charging points
UK - Individuals receive a £500 subsidy when installing a home charger; business receive a subsidy worth £300 per socket
- Local Authorities can receive a 75% refund (capped at £7,500) from the government for costs relating to installation of roadside charge points in
residential areas
Source: Citi Research, EAFO, ESB, EV Fleet World
© 2018 Citigroup
European Infrastructure
Below we illustrate the distribution of BEV sales and charging stations across
Europe. There is a notable divide between the levels of infrastructure development
in Western Europe and the rest of the continent. Clearly, charging stations are far
more prevalent in markets such as Norway, the Netherlands, Germany, France, and
the U.K. than in Southern and Eastern Europe.
Figure 42. European Charging Infrastructure and BEV Unit Sales
Source: Citi Research, ACEA, EAFO
© 2018 Citigroup
U.S. Infrastructure
Below we show U.S. BEV unit sales and charging stations by state in order to
illustrate the geographical spread of charging infrastructure in the U.S. The average
BEV unit sales per state in 2016 was 1,300 units, and sales were clearly driven by
the state of California, where over 30,000 units were sold. The average number of
charging units per state is just 412, and again these are disproportionately found in
California, where there are almost 5,000 charging stations available.
Figure 43. U.S. Charging Infrastructure and BEV Unit Sales
Source: Citi Research, Polk Automotive, Autodata, U.S. Department of Energy. Note BEV Unit Sales relate to 2016 volumes.
© 2018 Citigroup
U.S. infrastructure and sales highly concentrated in California. The total

number of BEVs sold in the U.S. in 2016 was 66,000 compared with total vehicle
sales in excess of 17.5 million, implying penetration of 0.4%. California accounted
for 46% of U.S. BEV sales, or ~0.2% of overall U.S. volumes. No other U.S. state
saw penetration higher than 0.02%. Unsurprisingly, our analysis of the geographical
split of EV charging points (Figure 44) tells a similar story: 24% of charging points in
the U.S. can be found in California. Only 6% are in Florida, which has the second
largest share, and 30 states have less than 1%.
Figure 44. Charging Infrastructure Across the Top 25 U.S. States by 2016 BEV Sales
0.20% 30%
0.18%
25%
0.16%
0.14%
20%
0.12%
0.10% 15%
0.08%
10%
0.06%
0.04%
5%
0.02%
0.00% 0%
GA
IL
OR
AZ
PA
VA
OH
CA
FL
TX
NY
CO
NJ
HI
NC
TN
WA
WI
MA
MD
UT
MN
CT
NV
MO
BEV U.S. Market Share
Share of U.S. charging points (rhs)
Source: Citi Research, Polk Automotive, Autodata, U.S. Department of Energy
© 2018 Citigroup
3. Battery Longevity: “How Long Will the Battery Last?”

Battery degradation continues to weigh on consumer confidence, despite
signs of durability from Tesla. The relatively recent, and limited, adoption of EVs
means the data available on battery longevity is limited; however, a crowdsourced
dataset containing inputs from almost 1,000 Tesla owners offers an insight into the
longevity of the battery in a Tesla Model S (see Figure 45). Interestingly, at 200,000
miles (often cited as the life expectancy of a conventional car), the Tesla Model S is
on average still operating at 93% of its original range, suggesting fears of battery
degradation could be overstated.
Figure 45. Tesla Model S Battery Degradation Figure 46. Nissan Leaf Battery Degradation
100% 100% 24 kWh 30 kWh Nissan estimate
Remaining Original Range
Battery State of Health

95% 90%
90% 80%
85% 70%
80% 60%
0 50,000 100,000 150,000 200,000 250,000 300,000 0 1 2 3 4 5 6 7 8
Mileage (km) Age (years)
Source: Citi Research, Tesla Motors Club Source: Accelerated Reported Battery Capacity Loss in 30 kWh Variants of the Leaf
8
However, the battery life issues at Nissan Leaf seem to have already dented
the credibility of electric car batteries. The state of health of the 24k-Wh and 30-
kWh Leaf batteries has been shown to decline by 3% and 7% per year,
respectively, after one year (Figure 46). Nissan has stated they are aware of battery
degradation issues in some models and are investigating, though it is concerning
that the larger battery has shown a much worse rate of decline. The latest 40-kWh
battery and the 60-kWh battery, which will be launched in 2019, are likely to improve
on this level of degradation, but the issues with earlier models have already
contributed to consumer concerns over battery longevity.
Vehicle-to-grid (V2G) charging could offer a solution by curtailing capacity

fade. V2G technology transfers energy from EV batteries to power the grid when
energy is not required by the vehicle, such as when a car is left to charge overnight.
This could present a significant benefit to the system through increased flexibility as
well as increasing the use of renewable energy and helping to address the
intermittency problem, especially if one considers that over 90% of vehicles are
parked at any given time. Further, a recent study has shown the process can
9
improve battery life by 9% over a year, suggesting that the decline in the state of
health of a battery is not irreversible and battery performance can be improved
without the need to pay for a replacement.
8
Myall, D., Ivanov, D., Larason, W., Nixon, M., and Moller, H. Accelerated Reported
Battery Capacity Loss in 30-kWh Variants of the Nissan Leaf. Preprints 2018.
9
Uddin, K. et al. On the possibility of extending the lifetime of lithium-ion batteries
through optimal V2G facilitated by a flexible integrated vehicle and smart-grid system.
Energy, 2017.
© 2018 Citigroup
Figure 47. EV Battery Warranties Available on Popular Models

Brand Model Warranty Duration Warranty Distance Capacity
(yrs) ('000 miles) Threshold (%)
Kia Soul EV 10 100 70
BMW i3 8 100 70
Chevrolet Bolt 8 100 60
Mercedes B250e 8 100 70
Nissan Leaf 8 100 70
VW e-Golf 8 100 70
Source: Citi Research, fleetcarma
Carmakers hope to ease consumer concerns over battery longevity by

offering warranties. In order to reinforce confidence in the longevity and quality of
their products, some carmakers offer warranties on the batteries contained within
their EVs. A battery is eligible for replacement if it fails once capacity has degraded
below a certain threshold before either a set number of years or a distance driven.
The examples listed in Figure 47 give an indication of the level of confidence
manufacturers have in the lifetime of the batteries in their vehicles. On average, the
warranties offered suggest batteries should retain 70% capacity after 100,000 miles
or 8 years.
© 2018 Citigroup
4. Residual values and TCO: “Why Should I Pay over 30%

Higher List Price for an EV?”
BEV residuals are weak, but improving. As well as having a higher entry price,
BEVs also suffer from worse three-year residual value assumptions. If we exclude
the premium end of the market, the residual values of BEVs are 6 percentage points
lower than their ICE alternative (Figure 48). As BEV technology improves and is
more widely understood by consumers, we expect residuals will improve.
Interestingly, when we compared the change in residual value assumptions
between June 2016 and November 2017 (Figure 49), we observed that BEV
residuals deteriorated less than their ICE alternatives. This may be the first signs of
improving BEV residuals, which is crucial for total cost of ownership (TCO).
Figure 48. 3-Year Residuals: Selection of Popular BEVs vs. ICE Figure 49. Change in 3-Year Residual Value (%, Nov-17 vs Jun-16)
Alternatives
60% 3%
50% 2%
1%
Residual Value %
40%
0%
30%
-1%
20% -2%
10% -3%
-4%
0%
VW Golf
Audi Q7
Tesla X (P)
Ford Focus
Nissan Pulsar
Ford Focus BEV
Tesla S
BMW 5er
Tesla X
Audi SQ7
VW e-Golf
Nissan Leaf
-5%
-6%
Tesla BMW VW e- VW Golf Ford Ford Nissan Nissan
Model S 5er Golf Focus Focus Leaf Pulsar
BEV
Source: Company data, Whatcar Source: Company data, Whatcar
Price is the single-biggest barrier, and it is clear to see why that is. The results
of the surveys we analyzed showed that price was the greatest hurdle for BEV
adoption. We compared the prices of a selection of popular BEVs in the U.K. with
their internal combustion engine alternative (see Figure 50). If we strip out the more
premium end of the market (e.g., Tesla Model X / Audi Q7), the list price of an ICE
alternative is 33% below that of its BEV equivalent (see Figure 51).
Figure 50. List Price of a Selection of Popular BEVs vs. ICE Alternatives Figure 51. ICE Alternative Price Relative to BEV Model
160,000 Price (£) 100%
140,000 90%
80% 74%
120,000 70%
70% 65%
100,000 59%
60% 52% 53%
80,000 50%
60,000 40%
30%
40,000
20%
20,000 10%
0 0%
VW Golf
VW Golf
Audi Q7
Tesla X (P)
Ford Focus
Nissan Pulsar
Audi Q7
Tesla X (P)
Ford Focus
Nissan Pulsar
Ford Focus BEV
Ford Focus BEV

Tesla S
Tesla X
Nissan Leaf
Tesla S
BMW 5er
Tesla X
Audi SQ7
VW e-Golf
Nissan Leaf
BMW 5er
Audi SQ7
VW e-Golf
Source: Company data, Whatcar Source: Company data, Whatcar
© 2018 Citigroup
The total cost of owning a BEV is not as high as you might think. Given the
large delta in price and lower residual-value assumptions, there is a perception that
BEVs are not priced attractively (compared with their ICE alternatives). We
analyzed the total cost of ownership over three years of several models (Figure 52),
and found the delta to be lower than what is commonly perceived. In fact, the
electric versions of the VW Golf and Ford Focus are 3% and 16% cheaper than
their ICE versions, on a three-year view. Our calculation assumes the average
numbers of miles driven per year to be 7,500 (which in 2016 was the average
distance traveled per year for a private owned car in the U.K.). In Figure 53 we
show how much of the total cost of ownership (over three years) relates to vehicle
depreciation and fuel cost. For the BEVs, vehicle depreciation accounts for 97% of
the three year TCO, while for the ICE alternative, depreciation accounts for 63% of
the TCO.
Figure 52. 3-Year Total Cost of Ownership: Various Models Figure 53. 3-Year TCO Split by Vehicle Depreciation & Fuel Cost
3 year cost Vehicle % of TCO Fuel % of TCO
60,000 (£) 100% 2% 2% 1% 4% 4% 3%
50,000 27%
89,268
35% 33% 37%

52,528
80% 47% 43%

44,600
40,000
39,850
60%
35,071
34,462
30,000 98% 98% 99% 96% 96% 97%

40%
26,218
73%
24,757
20,000 65% 67%

22,790
63%
22,091
22,085
57%
20,067
53%
10,000 20%
0 0%
Tesla Model S
Tesla Model X
VW Golf
Audi Q7
Ford Focus
Nissan Pulsar
Tesla Model X (P)
Ford Focus (BEV)

BMW 5er
Audi SQ7
VW e-Golf
Nissan Leaf
Tesla Model S
Tesla Model X
VW Golf
Audi Q7
Ford Focus
Nissan Pulsar
Tesla Model X (P)
Ford Focus (BEV)

BMW 5er
Audi SQ7
VW e-Golf
Nissan Leaf
Source: Citi Research, Whatcar, eOn, DfT Source: Citi Research, Whatcar, eOn, DfT
We are cognizant that an increasing proportion of new vehicle purchases are via a
finance plan (i.e. loan). In Figure 54 we show the funding-gap between a BEV and
ICE to be on average of 26% of the list price of a BEV; therefore, purchasing an ICE
saves 39% of the vehicle’s list price. The trade-off is the cost of fuel, and in Figure
55 we show how much three years of fuel costs vs. the ICE list price. Our analysis
suggests that an ICE remains ~5% cheaper than a BEV on a three-year view
(typical lease length), but we expect this to fall as battery costs fall.
© 2018 Citigroup
Figure 54. BEV vs. ICE - Funding the Gap Figure 55. 3-Year Fuel Cost as % of ICE List vs. Saving From Buying an
ICE
60%
100% LIST PRICE: -33%
100%
50%
90%
32% 39%
80% 67% 40%
70% 35%
60% 30%
25%
50%
40% 20%
68%
30% GAP OF -26%
20% 42% 10%
10%
0% 0%
EV ICE BMW Audi Audi VW Ford Nissan Average Funding
5er Q7 SQ7 Golf Focus Pulsar vs ICE
Funding Residual Value list price
Source: Citi Research, Whatcar, eOn, DfT Source: Citi Research, Whatcar, eOn, DfT
Figure 56. Dr Menahem Anderman, PhD Deep Dive into Battery Cost
Figure 57 below presents EV-battery cost curves between 2012 (best estimate) and
2024 (projected). The associated key battery design aspects and production volume
are tabulated in Figure 58. Battery cost can be divided into 4 relatively similar cost
components: (1) the cell cathode; (2) the balance of cell materials; (3)) cell
assembly; and (4) the battery module and pack components and assembly. Pack
cost has come down from about $375/kWh in 2012 to around $200/kWh today. Cost
reduction was achieved by: (1) cell designs with lower power-to-energy ratios; (2)
economy of scale; (3) the use of more energetic materials; (4) the use of lower-cost
solutions in the pack; (5) the maturity of the manufacturing process; and (6) the drop
Dr Anderman is a globally renowned in metal pricing between 2013 and 2016.
expert on automotive battery
technology with 35 years of industry Figure 57. EV battery cost-curves $/kWh
experience. He founded Total Battery
400 $/kWh
Consulting Inc. in 1996 and Advanced
Automotive Batteries Inc in 2000, and is 350
President of both companies. He holds
a PhD in Physical Chemistry from the 300
University of California and has written
250
extensively on the EV industry.
200
150
100
50
0
2012 2016 2020 2024
Pack Cell Total Materials Cathode
Source: Total Battery Consulting
© 2018 Citigroup
Figure 58. Key Battery Design Aspects and Production Volume
Year 2012 2016 2020 2024

Pack size kWh 24 42 60 65
Cathode chemistry LMO-NMC 1,1,1: 70-30 NMC 5,3,2 NMC 6,2,2 NMC 8,1,1
Power to energy ratio kW/kWh 5 3 2 2
EV volume* 000 units 62 247 650 1,600
Production volume GWh 1.5 10.4 39 104
*Excluding China
With production volumes ramping up at an annual rate of 30%–50%, we expect a

steady reduction in cost — averaging 6%–10% per year — for all aspects of battery
cost excluding the cathode. The latter has seen a significant rise in the price of key
raw materials, which has resulted in a price increase for current quotations (for the
2020 production year) versus those from 2016 (for 2018–19). Figure 59 exhibits a
cost breakdown for EV packs featuring the common NMC ‘6,2,2’ cathode for 2020
production.
Figure 59. Cost Breakdown for $170/kWh EV Packs Featuring the Common NMC ‘6,2,2’ Cathode
for 2020 Production
Warrentee & SG&A R&D

profit 5% 2%
Pack & 5%
Cathode
module
28%
assembly
7%
Other cell
factory costs
7%
Factory
depreciation
8%
Balance
Pack & materials
module 20%
components
18%
Future cost improvements are likely to mainly come from the same sources as
historical cost reductions, with economy of scale and improvements in battery
design allowing further gains. However, given that the cathode material already
accounts for over 25% of battery costs, with that percentage set to rise as more
scale efficiency is developed, we expect continued margin pressure on cathode-
material producers in order to support overall cost improvement. Larger battery
manufacturers are already reported to be operating on razor-thin margins in order to
gain share.
© 2018 Citigroup
EV penetration beyond regulated or highly-incentivized volumes is largely

dependent on reaching parity or better in terms of total cost of ownership compared
with ICEs, and it assumes that customers will accept a vehicle with a 200–250 miles
range and a refueling time of 30 minutes or longer. Batteries are the single most-
expensive component in an electric vehicle; for example, the 2017 Chevy Bolt
battery is listed at $15,734 (although its cost to GM is not known), representing an
estimated 40% of the vehicle price. Greater BEV and PHEV penetration is above all
limited by higher relative cost vs. ICE; the issues of dependence on subsidy and
high total cost of ownership are primarily caused by this cost disparity. Falling
battery costs as well as increasing battery energy per unit volume and increasing
charge rate are critical to achieving competitiveness on an unregulated basis.
For similar power levels, the cost of the all-electric powertrain, excluding the battery,
is currently similar to that of conventional gasoline engines and is about $1,500–
$2,000 lower than that of the advanced diesel powertrains used in premium
European brands. Thus, cost parity with ICE powertrains at the vehicle point of
purchase for an EV with a 60-kWh battery is not achievable. At our projected future
battery pricing of $130/kWh, the equivalent total cost of ownership for the European
market at current European fuel and electricity prices will require fuel amortization
over 4–6 years, and for the U.S. market, over 10 years. This drives the battery price
targets communicated by automakers to below $100/kWh.
Figure 60 and Figure 61 below provide our best estimates for production volume in
units and GWh for the key electrified vehicle architectures projected for 2020. Note
the dominant position of China in the EV unit count and the dominant aspects of
EVs on MWh consumed. In dollar value, the full electric vehicles present the largest
opportunity for the battery business including battery and cell production as well as
cell materials and pack components.
Figure 60. Projected Global xEV Production Volume, ‘000 Units Figure 61. Projected 2020 Global xEV Battery Demand, GWh
ROW EV Buses HEVs Strong,

Household China, 145 3.5 HEVs Mild, 0.5
EVs, 652
PHEV all, 14
China
HEVs Mild, Household
760 EVs, 42
HEVs Strong,
3,175
EV Buses
China China, 31
Household
EVs, 1,108
PHEV all, ROW

1,115 Household
EVs, 32
Source: Total Battery Consulting Source: Total Battery Consulting
© 2018 Citigroup
Demand-Driven Factors
It’s clear from volumes that we are still in the early stages of EV adoption. In
2017, BEV and PHEV vehicles combined accounted for just 1% of new car sales
globally. While we expect this to increase to 18% by 2030, the path to full EV
adoption is certainly a long one. We categorize consumers into five types in Figure
62 and would argue that today’s EV owners are the “innovators” on the long journey
to full EV adoption.
For us to move along the curve towards full adoption, demand will need to inflect
among the “early majority” and “late majority” consumers. We’re not there yet as
demand for electric vehicles remains weak, and we would argue that, particularly in
the early stages of development, the EV market is being driven predominantly by
supply factors.
Figure 62. There’s a Long Way To Go Before Reaching Full EV Adoption Figure 63. Intentions Are Not Translating into Purchases Just Yet
50% Would you consider purchasing an EV?
100
Laggards
40%
80
BEV Penetration (%)
60 30%
Late majority
45%
40 20%
35%
Early majority 30%
20
10%
Innovators Early adopters
1.6% 1.9% 0.8%
0
Time 0%
2017 2020e 2025e Germany U.K. U.S.
2030e 2030e (Bull case) 2030e (Bear case) Yes 2017 EV Adoption
Source: Citi Research estimates Source: Citi Research, McKinsey, CarGurus, ACEA, LMC
The public is warming to the idea, but barriers to adoption are inhibiting
demand. Although interest in electric vehicles has been growing and consumers
are increasingly open to the idea of buying an EV (Figure 63), this is yet to translate
into meaningful sales figures. The main barriers to adoption discussed earlier
(range, infrastructure, battery longevity, residual values) clearly contribute to this
disparity, and the relatively limited range of EV products currently available is also
likely to be impeding demand.
Fleet buyers are a potential catalyst for driving EV demand. While an

individual’s decision to buy a new car is determined by a range of factors including
more subjective criteria such as brand loyalty and public perception, fleet buyers are
more rational actors, driven by economics and costs. They therefore offer a
potential catalyst for adoption rates whereby demand inflects among fleet buyers
when it makes financial sense to switch to an electric fleet. This is not to say cost
parity is not also a key driver among individuals, but rather that it is the main driver
for fleet buyers and a point that is often overlooked despite the fleet market
10
accounting for roughly 23% of annual new car sales in Europe.
10
Peugeot, Ford, Renault lead in high-margin sales to private customers. Automotive
News Europe, June 2017. link
© 2018 Citigroup
Will Consumers Pay?
Ingrained buying habits present a challenge for EVs. We have long argued (see
here) that consumers are generally open to paying for power, but less willing to pay
for fuel efficiency. This is especially true when fuel prices are low, as they are today.
As such, OEMs have historically been reluctant to add fuel-saving content to
vehicles as there is limited financial return to them for doing so. We analyzed the
relationship between price and power (bhp) and that between price and CO2
emissions, and the results support our view (see Figure 64 and Figure 65). In our
analysis we captured the data for a range of engine sizes/types for the following
models: BMW 3-series, Ford Focus, Mercedes C-class, Nissan Qashqai, Renault
Clio and VW Golf.
Figure 64. Relationship Between Vehicle Price and Power Figure 65. Relationship Between Vehicle Price and CO2 Emissions
300 Price of vehicle- £

Power
(bhp) 0 10,000 20,000 30,000 40,000 50,000
80
250
90
R² = 0.6683
200 100
110
150
120
100
130 R² = 0.1829
50 140
150
0 CO2
0 10,000 20,000 30,000 40,000 50,000 160 (g/km)
Price of vehicle- £
Source: Citi Research, WhatCar (June 17 issue) Source: Citi Research, WhatCar (June 17 issue)
Battery costs are an obstacle for price-reductions. The above analysis

reinforces our view that the primary hurdle to mass EV-adoption is the cost of
batteries. If the average consumer is not willing to pay for fuel efficiency (as above)
it must be offered at little (or no) additional cost, and while battery costs are
prohibitively high (as economies of scale are not being attained) it is nigh-on
impossible for manufacturers to reduce the selling price of their electric vehicles
(without significantly damaging their financial performance). We believe list prices
need to fall to entice the average consumer, and until this happens electric vehicles
will remain in the ‘early-adopters’ phase.
© 2018 Citigroup
How Many EVs Are Required?

Model for European Powertrain Mix
Lower diesel penetration creates a major CO2 challenge for the industry. The
future may well be electric — maybe not wholly, but an increasing proportion of new
car sales will either be full electric (we see 6% by 2025) or partial electric. That
means the carmakers are still heavily reliant on diesel to deliver the CO2 savings
required to meet 2021 EU targets. However as we mapped out in this report, we
think diesel penetration will continue to fall. That begs the question of how many
EVs need to be sold to reach the emission targets, or conversely what is the
efficiency hurdle for gasoline engines given their popularity looks set to rise. We
have modeled this and present our findings below.
Model for European Powertrain Mix in 2020/2021, 2025, and 2030
We have built an industry model to gauge the magnitude by which either full-battery
electric vehicle (BEV) penetration needs to rise, or the efficiency of gasoline
engines needs to improve in order to meet the upcoming CO2 targets. We also
include the impact of super-credits {Note: in 2020 the carmakers will benefit from
super-credits for producing vehicles with < 50g/km]. Each vehicle that emits less
than 50g CO2 will be counted as two cars in 2020, 1.67 in 2021, and 1.33 in 2022.
The super-credits will be phased-out by 2023, when no multiplier will be awarded for
the sale of low-emission vehicles.
Figure 66. Europe: What Would BEV Penetration Need to be to Meet the Upcoming CO2 Targets?
Europe NEDC WLTP
Powertrain Mix 2015 2020 2021 2021 2022 2023 2024 2025 2030
Diesel 52% 33% 31% 31% 28% 26% 24% 21% 19%
Gasoline 46% 60% 62% 62% 62% 59% 59% 59% 52%
PHEV 1% 1% 2% 2% 2% 3% 3% 4% 9%
BEV 1% 5% 5% 5% 8% 12% 14% 15% 20%
Super-credits 2.00 1.67 1.67 1.33 1.00 1.00 1.00 1.00
CO2 (g/km) 2015 2020 2021 2021 2022 2023 2024 2025 2030
Diesel 108 92 90 104 101 99 97 96 86
Gasoline 135 115 113 129 127 124 122 119 108
PHEV 46 43 43 49 49 48 48 47 45
BEV 0 0 0 0 0 0 0 0 0.0
Fleet 119.5 95.0 95.0 109.2 104.9 100.7 96.7 92.9 76.5
Assumes a factor of 1.15 as we move from NEDC to WLTP

Assume that the fuel efficiency improvement for diesel and gasoline vehicles is 3.2% per year until 2020 (this is the same magnitude of improvement as seen between 2010 and
2015). Until 2020 we assume that PHEV efficiency also improves, but to a lesser extent (~1.2% per year)
In 2021 and beyond we assume fuel efficiency is harder to come by and assume diesel and gasoline fuel efficiency improves by 2.0% /year. For PHEVs we assume 0.8%/ year.
Source: LMC, IHS, Citi estimates. Yellow highlight indicates the balancing figure obtained by use of ‘goal-seek’, solving for fleet emissions
All else equal, we estimate that in order to meet the 2021 95g/km CO2 target the
penetration of BEVs will need to increase to 5% (from less than 1% today), which
compares to LMC’s expectation of ~4% BEVs in 2021. As we have outlined in this
report, there are a number of factors that are critical for greater adoption of electric
vehicles that still need to be overcome, and in that context a figure of 5% seems a
stretch. While we are in no doubt that electric vehicle penetration will accelerate in
the next decade, we are skeptical about the pace of ramp in the next few years. The
end of 2021 is a little over three years away, and while we estimate BEV volumes
will triple in that time, we expect it to still only account for a little over 2% of sales.
© 2018 Citigroup
Interesting observations from the analysis in Figure 66:
 Despite the reduction of super credits in 2021, BEV penetration need not rise as
the overall target remains 95g.
 The phasing out of super credits in 2022 and in 2023 has a material impact on
the number of BEVs required to achieve the fleet target (jumps 3–4 percentage
points per year).
Figure 67. Europe: How Much Would the Efficiency of Gasoline Engines Need to Increase to Meet the Upcoming CO2 Targets?
Europe NEDC WLTP
Powertrain Mix 2015 2020 2021 2021 2022 2023 2024 2025 2030
Diesel 52% 33% 31% 31% 28% 26% 24% 21% 19%
Gasoline 46% 64% 65% 65% 67% 68% 68% 68% 58%
PHEV 1% 1% 2% 2% 2% 3% 3% 4% 9%
BEV 1% 2% 2% 2% 3% 4% 5% 6% 14%
Super-credits 2.00 1.67 1.67 1.33 1.00 1.00 1.00 1.00
CO2 (g/km) 2015 2020 2021 2021 2022 2023 2024 2025 2030
Diesel 108 92 90 104 101 99 97 96 86
Gasoline 135 104 105 120 114 109 105 103 96
PHEV 46 43 43 49 49 48 48 47 45
BEV 0 0 0 0 0 0 0 0 0
Fleet 119.5 95.0 95.0 109.2 104.9 100.7 96.7 92.9 76.5
Gasoline fuel efficiency improvement (% per year) 4%* -1% 5% 5% 3% 2% 1%**
Assumes a factor of 1.15 as we move from NEDC to WLTP

* per year improvement in the period from 2015 to 2020 / **per year improvement in the period from 2020 to 2025
Source: LMC, IHS, Citi estimates. Yellow highlight indicates the balancing figure obtained by use of ‘goal-seek’, solving for fleet emissions
Required improvements in gasoline-efficiency are too tough to achieve target.

Assuming 2021 BEV penetration is as we predict (i.e., 2%), the efficiency of
gasoline engines needs to improve by 23% (or >4% per year) in the period to 2020
in order to meet the CO2 target. In 2022 and 2023 fuel efficiency will need to
improve by 5% (assuming BEV + PHEV penetration rises by only 100 basis points)
as super credits are phased out. For reference the European-wide fleet reduced
11
CO2 by 1.2% (from 119.5g to 118.0g) in 2016, but rose 0.4% in 2017 (to 118.5g )
as lower diesel limited the progress. On the basis of the historic evolution of fuel
efficiency, we do believe it will be possible to achieve the CO2 targets by exclusively
relying on an improvement here. To reach the targets we expect a mixture of both
EV penetration and improvement in combustion engine efficiency will be required.
Given the ever reducing popularity of diesel-cars (~20% more efficient than gasoline
equivalents), we believe that technologies such as 48-volt (which offers a 10%–20%
fuel efficiency improvement) will have to become far more prevalent if carmakers
are to achieve efficiencies of the magnitude that our analysis suggests is required.
Interesting observations from the analysis in Figure 67:
 Due to the increase in BEV and PHEV penetration in 2021 the efficiency of
gasoline in 2021 can deteriorate by 1% (vs 2020) and still meet the 95g target.
 Beyond 2023, the magnitude of improvement for the fuel efficiency of gasoline
vehicles is not particularly difficult (2.8% per year in 2024 and 2.1% in 2025),
provided PHEV and BEV penetration rise by ~100 basis points in each year.
11
https://www.eea.europa.eu/highlights/no-improvements-on-average-co2.
© 2018 Citigroup
 It is worth bearing in mind that the shift from NEDC to WLTP will make efficiency
improvements more difficult to achieve.
We have detailed the changes to the new-vehicle testing procedures in Appendix 6:

Navigating the Shift from NEDC to WLTP.
It’s all in the mix. It is worth noting our model assumes mix remains constant, and
so it is positively biased in that regard. We are, however, cognizant that OEMs are
making a conscious effort to downsize engines, thereby improving the fuel efficiency
of their fleet. This may be offset by the continued rise in demand for larger, and
therefore heavier, vehicles. SUV popularity (see Figure 69) and model offerings
mean the hurdle may even be higher than what we have derived.
Figure 68. The Engine Downsizing Trend Will Help Reduce CO2 Figure 69. Model Launches by Segment Shows a Raft of Upcoming
Emissions SUVs
160 160
150 140
No. of global model launches

120
140
CO2 (g/km)
100
130
80
120
60
110
40
100 20
800 1000 1200 1400 1600 1800 2000 2200
Engine Capacity (cc) 0
B-Segment C-Segment SUV Other
VW Golf Ford Mondeo BMW 3er 2017 2018 2019 2020
Source: Citi Research, www.parkers.co.uk Source: Citi Research, IHS
© 2018 Citigroup
Model for China Powertrain Mix

The Chinese have made it clear that they want to be global leaders in electric
vehicles, and they have set ambitious targets for the number of New Energy
Vehicles (NEVs) they want to sell in 2020 and 2025: 2 million and 7 million. To
encourage supply of NEVs the Ministry of Industry and Information Technology
(MIIT), the country’s regulatory body, has set for manufacturers an ‘NEV target
score’. The methodology for calculating the NEV target score is as follows:
– A) In 2018, 2019, and 2020 the percentage requirement is 8%, 10%, and 12%
respectively. The percentage beyond 2020 is to be formulated separately.
– B) The percentage (A) is applied to the total ICE passenger car production for
the corresponding year = “NEV target score”
– C) The actual NEV score is determined by applying a multiple to the volume of

NEVs produced/ imported. [Note: BEVs with a driving range in excess of
100km have a higher multiple than plug-in-hybrids.]
Figure 70. Dual Credit Management System

Regulatory authority Ministry of Industry and Information Technology (MIIT)
Two parallel system CAFC credit NEV credit
Calculation method CAFC credit = (Target CAFC - Actual CAFC) x # of vehicles NEV credit = NEV point / vehicle x # of vehicles
BEV with R>50km has multiplier impact of 5x/3x/2x in 2016-17/2018- NEV point per BEV = R x 0.012 + 0.8 (cap at 5)
19/2020 when calculating CAFC
PHEV with fuel consumption < 2.8L/100km has multiplier impact of NEV point per PHEV = 2
3.5x/2.5x/1.5x in 2016-17/2018-19/2020 when calculating CAFC
Management method - CAFC negative credit can be offset by CAFC positive credit earned - NEV negative credit can only be offset by NEV positive points via
from previous year, transferred from related corporates, or by NEV purchases from other manufacturers
positive credit
- CAFC credit is allowed to be carried forward for at most 3 years (with - NEV credit can trade freely on MIIT's platform, but cannot be re-sold
80% conversion ratio in 2018 and 90% in 2019 onwards) and can be
transferred within related corporates (shareholding at or more than 25%)
- NEV credit is not allowed to be carried forward
- The 2019/20 NEV balance will be examined together
Assessment companies All PV OEMs selling in China (including import) All PV OEMs with annual production volume or import volume greater
than 30k units in China
Assessment criteria A positive balance under GB 27999-2014 2019/2020 NEV point to # of non-NEV vehicles ratio at 10%/12%
Assessment period 2016+2017, 2018, 2019, 2020 2019+2020
Penalty measures Suspend application of car models that do not meet GB27999 standard Suspend partial production of gasoline models
and suspend partial production of high fuel consumption models
Source: MIIT, Citi Research
In addition to the NEV targets, the MIIT has also set a target to reduce the overall
fleet’s fuel consumption by ~28% by 2020 (to 5L/100km vs 6.9L/100km in 2015);
this is referred to as Corporate Average Fuel Consumption (CAFC). Each OEM has
a specific target, and like in Europe, the more fuel-efficient vehicles benefit from
super-credits.
© 2018 Citigroup
Model for Chinese Powertrain Mix in 2020 and 2025
We calculated the sensitivity of BEV volumes in China by modeling two different

scenarios: (1) how many BEVs are required in China to comply with the 2020 and
2025 CAFC targets; and (2) assuming the fleet reaches the required NEV score (in
2020 and 2025), how much does fuel efficiency need to improve. We present the
results of our sensitivity in Figure 71 and Figure 72.
Figure 71. Scenario 1: Solving for 2020 and 2025 CAFC Targets (number of NEVs)
2015 2016 2020 2025
Fuel consumption (L/100km) 7.02 6.65
vs 2012 (annual improvement) 2.3% 3.8%
After credits 6.67 6.44
Calculated fuel consumption 6.78 6.42 5.13 4.10
NEV multipliers 5 5 2 1
TARGET 6.90 6.70 5.0 4.0

vs 2015 (annual improvement) -6.2%
vs 2012 (annual improvement) -2.9% -5.0%
NEV vehicles (million) 2.0 7.0
Total Pas-Cars ('000) 21,248 24,788 31,181 35,313

o/w conventional fuel 21,064 24,448 29,181 28,313
NEVs 185 340 2,000 7,000
BEV 124 245 1456 5290
PHEV 61 94 544 1,710
Gasoline 99.1% 98.6% 93.6% 80.2%

BEV 0.6% 1.0% 4.7% 15.0%
PHEV 0.3% 0.4% 1.7% 4.8%
Fuel efficiency
Gasoline 7.08 6.86 5.80 5.02
BEV 0 0 0 0
PHEV 1.96 1.93 1.79 1.67
NEV "credit" targets 12% 22%

NEV credit requirement 3502 6229
NEV score 5449 26350

BEV 2.99 4.33
Range (km) 183 295
PHEV 2.0 2.0
Source: Citi Estimates, CAFC
Yellow highlight indicates the cells we used to ‘seek our goal’
In deriving our forecast for BEV sales in 2020, we assume gasoline fuel efficiency
will improve by 4% per year in China (from 2016 until 2020). It’s worth noting the
efficiency of gasoline vehicles in China improved at a rate of 1.8% per year in the
period from 2006 until 2015. And for reference we assume the improvement in
Europe will be ~3% per year over the same period. Our model tells us that in order
to meet the fleet-wide fuel consumption target of 5L/100km (in 2020), ~1.46 million
BEVs must be sold. To reach the 2025 target of 4L/100km, 5.3 million BEVs will
need to be sold. It is possible that with higher penetration of fuel-efficiency
technologies (e.g., turbochargers enabling engine downsizing, 48-volt systems,
direct injection), the efficiency improvement may be greater than our base-case
assumption, which will reduce the number of NEVs that need to be sold.
© 2018 Citigroup
Interesting Observations from the Analysis in Figure 71:
 The target of 2 million and 7 million NEV vehicles to be sold in 2020 and 2025
seems quite optimistic in the context of the government’s NEV credit
requirements.
Figure 72. Scenario 2: Solving for NEV Score and Required Fuel Efficiency Improvements
2015 2016 2020 2025
Calculated fuel consumption 6.78 6.42 5.00 4.00
NEV multipliers 5.00 5.00 2.00 1.00
TARGET 6.90 6.70 5.00 4.00
Total Pas-Cars ('000) 21,248 24,788 31,181 35,313

o/w conventional fuel 21,064 24,448 29,806 32,731
NEVs 185 340 1,375 2,582
BEV 124 245 831 872
PHEV 61 94 544 1,710
Gasoline 99% 99% 96% 93%

BEV 1% 1% 3% 2%
PHEV 0% 0% 2% 5%
Fuel efficiency
Gasoline 7.08 6.86 5.43 4.23
BEV 0.00 0.00 0.00 0.00
PHEV 1.96 1.93 1.79 1.67
Improvement in consumption 23% 22%

vs 2015 (p.a improvement) 5.2% 5.0%
vs 2020 (p.a improvement) 4.9%
NEV "credit" targets 12% 22%

NEV credit requirement 3577 7201
NEV score 3577 7201

BEV 2.99 4.33
Range (km) 183 295
PHEV 2.00 2.00
Source: Citi Estimates, CAFC
Yellow highlight indicates the cells we used to ‘seek our goal’
Calculating the required improvement in fuel efficiency in order to meet the fleet-
wide fuel consumption targets (5L/100km in 2020 and 4L/100k in 2025) cannot
ignore the stated NEV credit requirements, and so this is the first-step in this model.
[Note: We assume the average BEV range improves to 183km by 2020 and to
~295km by 2025. Given diesel cars are not relevant for the Chinese market we can
ignore these, and simply calculate by how much the efficiency of gasoline vehicles
need to improve to reach the CAFC targets. We calculated that by 2020 fuel
consumption needs to improve by 5.2% per year (recall it improved by an average
of 1.8% per year in the 9 years to 2015), and by ~5% per year in the five years to
2025.
Interesting Observations from the Analysis in Figure 72:
 Despite the precondition of this model being the NEV credit requirement, the
number of NEVs in 2020 is 31% below the target (~1.4 million vs target of 2
million), and more than 63% below in 2025 (~2.6 million vs target of 7 million).
© 2018 Citigroup
The Future of Electric Vehicles

In the research for this report, we spoke with many experts, one of whom was Carl
Sanderson, Former GM of BMW i-brand. Mr. Sanderson confirmed that an electric
vehicle is less complex, insofar that it only has 200 moving parts vs. 2,000 in a
conventional vehicle. At this stage of development, lower complexity does not mean
lower cost. The biggest hurdle for the mass-production of EVs is the cost of
batteries, but if we are to assume this is inevitable then the question that follows is
how much additional room for improvement exists? We considered a handful of the
components in an EV with the aim of understanding the potential for improvement,
and the impact this could have on performance.
Figure 73. The Evolution of Electric Motors
TODAY TOMORROW
• Our research suggests most motors are • Directly wound motors offer a 20% space
indirectly wound and in-house by OEMs saving and use ~50% less copper wire
• We think this is largely as a function of • In one test case, one of these motors resulted
scale (i.e., low EV volumes) in a 17% better performance
• It is cost effective at >100k units per year
Source: Aumann, Citi Research
© 2018 Citigroup
Figure 74. The Evolution of Battery Thermal Management Systems
TODAY TOMORROW
Battery
Battery
Cell
Battery
Battery Chiller – Input and
Supplemental Electric Refrigerant and Output
Battery Heater - Glycol Glycol
• Traditional engines required heating • Nylon tubes offer a 30-60% weight

and cooling, and therefore needed more saving vs conventional tubes
sophisticated tubing
• The coolant is not heated and
• The current EV generation utilize these therefore does not require heat
tubes (to maximize economies of scale) resistant tubes
Source: TI Fluids, Citi Research
Figure 75. The Evolution of Power Electronics
TODAY TOMORROW
• The components used today are primarily • Silicon Carbide enables a smaller form factor and
made using Silicon instead of the more less weight, higher efficiency, and density vs.
complex compound semiconductor materials Silicon-based products.
such as Silicon Carbide (SiC) for example. • Product wise, Silicon Carbide allows 50% volume
• The reason for the widespread use of Silicon reduction for onboard chargers and ~5% efficiency
is because of relatively low cost and high gains in real-life driving cycle in Main Inverter.
availability of substrates.
Source: Infineon, Citi Research
© 2018 Citigroup
Figure 76. The Evolution of Battery Technology
TODAY TOMORROW
• Battery technology is ever-evolving; the • Solid state cells can store double the energy
convention today is for a liquid • The cells are non-combustible
electrolyte
• They are ~75% smaller and also much lighter
• Battery manufacturers currently favor
NMC due to its higher energy density • Cathode of the future will likely have higher
nickel content
• Today anodes are mostly made of
graphite (which has a limited capacity) • Anode of the future will be made of silicon
Source: Bosch, Citi Research
Solid-State Batteries Will Help Overcome Range Issues
Of these four components, we believe solid-state batteries to be the biggest “game-

changer,” insofar that they will directly address some of the main barriers to
adoption: range, charging time, and battery degradation. We dig a bit further into
this below.
All-solid-state batteries are a leading candidate for mainstream next-

generation battery technology. Current lithium-ion batteries are made up of the
cathode, the electrolytic solution, the separator and the anode, whereas the
electrolyte is solid in a solid-state battery. In fact, all components and materials are
solid, hence the “solid-state” terminology. The properties of all-solid-state batteries
will depend on which materials are used, but research to date reveals clear potential
in terms of safety, resistance to leakage, resistance to combustion (simplified
cooling structure), miniaturization, flexibility of design in terms of direct layer
formation for cells, relative long discharge cycle lifespan, lack of degradation thanks
to good high/low temperature properties, short charge times, high energy density,
and high power density. In the past, low power density has been seen as a
weakness, but the Tokyo Institute of Technology and the Toyota Group’s research
team have together developed an all-solid-state battery with three times the power
density and twice the energy density of existing lithium ion batteries. We think that
all-solid-state batteries have the potential to overcome the disadvantages of EVs.
© 2018 Citigroup
Figure 77. Lithium-ion Battery Versus All-Solid-State Battery
－＋－＋
Negative Positive Negative Positive
electrode electrode electrode electrode
All-solid-
state
Li+ Li+ X- Li+ Li+ Li+ Li+ Li+ Li+ Li+ Li+ Li+
Li+ X-
Li+ Li+ Li+ Li+ Li+ Li+ Li+ Li+ Li+ Li+
X- Li+
Inorganic solid electrolyte

Organic electrolyte Separator
Source: Nature Energy, Citi Research.
Figure 78. Battery Power and Energy Density
All-solid-state
output
battery
High
1000
Super capacitor
Lithium-sulfur battery
100
Magnesium
Power density (kW/kg)
Lithium-ion battery battery

10
Aluminum-ion
1 battery
Sodium-ion battery Lithium-oxygen
battery
0.1
1 10 100 1000 10000
Cell energy density (Wh/kg) High energy
Source: Joint press release by the Tokyo Institute of Technology and Toyota (March 17, 2016), Citi Research.
© 2018 Citigroup
Figure 79. All-Solid-State Battery Potential and Hurdles to Commercialization

All-Solid-State Battery Potential
- All-solid-state batteries are highly fire-resistant and very safe as they have no leakage
- Lithium-ion is the only molecule that moves within the electrolyte and side-effects on the positive electrode surface are therefore unlikely. This lengthens battery life.
- Performance deterioration is limited across a broad temperature range (works well in low and high temperatures)
- Resistance to high temperatures allows the cooling structure to be simplified
- Cell design allows a high degree of flexibility
- Cells can be stacked directly on top of one another without packaging
- Discovery of superionic conductors enables high diffusivity and high output
- Energy density can be increased because batteries can be used at high voltage
- High-speed charge/discharge that is not possible with electrolytes; shortens discharge time
Hurdles to Commercialization
- Reducing interface resistance between electrodes and solid electrolytes
- Cell development process
- Sulfide solid-state electrolytes offer high conductivity but measures to deal with the hydrogen sulfide that results are an issue ⇒ additives being mixed in and research being
undertaken into blends with oxidized materials
- Demonstrating the reliability required for use in vehicles
- Establishing a battery pack structure
- Developing low-cost mass production technology
Source: Various materials, Citi Research.
How Do We See the All-Solid-State Battery Market Developing?
Based on the timing of development, we see little likelihood that an all-solid-state

battery using leading-edge materials capable of realizing the full potential outlined
above will emerge by 2022. When first introduced, the appeal of the batteries is
likely to be in the reduction in battery pack size (increase in cabin space and greater
battery capacity) and safety.
Aside from the aforementioned developments on the part of Toyota, other related
names are making similar in-roads in the solid-state battery space. U.K. household
appliance manufacturer Dyson announced recently that it would put an EV with all-
solid-state batteries on sale by 2020. Some media reports indicate that it plans to
use technologies from Sakit3, acquired in 2015, or from its in-house development
team; it appears the batteries’ technical characteristics are a closely guarded secret.
Reports also suggest that VW, Hyundai, BMW, and others are developing EVs that
use all-solid-state batteries. Audi exhibited a concept car called the Aicon, which
was equipped with all-solid-state batteries at the Frankfurt Motor Show 2017. Audi
says it can travel 800km or more on a single charge and that the batteries can be
replenished to 80% of capacity in 30 minutes or less. It is of course possible that
there will be performance innovations in existing lithium-ion batteries on the shift to
mass production and lower prices, such as at the joint Tesla-Panasonic Gigafactory.
It is conceivable that the all-solid-state battery narrative will be interrupted if there
are innovations with lithium-ion batteries in key areas such as charging times,
range, battery depletion, and costs. As things stand, the possibilities of any future
propulsion technology cannot be completely discounted.
What About Buses and Commercial Vehicles?
This report has largely focused on the future for electric passenger cars, but electric
bus and commercial vehicles (c-EV) also warrant a mention given their relevance to
the EV debate is likely to increase as the number of such vehicles on the road
grows. Penetration in Europe and U.S. is negligible today, but the markets are
expected to grow in the coming years.
© 2018 Citigroup
While the European commercial vehicle market is expected to grow at a rate of 4%

per year (to 2025), the e-CV market is expected to grow by 56% per year. A similar
trend is expected in the U.S., where sales of medium and heavy commercial
vehicles are expected to grow at a rate of 1% per year, while the e-CV market is
estimated to grow at a compound annual growth rate (CAGR) of 30%.
Figure 80. Forecast Commercial Vehicle Sales Growth in Europe and Figure 81. Forecast Bus Sales Growth in Europe and the U.S.
the U.S.
700,000 Units 546k 616k 60,000 Units
45k 50k 43k 53k
600,000 1% 2%
2%
50,000 1%
1%
310k 460k
500,000
4% 40,000
400,000
30,000
300,000
20,000
200,000
100,000 10,000 22%

20%
56% 17,500 30% 36,960 48% 9,010
0 5,000
0
2015 2025 2015 2025 2015 2025 2015 2025
Europe U.S. Europe U.S.
Electric Non-Electric 2015 - 2025 CAGR Electric Non-Electric 2015 - 2025 CAGR
Source: Company presentation, Roland Berger data Source: Company presentation, Roland Berger data
With regards to the buses, in Europe overall bus and coach volumes are expected
to grow 1% per year (to 2025), while e-buses are estimated to grow at a rate of 48%
per year. The same figures for the U.S. forecast market growth of 2% per year and
e-buses estimates to grow at a CAGR of 20%.
Although the e-bus market is in in the very early stages of development in Europe
and the U.S. with 1% and 3% penetration of new bus registrations, respectively, in
2015, in China the e-Bus market is already well established.
Figure 82. China Electric Bus Sales Figure 83. Global E-Bus and Passenger Car Lithium-Ion Battery
Demand
140 Units ('000s) 100 GWh Global passenger EV battery demand
Global e-bus battery demand
90
120 Global EV battery manufacturing capacity
80
100 ~43% of e-buses that received the
subsidy in 2015 were fraudulant 70
38
80 60
116
90
60 50
50 40
40
30
20 13 21
22 20 12
17 19 16
0 10
2011 2012 2013 2014 2015 2016 2017 7 13 16
2 5 4
PHEV buses BEV buses Est. fraud vehicles 0
2012 2013 2014 2015 2016
Source: Citi Research, Bloomberg New Energy Finance, Bloomberg Industries, Source: Citi Research, Bloomberg New Energy Finance, China Automotive
OFweek, EV Sales Blogspot Technology & Research Centre
The Chinese market, supported by very generous subsidies, has already grown
significantly (Figure 82). This in turn is helping drive lithium-ion battery demand as
shown in Figure 83. As a proportion of global EV lithium-ion battery manufacturing
capacity e-bus demand has increased from just 1% in 2012 to 18% in 2016.
© 2018 Citigroup
Citi’s Powertrain Forecasts

Figure 84. Global EV Forecasts by Type (Citi’s Base Case) Figure 85. Global BEV Forecasts by Region
25 Units (mn) 14 Units (mn)
12
20
10
15
8
6
10
4
5
2
0 0
'14 '15 '16 '17 '18 '19 '20 '21 '22 '23 '24 '25 '26 '27 '28 '29 '30 '14 '15 '16 '17 '18 '19 '20 '21 '22 '23 '24 '25 '26 '27 '28 '29 '30
BEV PHEV Europe China U.S. RoW
Source: Citi Estimates Source: Citi Estimates
We forecast global BEV penetration to reach 2% by 2020, 5% by 2025 and 10%

by 2030. If we include PHEVs our estimate rises to 18% in 2030 and when also
including full-hybrids, we expect penetration to reach 22%. We estimate BEV
penetration (of new car sales) will reach 14% in Europe, China, and the U.S. by
2030.
As illustrated in Figure 85, we expect China to continue to be the key region driving
the BEV market in the near to medium term due in part to the size of the market but
also higher penetration rates. European BEV penetration currently lags China by 1
percentage point, and we do not expect it will catch up until 2030.
Figure 86. European EV Forecasts by Type Figure 87. China EV Forecasts by Type
5 Units (mn) Units (mn)
2030 Target
10
4
8 2025 Target
3
6
2
4
2020 target
1 2
0 0
'14 '15 '16 '17 '18 '19 '20 '21 '22 '23 '24 '25 '26 '27 '28 '29 '30 '14 '15 '16 '17 '18 '19 '20 '21 '22 '23 '24 '25 '26 '27 '28 '29 '30
BEV PHEV BEV PHEV
We expect Europe and China to remain the largest EV markets with BEV sales
growing at a CAGR of 27% and 21%, respectively, between 2017 and 2030.
We have been conservative in our base case assumptions for European EV
penetration, insofar that our forecasts suggest the European emissions targets will
not be met without significant improvement in ICE fuel efficiency. Similarly in China,
© 2018 Citigroup
we assume the NEV targets will not be met [Note: full hybrids do not count towards
meeting these thresholds] — see Figure 87.
Figure 88. U.S. EV Forecasts by Type Figure 89. Rest of World EV Forecasts by Type
4 Units (mn) 6 Units (mn)
5
3
4
2 3
2
1
1
0 0
'14 '15 '16 '17 '18 '19 '20 '21 '22 '23 '24 '25 '26 '27 '28 '29 '30 '14 '15 '16 '17 '18 '19 '20 '21 '22 '23 '24 '25 '26 '27 '28 '29 '30
BEV PHEV BEV PHEV
Our 2030 BEV estimate rises to 18% in our bull case and falls to 5% in our
bear. In our bull scenario, we assume that both European emissions targets will be
met (as a result of higher EV penetration) and China NEV targets will be reached,
resulting in 26% BEV penetration in Europe and 24% in China. We also upgrade
our growth rates for the U.S. and Rest of World (RoW) resulting in 30% and 6%
penetration, respectively. Our bear case sees a more significant miss on European
CO2 targets and on China NEV targets, and slower growth in both the U.S. and
RoW.
Figure 90. Bull-Bear Case for Global BEV Penetration Figure 91. 2030 BEV Penetration Bull-Bear
20% BEV 20% BEV
penetration penetration
18% 18%
16% 16%
14% 7.9%
14%
12%
12%
10%
10%
8% 18.1%
8% 5.1%
6%
6%
4% 10.2%
4%
2%
5.1%
0% 2%
'14 '15 '16 '17 '18 '19 '20 '21 '22 '23 '24 '25 '26 '27 '28 '29 '30 0%
Base Bull Bear Bear Base Bull
© 2018 Citigroup
ICE new car sales should continue to dominate, but their market share will
suffer. In Figure 92 we show growing BEV sales at the expense of ICE vehicle
market share. However, in 2030, we still expect more than 90% of new cars sold to
contain an internal combustion engine (including mild, plug-in, and full hybrids). The
penetration of electric vehicles is of course growing, but it is exaggerated to predict
the death of the combustion engine in the near term. Instead, as shown in Figure
93, new cars sold are likely to contain both a combustion engine and also a level of
electrification (i.e., be a full hybrid electric vehicle (FHEV) or plug-in hybrid vehicle
(PHEV)). In 2030, we expect at least 20% of vehicles sold will have a battery pack
(in excess of 60-volts), while only 80% of vehicles will contain conventional ICE
powertrains (of which 20% will have some form of electrical assistance, over and
above the standard 12-volt battery).
Figure 92. BEVs Are Gradually Eating into ICE Market Share Figure 93. Traditional Vehicles as We Know It Are on the Decline
Units (mn)
Units (mn)
12% 100%
120 120
10%
100 100 90%
80 8%
80
80%
60 6%
60
70%
40 4%
40
20 2% 60%
20
0 0% 0 50%
'14 '15 '16 '17 '18 '19 '20 '21 '22 '23 '24 '25 '26 '27 '28 '29 '30 '14 '15 '16 '17 '18 '19 '20 '21 '22 '23 '24 '25 '26 '27 '28 '29 '30
BEV PHEV
ICE (inc. MHEV & FHEV) BEV % of new sales (RHS) New cars sold ICE (inc PHEV) % of new sales (RHS)
Note: MHEV = Mild Hybrid Electric Vehicle Source: Citi Estimates
Source: Citi Estimates
Figure 94. Global Base Case EV Forecasts

Global 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
Volumes ('000s units)
84,041 89,328 93,393 95,371 99,376 100,858 104,744 107,718 110,746 113,858 117,045 118,278 119,488 120,674 121,835 123,020 124,230
% change 6% 5% 2% 4% 1% 4% 3% 3% 3% 3% 1% 1% 1% 1% 1% 1%
ICE 81,723 86,762 90,303 91,450 93,556 93,068 93,572 93,634 92,792 90,514 88,341 85,522 82,623 80,080 77,610 75,604 73,504
Mild-hybrid 386 227 247 397 509 523 530 543 554 566 580 588 597 606 616 627 639
48-volt 0 0 0 0 255 967 2,672 4,418 6,967 10,438 13,380 15,367 17,367 18,803 20,248 21,466 22,703
Hybrid 1,608 1,758 2,018 2,235 3,004 3,547 4,172 4,282 4,393 4,507 4,624 4,568 4,515 4,468 4,424 4,387 4,355
PHEV 137 253 340 482 819 1,221 1,729 2,231 2,783 3,635 4,852 5,956 6,897 7,848 8,677 9,518 10,274
BEV 187 329 485 808 1,233 1,532 2,069 2,610 3,259 4,196 5,269 6,277 7,490 8,869 10,260 11,418 12,755
Penetration %
ICE 97% 97% 97% 96% 94% 92% 89% 87% 84% 79% 75% 72% 69% 66% 64% 61% 59%
Mild-hybrid 0% 0% 0% 0% 1% 1% 1% 1% 1% 0% 0% 0% 0% 1% 1% 1% 1%
48-volt 0% 0% 0% 0% 0% 1% 3% 4% 6% 9% 11% 13% 15% 16% 17% 17% 18%
Hybrid 2% 2% 2% 2% 3% 4% 4% 4% 4% 4% 4% 4% 4% 4% 4% 4% 4%
PHEV 0% 0% 0% 1% 1% 1% 2% 2% 3% 3% 4% 5% 6% 7% 7% 8% 8%
BEV 0% 0% 1% 1% 1% 2% 2% 2% 3% 4% 5% 5% 6% 7% 8% 9% 10%
© 2018 Citigroup
Figure 95. Europe Base Case EV Forecasts

Europe 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
16,503 16,804 17,489 18,145 18,890 19,248 20,033 20,390 20,797 21,149 21,486 21,503 21,477 21,405 21,287 21,172 21,059
% change 2% 4% 4% 4% 2% 4% 2% 2% 2% 2% 0% 0% 0% -1% -1% -1%
ICE 16,211 16,402 16,984 17,422 17,861 17,526 17,617 17,167 16,465 15,823 15,039 14,258 13,421 12,558 11,674 10,801 9,937
Mild-hybrid 14 12 6 20 20 21 19 18 15 13 11 9 7 6 5 4 3
48-volt 0 0 0 0 94 577 1,002 1,631 2,496 3,172 3,867 4,193 4,510 4,816 5,109 5,399 5,686
Hybrid 184 206 290 435 545 651 775 788 802 815 827 785 744 704 665 627 592
PHEV 37 97 118 136 195 237 297 364 454 568 727 943 1,156 1,367 1,572 1,775 1,976
BEV 58 88 91 132 174 236 323 423 565 757 1,014 1,314 1,638 1,955 2,263 2,566 2,864
Penetration %
ICE 98% 98% 97% 96% 95% 91% 88% 84% 79% 75% 70% 66% 62% 59% 55% 51% 47%
Mild-hybrid 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%
48-volt 0% 0% 0% 0% 1% 3% 5% 8% 12% 15% 18% 20% 21% 23% 24% 26% 27%
Hybrid 1% 1% 2% 2% 3% 3% 4% 4% 4% 4% 4% 4% 3% 3% 3% 3% 3%
PHEV 0% 1% 1% 1% 1% 1% 1% 2% 2% 3% 3% 4% 5% 6% 7% 8% 9%
BEV 0% 1% 1% 1% 1% 1% 2% 2% 3% 4% 5% 6% 8% 9% 11% 12% 14%
Figure 96. China Base Case EV Forecasts

China 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
19,652 21,248 24,788 25,387 26,936 28,417 30,833 32,220 33,670 35,185 36,769 37,136 37,508 37,883 38,261 38,644 39,031
% change 8% 17% 2% 6% 5% 9% 4% 4% 4% 4% 1% 1% 1% 1% 1% 1%
ICE 19,573 21,026 24,333 24,622 25,709 26,602 27,459 27,632 27,428 26,763 26,240 25,319 24,299 23,179 22,032 20,858 19,656
Mild-hybrid 2 0 0 5 5 5 5 5 4 4 3 3 2 2 1 1 1
48-volt 0 0 0 0 54 142 925 1,611 2,694 4,222 5,515 6,127 6,751 7,387 8,035 8,695 9,367
Hybrid 27 37 115 182 328 488 683 714 746 780 815 782 750 720 691 663 636
PHEV 17 61 94 114 194 337 555 741 942 1,196 1,507 1,819 2,212 2,613 3,022 3,439 3,863
BEV 33 124 245 465 646 843 1,205 1,517 1,855 2,220 2,688 3,086 3,492 3,981 4,480 4,989 5,507
Penetration %
ICE 100% 99% 98% 97% 95% 94% 89% 86% 81% 76% 71% 68% 65% 61% 58% 54% 50%
Mild-hybrid 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%
48-volt 0% 0% 0% 0% 0% 1% 3% 5% 8% 12% 15% 17% 18% 20% 21% 23% 24%
Hybrid 0% 0% 0% 1% 1% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2%
PHEV 0% 0% 0% 0% 1% 1% 2% 2% 3% 3% 4% 5% 6% 7% 8% 9% 10%
BEV 0% 1% 1% 2% 2% 3% 4% 5% 6% 6% 7% 8% 9% 11% 12% 13% 14%
Figure 97. U.S. Base Case EV Forecasts

U.S. 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
16,489 17,445 17,539 17,184 17,400 16,000 16,000 16,105 16,206 16,308 16,380 16,380 16,380 16,380 16,380 16,380 16,380
%change 6% 1% -2% 1% -8% 0% 1% 1% 1% 0% 0% 0% 0% 0% 0% 0%
ICE 15,915 16,944 17,039 16,627 16,437 14,906 14,346 14,022 13,510 11,965 10,544 9,345 8,145 7,435 6,854 6,764 6,607
Mild-hybrid 35 24 8 3 3 3 3 2 2 2 1 1 1 1 1 0 0
48-volt 0 0 0 0 35 80 480 805 1,296 2,446 3,276 4,095 4,914 5,160 5,406 5,406 5,406
Hybrid 417 360 340 365 457 500 580 584 587 591 594 564 536 509 484 459 436
PHEV 59 51 77 94 121 143 175 209 242 489 983 1,310 1,474 1,638 1,671 1,704 1,638
BEV 63 65 74 95 348 368 416 483 567 815 983 1,065 1,310 1,638 1,966 2,048 2,293
Penetration %
ICE 97% 97% 97% 97% 94% 93% 90% 87% 83% 73% 64% 57% 50% 45% 42% 41% 40%
Mild-hybrid 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%
48-volt 0% 0% 0% 0% 0% 1% 3% 5% 8% 15% 20% 25% 30% 32% 33% 33% 33%
Hybrid 3% 2% 2% 2% 3% 3% 4% 4% 4% 4% 4% 3% 3% 3% 3% 3% 3%
PHEV 0% 0% 0% 1% 1% 1% 1% 1% 1% 3% 6% 8% 9% 10% 10% 10% 10%
BEV 0% 0% 0% 1% 2% 2% 3% 3% 4% 5% 6% 7% 8% 10% 12% 13% 14%
© 2018 Citigroup
The Outlook for Supply (from OEMs)

EV product range is set to expand significantly in the next 3 years. When it
comes to deciding on what EV to buy today, customers are hardly spoilt for choice.
Manufacturers offer a limited number relative to the number of conventional models
available. However, the number of new BEV and PHEV models being launched
globally doubled last year and is expected to almost double again in 2018 (see
Figure 98). We only show those vehicles that LMC deem to be “highly likely” (i.e.,
where there is strong evidence of commercialization) to reach series production.
Figure 98. Number of BEV and PHEV Models to be Launched Globally* Figure 99. BEV and PHEV Model Launches Split by Region*
140 160
120 140
64
42 120
100
Model Launches
Model Launches
45
100
80
80
73
60 20
59 60
55
40 44 40
16 20
20 7
14 17
0
0
2015 2016 2017 2018 2019 2020
2015 2016 2017 2018 2019 2020
Asia-Pacific Europe Big 5 North America
BEV PHEV
* “highly likely” to reach series production per LMC * “highly likely to reach series production per LMC
Source: Citi Research, LMC Source: Citi Research, LMC
The growth in the number of models is predominantly being seen in China. In

2017, the number of BEV and PHEV models launched grew by almost 3x and is
expected to increase by another 70% in 2018. While the number of new BEV and
PHEV models being launched in Europe’s top five markets is also expected to grow
significantly in 2018 and 2019, in North America the number of new models is
expected to be lower in each of the next three years than the 14 models that were
launched in 2017 (Figure 99).
Expect more electric SUVs as carmakers expand their EV product range.

Manufacturers will be able to provide the consumer with greater choice as their
electric product ranges grow and develop to meet consumer preferences with
regard to vehicle size. Currently, the BEV market is dominated by smaller A-
segment and B-segment vehicles (combined 55% of global BEV market share) and
only 5% of BEV units sold in 2017 were SUVs (Figure 100). By comparison, the
segment mix of ICE vehicles is dominated by larger vehicles; A-segment and B-
segment vehicles represented only 17% of sales and SUV market share was over
three times the size of BEV SUV market share at 16%.
Car segmentation is based on LMC classification, which does not formally define
characteristics of each segment. A-segment includes “basic” cars such as the VW
Up!, B includes “sub-compact” cars such as the Fiat 500X, and C includes
“compact” cars such as the Nissan Leaf. Larger segments include the BMW 3
Series (D-midsize), Tesla Model S (E-Large), BMW 7 Series (F-Large-Plus) and
Porsche Mission E (G-Sport), as well as SUVs such as BMW X5 and Audi Q e-tron.
© 2018 Citigroup
Figure 100. 2017 BEV Sales by Segment Figure 101. 2021 BEV Sales by Segment
SUV SUV
MPV 5% 11% A-segment
1% MPV 12%
A-segment 2%
21%
B-segment
C-segment &
26%
above
39%
C-segment &
B-segment
above
34%
49%
Note: Data excludes the ‘Unclassified’ segment Note: Data excludes the ‘Unclassified’ segment
EV line-ups will soon better reflect conventional car model mix. The swathe of
BEV models to be launched in the coming years should bring the BEV segment mix
closer in line with ICE mix. While segment market share of ICE vehicles is not
expected to change materially from 2017 to 2021, the opposite is true in the BEV
market (Figure 101).
A-segment and B-segment combined share for BEVs is expected to fall from 55% to
38% in 2021 as manufacturers introduce larger models to the market that were
previously unavailable with a fully electric powertrain. As a result, LMC forecasts
market share of larger vehicles classified as C-segment & above to grow from 39%
to 49% and SUV share to more than double from 5% to 11%, bringing it closer to
the 17% 2021 market share for ICE vehicles.
The implication here is that consumers will soon have the choice between
conventional and fully-electric powertrain without being restricted by choice of
model.
Figure 102. What Size BEVs Are Available Today from Top 10 Global BEV Manufacturers?
BEV 2017 Sales (Units 000's) A B C & Above SUV Unclassified Total
BAIC 0 88 13 0 0 101
Geely 42 15 23 0 0 81
Tesla 0 0 46 34 0 80
Renault-Nissan 31 0 49 0 0 80
BYD 0 0 47 0 0 47
Chery Group 10 19 2 0 0 30
Jianghuai 0 0 30 0 0 30
Changan 8 17 3 0 0 29
Hyundai 0 8 16 0 0 25
GM 0 24 0 0 0 24
Average 9 17 23 3 0 53
Source: Citi Research, LMC
© 2018 Citigroup
Figure 103. What Size BEVs Will We Be Buying in 2021?

BEV 2021 Sales (Units 000's) A B C & Above SUV Unclassified Total
VW 10 1 180 86 160 438
BAIC 14 171 45 0 16 246
Tesla 0 0 211 33 0 244
Renault-Nissan 39 7 129 2 64 241
SAIC 43 4 25 0 141 212
Geely 39 34 80 30 19 201
Hyundai 0 32 56 0 65 154
BYD 0 13 109 0 13 135
Toyota 0 1 85 0 42 128
Jianghuai 25 12 65 1 12 116
Average 17 28 99 15 53 212
In Figure 102 and Figure 103 we demonstrate how this trend develops across the
top 10 global BEV manufacturers. Interestingly, Tesla appears to be ahead of the
trend, having sold 34,000 BEV SUVs (Model X) in 2017, whereas the other nine top
OEMs did not offer a fully-electric SUV model last year. By 2021 however, VW is
expected to sell 86k SUV BEV units, almost 3 times as many as Tesla will sell of its
Model X.
Elsewhere the C-segment & above category on average should grow by 330%
while average combined A-segment and B-segment grows by only 35%. The
increase in unclassified units being sold is a result of where the segment details of
future models is unclear.
Outlook by Carmaker
Europe
BMW: 25 electrified models by 2025, including 12 fully electric. BMW is explicit
in its ambition to be a world leader in e-mobility and will launch a fully electric Mini
and X3 in 2019, 3 Series BEV in 2020 (it already offers PHEV powertrain), and the
iNext and i5 in 2021. It will offer a range of 12 BEV models by 2025, which is two
more than Mercedes.
Daimler: Mercedes to offer electrified version of all models and Smart range
to be fully electric by 2022. Mercedes will launch its EQ range, which is expected
to include 3 SUV models, 3 sedans, and a hatchback by 2021, and Daimler plan to
introduce more than ten new all-electric vehicles by 2022.
VW Group: All 300 VW Group models will have electrified versions by 2030.
Audi plans to launch its fully electric E-Tron SUV in 2018 and is expected to follow
this with the E-Tron coupe the following year. It will also offer BEV models from its A
and Q ranges and a compact EV in 2020. The brand plans to offer more than 20
different electrified models by 2025.
The VW brand will launch its fully electric ID range on its new MEB platform with
the VW I.D. hatchback (2019) and the ID Crozz SUV (2020). In China, FAW-VW is
also expected to launch a fully electric New Bora in 2018.
Skoda’s first BEV will be a fully electric Citigo expected in 2019, and this will be
followed in 2020 by an SUV based on the Vision E concept that was on display at
the 2017 Frankfurt Motor Show.
Lastly, production of Porsche’s first fully electric car, the Mission-E, is expected to
begin in 2019 after first being displayed as a concept in 2015.
© 2018 Citigroup
Groupe Renault: 8 new electric and 12 new electrified vehicles by 2022. The
Group already produces the Renault Zoe and Nissan Leaf, which have a combined
BEV market share of 11%, as well as the Mitsubishi Outlander, which is the third-
best-selling PHEV globally. In 2020, Mitsubishi and Nissan are both expected to
start production of an SUV BEV, and Renault is expected to launch a fully-electric
Kwid.
Groupe PSA: All car and light-trucks to offer electrified versions by 2025. PSA
already offer fully-electric passenger cars in the Citröen C-Zero and Opel Ampera-e,
and the Group’s other brands both plan to launch their first BEVs in 2019, with a
fully-electric Peugeot 208 and the DS 3 Crossback.
FCA: 10 BEV and 25 PHEV nameplates by 2022. To date, the company lags its
peers on in the number of electrified vehicles it offers (only 3,000 Pacifica Hybrid
van and 5,000 Fiat 500 BEVs were sold last year), and CEO Marchionne has been
critical of electric vehicles in the past. Nonetheless, the company has committed to
spend €9 billion on EVs over the next five years and is targeting fleet electrification
of 20% in EMEA and NAFTA and 15% in APAC by 2022.
Volvo Cars: All new models to be electrified from 2019. Volvo will launch five
fully electric BEV models between 2019 and 2021, of which three will be Volvo
models and two will come from its high-performance brand Polestar. Other
electrified options will include both plug-in hybrids and 48-volt mild hybrids, but pure
ICE vehicles will be gradually phased out. The company plans to have sold 1 million
electrified cars by 2025.
China
Beijing Automotive Group (BAIC): To go fully electric by 2025. The state-
backed company plans to completely stop all production and sales of traditional ICE
vehicles from its own brand range in China by 2025. It is the second Chinese
manufacturer to make this pledge (Changan Automobile has set the same target).
The BAIC EC-Series was the best-selling BEV globally in 2017, with 78k units sold
compared with 45k Tesla Model S and 45k Nissan Leaf units.
Geely Group: 30 new energy vehicles (NEV) to be launched by 2020. The

company has pledged to release several NEVs capable of driving over 500km on a
single charge by 2020, which would rival the range that can be achieved on current
Tesla models. In 2018 the company also launched its flagship B-segment hybrid
sedan with MHEV and PHEV systems. The company claims the systems used in
12
this model are the world’s most efficient hybrid and plugin hybrid systems
Japan
Honda: By 2025, two-thirds of European sales will be electrified powertrains.
The Honda DF HR-V will be the company’s first fully-electric BEV in 2018, and in
2019 it is expected the Urban BEV will go into production.
Toyota: 10 news BEVs by 2020 and electrified options across entire lineup by
2025. Toyota does not currently offer a fully electric vehicle, but announced in
December 2017 that it will offer 10 BEVs by 2020. LMC forecasts sales for a
Compact BEV from 2018 and an SUV from 2019.
12
Geely Announces New Energy Strategy with Launch of Smart Hybrid Flagship, Geely
Global Media Center, May-18, link
© 2018 Citigroup
South Korea
Hyundai: Expect electric vehicles to represent 10% of sales by 2025. Hyundai

will launch a fully electric Kona SUV in 2018 and is also expected to offer a BEV
Elantra/Avante in 2020. The Kia Niro and Kia Stonic will add additional BEV SUVs
to the range in 2018 and 2019.
© 2018 Citigroup
U.S.
Tesla: Plans for the Model Y and Roadster to follow Model 3. Production of the
highly anticipated Model 3 sedan is expected to ramp up in 2018 after further delays
in 2017 and will be followed by what Tesla is currently referring to as the ‘Model Y’
crossover in 2019. The Model Y is yet to be officially unveiled by Tesla, but is
understood to be in final development stages. In 2020 Tesla hopes to also launch
the second generation Roadster.
Ford: 40 electrified vehicles will arrive by 2022. In 2017 Ford announced seven
of the thirteen electrified vehicles it initially planned to launch in the next five years,
which includes two passenger cars, a fully electric small SUV, and a hybrid Mustang
both expected in 2020. The range also includes a hybrid “high-volume autonomous
vehicle designed for commercial ride hailing,” a hybrid F-150 pickup, a hybrid
13
Transit Custom plug-in, and two “pursuit-rated Hybridge police vehicles” . At the
Detroit auto show in January of this year, however, it significant increased its
electrification ambitions and announced plans to invest up to $11 billion by 2022
(the previous figure was $4.5 billion). The ambition is to offer 40 electrified models,
of which 16 will be BEV.
GM: 20 all-electric vehicles launched by 2023, including two by mid-2019. In

2017, GM announced its ambitions for the EV market with plans to offer 20 fully
electric vehicles by 2023. LMC expect Buick to produce a midsize SUV, compact
MPV, and Encore with fully electric powertrain in 2019 as well as a Cadillac XT4
and a new Chevrolet BEV in 2020.
13
Ford Adding Electrified F-150, Mustang, Transit By 2020 In Major EV Push, Ford
Media Center, Jan-17, link
© 2018 Citigroup
Conclusion
Figure 104. There’s a Long Way to go Before Reaching Full EV Figure 105. Penetration of Electric Vehicles Within the Global Car Fleet
Adoption
100
100
90
Laggards
80
80
BEV Penetration (%)
70
60
60
Late majority 50
40
40
30
Early majority
20
20
Early adopters 10
Innovators
0
0
Time 20182024203020362042204820542060206620722078208420902096
2017 2020e 2025e
Parc ICE New Car ICE Parc BEV New Car BEV
2030e 2030e (Bull case) 2030e (Bear case)
Source: Citi Research estimates Source: Citi Research estimates
We have no doubt that the Car of the Future will not be powered in the same way
that it has been for the past century. For us it is a question of timeframe; ‘how do
you define the future’? Are we talking about five years, 10 years or 50 years? In our
forecasts, our base case assumption is that in 2030 BEVs will account for 10% of
global new vehicle sales. Our bull case scenario foresees global BEV penetration of
18% in 2030. Both of these outcomes would still only put us within what we define
as the ‘early-adopters’ phase (see Figure 104).
The barriers will (most probably) be broken down, they always are — it is a question
of time. As we see it, the challenge that EVs face is they do not change the utility of
a vehicle, and therefore the adoption curve will take longer to navigate (compared
with truly disruptive tech). Given the lack of improved utility for consumers, having to
pay a price premium is more-difficult to accept. Given consumer reticence to pay for
fuel efficiency, the hockey stick effect (for demand) will occur when the total cost of
ownership is obviously below that of a traditional combustion engine vehicle. As we
have said in this report we think this is feasible, but we are not there yet.
© 2018 Citigroup
Appendix 1
Vehicle Types
Electric vehicles can broadly be split into four categories, as shown in Figure 106
below. Hybrid Electric Vehicles (HEVs), Plug In Hybrid Electric Vehicles (PHEV),
and Electric Range Extended Vehicles (E-REV) all include internal combustion
engines that are supplemented (usually at low speeds, in urban areas) by electric
batteries. Battery Electric Vehicles (BEVs), however, are entirely powered by
electricity from the grid. The distinction between HEVs and PHEVs is that the
batteries for the HEVs are charged from energy recuperation, while PHEVs, like
EVs, are charged from the mains.
Figure 106. Structure of HEVs PHEVs, E-REVs and EVs
 Conventional hybrids: A conventional hybrid combines an electric motor while

still having a gasoline engine. Their battery cannot be charged using a plug, but
only recuperates electricity from brake energy, converting kinetic to electric
energy, which normally goes to waste with traditional gasoline engines. This
efficiency gain allows cars to run electrically and thereby reduces gasoline use.
 Plug-in hybrid electric vehicles (PHEVs): The basic building blocks of this car
are the same as in a conventional hybrid, as PHEVs also have an electric and a
combustion engine. The difference (as one can guess from the name) is that the
battery of these vehicles can be charged using a plug which increases the range
of the car in electric-only usage using “grid-electricity” instead of gasoline. Even
their short reach allows users to commute in electric-only mode, as many people
commute less than 50km every day. Apart from a positive environmental impact,
electric usage also has a variable cost benefit as running the car is cheaper.
 Battery electric vehicles (BEVs): These vehicles represent the other spectrum
of cars and forego the internal combustion engine. Due to missing the
combustion engine, these vehicles have larger batteries, which allow them to
drive longer ranges compared with PHEVs in electricity-only mode. Even though
these cars don’t create emissions while driving, one may not forget that there are
significant emissions when the electricity for charging these cars is produced.
© 2018 Citigroup
Appendix 2
Top Selling EVs by Region and OEM
Figure 107. Specification Comparison of the Best-Selling BEV and PHEV in Europe & N. America During 2017
Model Type Global Battery Battery EV Rapid Standard Average Average 2017 Sales % of total
Segment capacity Range charging to charging Price (€) Price ($) Volume BEV &
(kWh) (km) 70/80% (min) (hour) (units) PHEV
volume
Tesla Model S 75 BEV E Li-ion 75.0 489 40 11.0 105,659 124,678 39,683 8.9%
Renault ZOE ZE 40 BEV A Li-ion 40.0 402 60 6.0 24,585 29,101 30,381 6.8%
Nissan Leaf BEV C Li-ion 30.0 249 30 4.5 32,125 37,908 29,728 6.7%
Tesla Model X 75D BEV SUV Li-ion 75.0 565 40 11.0 115,465 136,249 26,092 5.9%
Chevrolet Bolt BEV B Li-ion 60.0 238 60 5.9 50,148 59,175 23,343 5.3%
BMW i3 BEV C Li-ion 33.0 314 35 4.5 65,816 77,663 16,308 3.7%
VW e-Golf BEV C Li-ion 35.8 299 35 5.0 42,404 50,037 16,137 3.6%
Weighted Average 52.6 381 43 7.3 65,149 76,867
BEV
Toyota Prius PHEV C Li-ion 8.8 63 N/A 2.8 29,778 35,138 21,621 4.9%
Chevrolet Volt PHEV C Li-ion 16.0 53 N/A 4.4 46,134 54,438 20,125 4.5%
VW Golf PHEV C Li-ion 8.7 50 N/A 2.0 31,717 37,426 16,137 3.6%
Mitsubishi Outlander PHEV SUV Li-ion 12.0 53 25 3.5 39,131 46,175 14,390 3.2%
BMW 330e PHEV D Li-ion 7.7 40 N/A 2.0 40,445 47,725 14,022 3.2%
Volkswagen Passat PHEV D Li-ion 9.9 35 N/A 3.0 39,240 46,303 13,168 3.0%
Audi A3 e-tron PHEV C Li-ion 9.0 50 N/A 3.0 37,277 43,987 10,891 2.5%
BMW 2-series AT 225xe PHEV C Li-ion 7.6 40 N/A 2.0 37,456 44,198 10,835 2.4%
Mercedes GLC 350e PHEV E Li-ion 6.2 31 N/A 1.8 43,052 50,801 10,356 2.3%
BMW 530e PHEV E Li-ion 9.2 47 N/A 3.0 50,111 59,131 10,033 2.3%
BMW X5 xDrive40e PHEV SUV Li-ion 9.0 31 N/A 3.0 62,963 74,296 9,791 2.2%
Volvo XC90 PHEV SUV Li-ion 9.2 40 N/A 3.0 72,587 85,653 9,295 2.1%
Ford C-Max PHEV C Li-ion 7.6 20 N/A 3.0 36,276 42,806 8,612 1.9%
Weighted Average PHEV 9.7 45.1 N/A 2.9 41,914 49,459
Source: Citi Research, LMC, Company Data, Zap Map, Smart EV, Pod Point, EV Box, Elektrek, Edmunds, AutoHaus, Broadspeed, VoitureNeuve
Figure 108. Specification Comparison of the Best-Selling BEV and PHEV in China During 2017
Model Type Type 2 Battery Battery EV Range Fast Standard Min price Max price 2017 sales % of total
capacity (km) (MIIT) charging to charging (USD) (USD) volume NEV PV
(kWh) 70/80% (hour) (units) volume
(min)
BJEV EC180 BEV Sedan NCM 20.3 156 N/A 7 22,458 23,346 78,079 14.0%
BJEV EC200 BEV Sedan NCM 20.5 162 36 8 23,494 24,382
Zhidou D2 BEV Sedan NCM 18 155 N/A 6-8 22,458 27,932 42,342 7.6%
Chery eQ 2017 BEV Sedan NCM 23.6 200 30 8-10 25,136 25,432 25,784 4.6%
JAC IEV6S BEV SUV NCM 33 251 90 12 32,519 32,519 24,210 4.4%
BYD e5 EV300 BEV Sedan LFP 43 305 80 7 28,983 31,942 23,601 4.2%
Geely Emgrand EV300 BEV Sedan NCM 41 300 45 7 28,968 31,927 23,324 4.2%
Zotye E200 BEV Sedan NCM 24.52 160 30 10-12 26,897 26,897 16,751 3.0%
Chang'an Benben EV180 BEV Sedan NCM 23.2 180 30 8 22,902 24,677 14,549 2.6%
Chang'an Benben EV210 BEV Sedan NCM 27.5 210 45 9 23,790 25,417
BAIC BJEV EU260 BEV Sedan NCM 41.4 260 30 6-7 30,462 31,942 13,158 2.4%
BAIC BJEV EU400 BEV Sedan NCM 54.4 360 30 6-7 33,273 33,273
Zotye Cloud 100 BEV Sedan NCM 18 155 90 8 23,509 25,136 11,069 2.0%
JMC E100 EV BEV Sedan LFP 15 152 N/A 7-9 23,080 23,080 10,663 1.9%
Weighted Average BEV 26 202 50 8 25,591 27,355
BYD Song DM PHEV SUV NCM 16.9 80 45 10 31,942 36,380 30,911 5.6%
BYD Qin 2017 PHEV Sedan NCM 13 100 N/A 6 26,024 29,574 20,738 3.7%
SAIC eRX5 PHEV SUV NCM 12 60 N/A 3 39,339 42,298 19,510 3.5%
BYD Tang 100 PHEV SUV NCM 22.8 100 N/A 7 39,339 44,369 14,592 2.6%
Weighted Average PHEV 16 84 N/A 7 34,161 38,155
Source: Citi Research, CPCA, AutoHome
© 2018 Citigroup
Appendix 3
Current Penetration by Region and OEM
Europe
EV sales growing rapidly, but market penetration remains modest. Electric
vehicle penetration is increasing; the number of plug-in EVs (BEV & PHEV) sold in
Europe in 2017 grew by 38% to 279k, and including HEVs this number rises to
735k. This represents a twelve-fold increase in electrified vehicles compared with
2013. Despite the heady growth rates, as a share of the total European car market,
EV volumes remain comparatively small; in 2017 HEV market share was 2% and
BEV and PHEV sales each represented just 1% of the market.
Figure 109. European Penetration is Growing but Remains Modest Figure 110. Penetration in Norway is Miles Ahead of the Rest
800 Units ('000s) 60%
700 50%
600
40%
500 2.4%
30%
400
1.7% 20%
300
1.2%
10%
200 0.8%
1.1%
0.6% 0.7%
100 0%
0.2%
Austria
France
Norway
Sweden
Denmark
Poland
Netherlands
Belgium
Switzerland
Germany
UK
Ireland
Spain
Finland
Italy
0.7%
Portugal
0.3% 0.5% 0.5%
0
2014 2015 2016 2017
BEV PHEV Hybrid BEV PHEV HEV
Source: ACEA, Citi Research Source: ACEA, Citi Research
Norway is the clear EV leader in Europe with 21% BEV penetration. The
Norwegian market has benefited from generous government subsidies (see
Appendix 4 for further details) that have supported high EV penetration. Including
hybrids, half of new cars sold in Norway last year were electrified, and a rate of 21%
pure-electric penetration is around 10 times that in the Netherlands, which saw the
second-highest level of BEV penetration (2%). Austria (2%), Switzerland (2%) and
France (1%) had the next highest BEV penetration.
In terms of type of EV, the development of the three main variants has been:
 BEV: Norway’s generous incentives support its position as market leader.

In 2017, 135,000 new BEV vehicles were sold in Europe at a growth rate of 51%
year over year, but BEVs still represent only 1% of total new car sales. In
Norway, however, BEV market share was 21% largely due to some of the most
generous incentives on the continent, which have helped BEV stock grow almost
20-fold from 5,000 in 2011 to 99,000 in 2016 according to the IEA. Although the
German and French BEV markets are the next-two-largest in absolute terms
(both saw sales of 25,000 units last year), BEV market share remains at just 1%
in both countries.
© 2018 Citigroup
 PHEV: U.K. & Norway lead the PHEV charge. In 2017, 144,000 new PHEV
vehicles were sold in Europe at a growth rate of 29% year over year. PHEV, like
BEV, also currently has just a 1% market share of total new car sales. Norway
again has the highest PHEV penetration, but penetration is also high in Sweden
(4%) and Finland (2%). In absolute terms, however, the U.K.’s PHEV market is
the largest; 31,000 units were sold in 2017, and the country has seen a boom in
PHEV stock from 1,000 in 2012 to 55,000 in 2016.
 U.K. grants catalyze PHEV expansion. EV supportive incentives have helped

reduce total cost of ownership of EVs and facilitated the growth in PHEVs. U.K.
plug-in grant, for example, which was launched in 2011, experienced just 109
14
claims in Q4 2011, but, by Q4 2015, this had risen to 8,453 claims
demonstrating the growth in popularity of such schemes. Other tax incentives
and non-financial incentives such as access to bus lanes and free parking have
also encouraged consumers to buy electric.
 Improving charging infrastructure key for sales growth. Improvements in

charging infrastructure have also reduced one of the key barriers to entry for
consumers. According to the European Alternative Fuel Observatory (EAFO), in
2011 there were 6,987 publicly accessible normal power and 31 high power
charging positions in Europe. This has grown to 104,656 normal and 13,475 high
power positions in 2017.
 The range of EVs available to consumers is broadening. Lastly, the range of

PHEV vehicles available has increased in recent years as manufacturers have
begun to focus on electrified powertrain and increased supply. In 2014, for
example, 99% of PHEV sales in Norway were from just 5 models, whereas in the
last 12 months the top 5 models represented only 62% of the new PHEV market.
 HEV: The electrified vehicle of choice for now. In 2017 456k new HEV
vehicles were sold in Europe; over 175k more than BEVs and PHEVs combined,
demonstrating the current consumer preference for plug-free hybrid technology
over EVs that require charging infrastructure. HEV sales grew 59% YoY, and
HEV market share is currently higher than both BEV and PHEV at 3%.
 Limited charging infrastructure in Italy leaves consumers with little choice.

Interestingly, Italy, which is only the 9th- and 11th- largest European country in
terms of PHEV and BEV sales respectively, had the third-highest number of HEV
sales in 2017. According to the EAFO, between 2012 and 2016 Italy increased
the number of normal power charging positions by just 33% whereas this same
figure increased by 91% in Norway and 242% in the U.K. over the same period.
Consumer preference in Italy for HEVs reflects this lack of investment in charging
infrastructure required for BEV and PHEV models.
14
Complete guide to the plug-in electric car grant, Carbuyer, 2nd March 16, link.
© 2018 Citigroup
Figure 111. European 2017 Plug-in EV Vehicle Market Share Figure 112. European BEV and PHEV 2017 Sales
70,000 Units
Hyundai PSA Other
Geely4% 2% 1% Renault-
Nissan 60,000
6%
25%
50,000
Tesla
10% 40,000
30,000
20,000
Daimler
10,000
12%
0
VW
Hyundai
Other
Renault-Nissan
BMW
Tesla
Daimler
Geely
PSA
VW
21%
BMW
19%
BEV PHEV
Renault is the European electrified vehicle market leader for now. Renault’s
25% share of the European plug-in EV (BEV & PHEV) vehicle market (see Figure
111) suggests it has benefited from acting early in the EV race; the top and third-
best selling BEVs in Europe are the Renault Zoe and Nissan Leaf, respectively. The
first generation Leaf was introduced to European markets in 2011, and first
deliveries of the Zoe were in late 2012, which appears to have given Renault a first-
mover advantage over competitors. VW’s 21% market share is supported by the
popularity of the e-Golf and Golf PHEV, which combined makes the Golf the
second-most popular model in Europe.
Renault also leads the race for BEV sales in Europe. Figure 112 illustrates the
lead Renault has over competitors in Europe, specifically in relation to fully electric
vehicles. The popularity of the Renault Zoe and Nissan Leaf in Europe has given
Renault a 38% share of the BEV market, greater than the Tesla (20%) and VW
(12%) BEV market shares combined. The Germans’ market share is largely
supported by sales of their PHEV models. At VW, 72% of sales were from PHEV
models and similarly, at BMW and Daimler, the rate was 71%.
If we agree that OEMs are in the process of transitioning to fleets of zero tailpipe
emission vehicles, then Renault is clearly leading this race. However, we expect
market share to remain relatively fluid as demand for electric vehicles increases and
new models hit the market in the coming years.
© 2018 Citigroup
China
China alone in 2017 sold 577k NEVs (defined as BEV and PHEV units for the
purpose of this analysis), representing a 52% global market share, significantly up
from 8% in 2012. Despite this, the share of NEV to the total vehicle market remains
low, at 1.3% globally and only 0.7% in China.
Figure 113. China’s Share of Global NEV Passenger Vehicle Market Has Been Increasing Over
the Past Five Years
700 NEV units

('000s)
600 52%
500
400
45%
300
38%
200
100 23%
8% 8%
0
2012 2013 2014 2015 2016 2017
Source: Citi Research, IEA
China accounts for over half of the global NEV market. The recent boom has been
driven largely by generous government NEV subsidies for both manufacturers and
customers. NEVs in China are produced overwhelmingly by local OEMs based on
Chinese technologies. The Chinese government promotes NEV in an attempt to
fight air pollution, improve energy safety (due to China’s high dependence for oil
import at 65%, according to Sinopec Economics & Development Research
Institute), and develop the country into a world leader for NEV and its supply chain
(such as battery) in terms of both shipment volumes and technology.
© 2018 Citigroup
Figure 114. Monthly NEV Passenger Vehicle Sales Volume in China

('000 units)
100
80
60
40
20
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
2012 2013 2014 2015 2016 2017
Source: CPCA, Citi Research
BEV should dominate the NEV industry in the coming years, but we expect the
growth rates for PHEV to pick up dramatically from 2018 and surpass BEV’s not
only due to the low base effect, but also given PHEV’s lower entry barrier. OEMs
without proper BEV experience or joint venture brands that also struggle to meet
CAFC targets under the vehicle upsizing trend can leverage on PHEV to fulfill both
CAFC and NEV requirements (see China regulation section of report).
PHEV has been more appealing to users given its better flexibility, especially
considering the low penetration for charging stations. The latest and a popular
PHEV model, SAIC’s eRX5, has a total range (electricity + gasoline) of around
500km, which is comparable to that of a gasoline vehicle, and has average fuel
consumption (without electricity) at a reasonable 7-8L/100km, much lower than
some older models, such as BYD Tang 100’s nearly 13L/100km. Apart from this, the
gradual phasing out of subsidies will more significantly decrease the attractiveness
of BEV for customers and OEMs given BEV’s higher reliance on subsidy vs.
PHEVs.
Big NEV makers getting bigger. The NEV passenger vehicle market is
concentrated despite the entry of new players. In the first nine months of 2017, the
top-5 makers took up a 65% market share and the top-8 took 82%. Among them,
BYD remains the leader with a 23% market share (although this has over time), with
four of its NEV models ranked as the top-15 best-selling ones in the first nine
months of 2017.
© 2018 Citigroup
Figure 115. Market Shares of Top-5 and Top-8 NEV OEMs Figure 116. Share of the NEV Market in Through September 2017
100% 97% 98% 98% 94%

92% 91% 89% Others,
90% 86% 18%
82%
78% BYD, 23%
80%
69%
70% 65%
60%
50% Zotye, 6%
40%
30% JMC, 6%
20%
10% BAIC BJEV,
JAC, 6% 17%
0%
2012 2013 2014 2015 2016 2017 Chery, 9%
Top 5 Top 8 SAIC, 10%
Source: CAPA Source: CAPA
Geographically, Tier-1 cities accounted for 60% of NEV sales, owing to better local
infrastructure (the charging stations, more specifically), and also because licensed
car plates for conventional vehicles can only be acquired through public lottery or
purchased at a high price in many Tier-1 cities such as Beijing and Shanghai.
Figure 117. 9M17 NEV Sales Volume Breakdown by City Tier Figure 118. NEV Sales Concentrated in Tier 1 and Tier 2 Cities
Tier 5 & 90% 84%
smaller, 5%
80% 72% 72% 74%
Tier 4, 10%
70%
60%
60%
Tier 3, 11% 50%
40%
30%
Tier 1, 60% 20%
Tier 2, 14% 10%
0%
2013 2014 2015 2016 8M17
Source: CAPA Source: CAPA
We see a similar trend in Figure 119, which shows NEV penetration is concentrated
in China’s more urbanized provinces. Unsurprisingly, NEV sales were the highest in
provinces home to China’s largest cities (e.g., Shanghai, Beijing and Guangzhou).
© 2018 Citigroup
Figure 119. 2017 NEV Sales by Province Split by Powertrain Type
100,000 Units
90,000
80,000
70,000
60,000
50,000
40,000
30,000
20,000
10,000
0
Qinghai
Tibet
Beijing
Anhui
Shanxi
Liaoning
Jilin
Fujian
Sichuan
Guizhou
Heilongjiang
Ningxia
Zhejiang
Tianjin
Henan
Jiangsu
Hunan
Hebei
Jiangxi
Hubei
Gansu
Inner mongolia
Xinjiang
Guangdong
Guangxi
Yunnan
Shaanxi
Chongqing
Hainan
Shanghai
Shandong
BEV PHEV HEV
Source: Citi Research, Thinkercar
Of the top-15 best sellers in the first nine months of 2017, 73%, or 11, of them were
BEV, all 11 of them were sedans, although BYD’s PHEV SUV Song DM managed to
come in at No. 3. Unlike gasoline vehicles that are gradually upsizing, small NEVs
are still favored, with 53%, or 8, out of the 15 top sellers being of class A00 and A0,
given subsidy remains a key reason behind the production and purchase of NEV.
While we are still in an early lithium-ion battery (LFP)-to-lithium nickel manganese
cobalt oxide battery (NCM) cycle, as pointed out by Citi’s Basic Materials team, we
find that NCM is becoming a standard configuration for the most popular NEVs, with
only two of them equipped with LFP ones.
Figure 120. Top-15 Best-selling NEV Models in September 2017 and Their Respective Market
Shares
14%
12.1%
12%
10.4%
10%
8%
6.4%
5.8%
6% 5.2% 5.0% 4.7%
4.3%
4% 3.4% 3.3% 3.3% 3.0%
2.6% 2.5%
1.8%
2%
0%
BAIC EC series
JAC IEV6S
Zhidou D2
SAIC eRX5
BYD Tang
BYD Song DM
Zotye E200
BAIC BJEV EU
BYD e5
Zotye Cloud 1
Chery eQ
Chang'an Benben
Geely Emgrand EV
JMC E100
BYD Qin
Source: CPCA
© 2018 Citigroup
U.S.
The EV theme in the U.S. isn’t new. Skeptics would probably point to the “EV rush”
of 2008–09 — when EV enthusiasm (amid rising gas prices) ultimately ended with
numerous companies failing amid overly optimistic projections. Indeed, the lessons
learned from that era probably contributed to the industry’s more cautious approach
to re-embracing EVs up until the last few years or so. Two things changed in the
past few years. First, Tesla’s success was a game-changer with respect to the
perception of EVs among consumers and industry observers. Second, continued
battery cost reductions got to a point where a path towards ICE parity was visible.
Auto companies are speaking a different language today. Rather than treating EVs
as a cost/endeavor of doing business or meeting regulations, today many
companies understand that EV disruption is a matter of when not if. Nobody wants
to be left behind, and no Automotive CEO wants to continue hearing about their
future demise from the likes of Tesla and other EV newcomers.
So a race has clearly begun for the future of automotive propulsion.
Figure 121. U.S. EV Penetration
12%
10% Baseline Upside
8%
6%
4%
2%
0%
2016 2017 2018 2019 2020 2021 2022 2023 2024
Source: Citi Research, Company Websites
© 2018 Citigroup
Current U.S. Penetration
EV penetration today however remains low at just 1% of total new car sales.
Furthermore, Figure 122 shows just how heavily concentrated BEV sales are in
California relative to other U.S. states (over 8x as many BEVs were sold in
California compared with Florida, the second-largest state by BEV sales). Apart
from California (CA), Florida (FL) and Washington (WA), no U.S. state had BEV
penetration over 0.02% in 2016.
Figure 122. 2016 U.S. BEV Market Share Appears Negligible and a California Phenomenon
0.20%
0.18%
0.16%
BEV Share of US Market
0.14%
0.12%
0.10%
0.08%
0.06%
0.04%
0.02%
0.00%
GA
IL
OR
AZ
PA
VA
OH
CA
FL
TX
NY
CO
NJ
WA
WI
MA
HI
NC
TN
MD
UT
MN
CT
NV
MO
Source: Citi Research, Polk, IHS, Company Reports
GM and Tesla are U.S. market leaders for now. A handful of popular models
provide GM and Tesla with a combined 47% share of the U.S. EV market (Figure
123), and it was the popularity of Chevrolet’s fully-electric Bolt (23,000 units) and
plug-in hybrid Volt (20,000 units) that put GM ahead of Tesla in 2017 U.S. sales.
Although Tesla’s Model S was the highest-selling electrified vehicle last year in the
U.S. (25,000k units), lighter Model X sales (15,000 units) and the well-documented
Model 3 production problems left Tesla with 23% market share, compared with
GM’s 25%.
Figure 123. U.S. EV Market Share by Manufacturer Figure 124. U.S. BEV & PHEV Sales by Manufacturer
50,000 Units
VW Hyundai OtherDaimler
4% 3% 2% 2% 45,000
GM
FCA 40,000
25%
5% 35,000
Renault- 30,000
Nissan 25,000
6%
20,000
15,000
BMW
10,000
9%
5,000
0
GM
VW
Hyundai
Other
Tesla
Ford
BMW
Toyota
Renault-Nissan
Daimler
FCA
Geely
Honda
Tesla
Toyota 22%
11%
Ford
11%
BEV PHEV
© 2018 Citigroup
Tesla dominates the fully-electric market in the U.S. When considering unit
sales by BEV and PHEV split (Figure 124) it is clear that Tesla’s all-electric range
means it dominates the BEV market, whereas GM’s sales are split almost evenly
between fully electric (Bolt) and PHEV (Volt). Elsewhere Ford, Toyota, and BMW
sales are almost entirely PHEV, whereas Renault-Nissan’s sales are mostly BEV
(95% of sales are generated by the fully electric Nissan Leaf).
What Lies ahead for EVs in the U.S.?
Unlike consumer electronics, the automotive supply chain is lengthy in nature and
can’t really accommodate an “overnight” conversion to any new propulsion system.
Even under the most aggressive timeline and assuming no supply-chain
bottlenecks, this is a matter of many years.
So rather than dwell on potential 2030+ EV penetration rates, we prefer to focus on

the most telling data points in 2018–19 that will shape the speed of EV adoption and
therefore EV investment. The next 12 months will prove very revealing for what we
call Gen-3 EVs—the Tesla Model 3, Chevy Bolt, Nissan Leaf, and others expected
to launch in 2018–19. These vehicles will test the market demand for EVs that offer
good yet less than a full-range (200–310 miles vs. ICE at 250–300) at lower yet not
quite mass-market sticker cost ($30–$50,000 as opposed to $20–$35k for small/mid
sedans).
Much of the recent excitement around EVs hasn’t necessarily stemmed from the
specifications of these Gen-3 vehicles, but rather the market’s apparent response to
the Tesla Model 3. The existence of >400k global Model 3 pre-orders sent waves
throughout the auto industry with the interpretation that the EV disruption was
coming far sooner than many expected. Though the pre-orders themselves weren’t
firm orders per se ($1k fully refundable reservations, no credit checks), they at least
gave the market a sense of the awareness and general intent to own an EV.
As Tesla scales up the Model 3, the rate of Model 3 order conversions — after initial
enthusiasts and some Model S/X owners take delivery — could be the most
important data point for the entire global EV story.
The Model 3 pre-orders do beg another question though. Why has the response to
other Gen-3 EVs, namely the Bolt, been far less impressive? To be sure, by all
accounts the Bolt has been well-received, reviewed favorably, and won awards, and
it is gradually gaining sales momentum. But it hasn’t generated nearly as much
buzz as the Model 3.
Now, the first reaction when we pose this question is that the Model 3 is better
styled than the Bolt. Even though styling is a matter of opinion, we’d still argue that
the Bolt should be doing better if Gen-3 EV specs are as compelling as Tesla Model
3 pre-orders suggest they are. After all, one could easily counter that, at the $35–
$40k price point, the Bolt is quite competitive against a similarly priced Model 3 —
more range, instrument cluster, Apple CarPlay, pedestrian automatic emergency
braking (AEB) systems, very strong regeneration based on media reports — though
the Model 3 clearly wins on semi-autonomous sensing suite (though one is coming
on the Bolt), over-the-air (OTA) updates, and acceleration. Away from one’s view
around styling, one could argue that the two products each have some pros/cons,
particularly at the lower price-point.
© 2018 Citigroup
Figure 125. Chevy Bolt vs. Tesla Model 3
Bolt Model 3
Starting Price $37,495 $35,000
Fully Loaded $43,510 $59,500
Base Range 238 220
Top Range 238 310
0-60 MPH (sec) 6.5 5.1-5.6
Center Touch Screen 10.2" 15"
Instrument Cluster? Yes No
Apple CarPlay? Yes No
Rear Camera Mirror? Yes No
Surround Vision? Yes No
Autonomous Features? No Yes
Supercharging No Yes
Vehicle AEB? Yes Yes
Pedestrian AEB? Yes Not yet
Length 164.0" 184.8"
Wheelbase 102.4" 113.2"
Source: Citi Research, Company Websites
That the Model 3 appears to be gaining far more orders than the Bolt begs a key
question: Is the Model 3 success really about “EVs” or is it about Tesla as a
brand/product? And to what extent are Tesla-exclusive features like large displays,
OTA, and the promise of full self-driving also a factor in demand?
If it’s the former, then it’s obviously a very bullish sign for future EV penetration and
one that every automaker will need to aggressively respond to (some arguably
already are). By definition, under this scenario competing vehicles like the Bolt and
Leaf would also stand to do better in 2018-19, as would new entries.
But if it’s the latter, then the broader EV implications wouldn’t be as powerful. That
doesn’t mean that the EV growth story would subside or that this is 2008–09 all over
again, but rather that automakers might reconsider their reaction to Tesla’s success
— perhaps by accelerating OTA and automated driving functionalities, or by re-
thinking other design principles.
Perhaps it’ll be a matter of both. If the initial Gen-3 EV enthusiasm is really about
Model 3, that would still result in a significant increase in public awareness and
word-of-mouth. Such a scenario would probably benefit the other Gen-3 players, so
the answer to the question posed above won’t necessarily play out concurrently with
Tesla’s Model 3 production ramp.
Besides demand for Gen-3 EVs, we think another important trend to monitor is the
price of battery raw-material costs such as lithium, cobalt, and nickel. In December
2017, Hyundai noted (per Reuters) that rising raw material costs threaten to slow
and then maintain a status quo for battery costs by 2020. If this materializes, this
too could slow EV demand depending on where battery prices settle at that time.
Assuming a $100/KwH battery pack cost, a mid-200 mile EV would likely still sell at
a modest price premium to an ICE though the EV would be more compelling on a
cost-of-ownership basis. Auto observers have long-debated the selling power of
cost-of-ownership vs. sticker price parity, though clearly the latter would be better
for EV demand to truly inflect in a disruptive fashion (particularly if range isn’t quite
at ICE levels in all weather conditions).
© 2018 Citigroup
In the U.S., the rush of higher-volume EV launches still appears to be more of a

2020+ event, though the next few years will see a handful of important launches,
such as the Gen-3 EVs, which will be closely monitored throughout 2018.
Toyota recently unveiled its EV plan that called for an EV “rollout in earnest” starting
in 2020 in China (mass-produced vehicle), followed by Japan, India, the U.S. and
Europe. By 2030, Toyota’s plan called on >50% of its vehicle sales to be electrified
(including hybrids) with pure EV (and fuel-cell) representing >10% of Toyota sales.
Ford is expected to roll out EVs more aggressively in 2020+. GM, on the other
hand, is expected to launch two new EVs in the next ~16 months.
For the U.S., we expect EV penetration to remain <10% through the early 2020s
even under somewhat optimistic projections, but the difference is that the industry at
that point could be prepared to scale much faster should supply-demand dictate so.
Under the most optimistic scenario for battery development as well as autonomous
driving (AV) development, the EV penetration story beyond the early/mid-2020s
could become a how fast can they scale it question at least in major passenger
vehicle segments. What’s clear is that automakers are investing capital now to
prepare for scalable xEV platforms in the 2020+ era.
In the meantime, the next 12 months will prove plenty interesting in shaping the EV
narrative. Demand for Gen-3 EVs (Model 3, Bolt EV, Leaf) and launches like the
Jaguar I-Pace will answer the following questions:
1. Whether consumers are purchasing EVs with the total-cost-ownership in

mind or sticker price: Gen-3 EVs are still more expensive than competing
ICEs on a sticker price level, but less so on a total-cost-of-ownership. If cost-of-
ownership starts to play a major role in the buying decision for Gen-3 EVs, then
the pace of disruption in the 2020+ timeframe could accelerate. Of course the
automakers selling Gen-3 EVs could benefit from being early with such
product.
2. How consumers weigh the cost/benefit of today’s EVs: At a modestly

higher cost of ownership, one could argue that consumers would still purchase
Gen-3 EVs because of other well-regarded EV benefits — smooth acceleration,
greater usable capacity, brake regeneration. But they’d still have to accept
modestly less range (200–300 miles, the high-end being the LR Model 3
variant) with the Gen-3 EVs. Will they or won’t they? The EV bull argues that
EVs are so compelling that consumers will merely adjust lifestyles for the
modest range sacrifice. The EV bear argues against that notion by pointing out
that U.S. consumers didn’t enthusiastically accept 100-mile EVs even though
most daily range needs are lower than 100 miles. Past studies done by
consultants suggest a desire for the upper-end of the 200–300 mile range,
though it’s unclear whether such studies are limited by low awareness of EV
benefits.
3. Is this an EV movement or a Tesla movement? A question that will be

answered by gauging Model 3 demand vs. competing Gen-3 EVs, not just
initially but throughout 2018–19. If it’s ultimately more a Tesla movement, the
industry’s reaction could shift somewhat towards accelerating the catch-up
within other Tesla-leading features like OTA, digital displays, and perhaps
standard semi-autonomous hardware. This doesn’t mean that automakers
would stop investing in EVs, but rather that the perception of what’s driving EV
penetration (in the near-term) would shift to a Tesla story more than an EV
story, at least for some period of time.
© 2018 Citigroup
Japan
Japan an HEV Leader While Next-Generation Batteries are Key for BEVs. In
Japan, the FY16 (FY3/17) weightings for BEV and PHEV sales were extremely low,
at 0.3%, respectively. PHEVs and BEVs could hardly be considered to have entered
a dissemination phase, as unit sales trends can largely be explained by the model
cycle of the Nissan Leaf in the case of BEVs and the Toyota Prius PHV and the
MMC Outlander PHEV in the case of PHEVs.
Figure 126. xEV Penetration in Japan Figure 127. Sales in Japan of the Nissan Leaf, the MMC Outlander,
PHEV, and the Toyota Prius PHV
(Units)
Units ('000s)
3,500 80% 60,000
3,000 70%
50,000
60%
2,500
50% 40,000
2,000
40% 30,000
1,500
30%
1,000 20,000
20%
500 10% 10,000
0 0%
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
-
2011 2012 2013 2014 2015 2016 2017
BEV PHEV FCV HEV xEV ratio Outlander Prius PHV Leaf
Source: Next Generation Vehicle Promotion Center, FOURIN, Citi Research Source: Marklines, Citi Research
The market penetration rate for HEVs, meanwhile, has risen sharply, reaching 26%
in FY16 and doubling over the past five years. The top three automakers have
actively launched HEV models, as in addition to Toyota and Honda, which honed in
on HEVs from the outset, Nissan introduced the Note and the e-POWER. In the
mini-vehicle space, Suzuki is stepping up the launch of mild hybrid models.
Japan does not have many incentives (i.e., regulatory, subsidies) aimed at
promoting BEV or PHEV adoption. We believe automakers are likely to expand their
line-ups to keep abreast of overseas regulations, with the effects likely to extend to
Japan. However, we expect only a moderately paced market penetration for BEVs
and PHEVs in Japan, given the few merits to consumers of switching to such
models.
We assume Japan’s BEV weighting increases to 2% in 2025 and to 6% in 2030.

Leaving aside the impact of radical regulatory changes or subsidies that are not
currently under consideration, we think battery cost and performance are likely to
have a major impact on BEV penetration rates. Toyota aims to commercialize all
solid-state batteries in the first half of the 2020s, but we expect these batteries to be
expensive at the outset and therefore we do not think the pace of BEV market
penetration will rise. However, we think BEV sales could increase sharply in Japan
in the latter half of the next decade, when commercial versions of all-solid-state
battery models become available.
We expect Japan’s PHEV weighting to inch up to 2% in 2025 and to 3% in 2030.

PHEVs tick many regulatory boxes in Europe, the U.S., and China, but there is no
clear need for motorists in Japan to make the switch to PHEVs at the moment.
However, motorists could shift from HEVs to PHEVs if fuel economy regulations
were tightened in a way that would favor PHEVs.
© 2018 Citigroup
The HEV still offers the greatest promise in Japan, and we expect the penetration
rate to rise to 47% in 2025 and to 66% in 2030. We think HEVs will remain a
mainstay at Toyota, which is achieving margins comparable to those of gasoline-
powered vehicles in strong hybrids. As a realistic solution, Nissan, which has been
focussing on BEVs since the start of this decade, is strengthening its hand with the
e-POWER (features a series hybrid system). The internal combustion engine is also
evolving thanks to thermal efficiency improvements, and it seems that the
superiority of HEVs, which can improve fuel economy and generate profit by
combining the internal combustion engine with electrification, will not change.
© 2018 Citigroup
Appendix 4
Regulation by Region
Europe
There are currently no specific quotas for EVs in Europe but rather a host of
regulations that favor low emission vehicles and are generating significant push
demand. We summarize the key European regulation in Figure 128 below.
Figure 128. Tougher Testing Standards and Falling CO2 Limits Give OEMs a Regulatory Headache and Spur on the Need for Powertrain Evolution
Overview of European regulatory environment
European Commission CO2 Targets 2008: In 1998 European OEMs agreed to an optional CO2 emissions target of 140g/km by 2008, however by 2007 it was clear the
hurdle would not be cleared and the voluntary target was scrapped with the intention of replacing it with a mandatory target.
2015: A target of 130g/km was set in 2009, which was comfortably reached according to the European Commission, which reports
that in 2016 average emissions were 118.1g/km of CO2.
2020/21: The current passenger car CO2 target was set in 2013 and aims for manufacturers to achieve 95g/km by 2021, being
phased in from 2020. This would represent a 5.1% annual reduction rate from 2015, considerably faster than the 1.7% rate set by
the 2015 target. The targets are currently based on NEDC terms (see Emissions Testing below) and therefore WLTP CO2 values
will be converted back into NEDC terms in order to be compared against the 95g/km target.
2025/2030: The European Commission proposed its post-2020 CO2 targets in November 2017. CO2 emissions will need to fall by
15% (2025) and 30% (2030) compared with the 2021 level. Similarly to the current 2020/21 targets, there is a degree of flexibility as
targets are partially dependent on the weight of a manufacturer's fleet. Manufacturers that sell more zero-emission and low-
emission vehicles (ZLEVs) are also rewarded with a reduction to their average fleet emissions target. The targets are defined in
WLTP testing terms and the base against which the reduction is measured (i.e., the OEMs 2020/21 target) will be converted from
NEDC terms to WLTP.
Super credits: Manufacturers receive super credits for producing low-emission vehicles, which can be used to offset their average
fleet emissions. For each low-emission car manufactured, OEMs are able to count that car as 2 vehicles in 2020, 1.67 vehicles in
2021, 1.33 in 2022, and then 1 from 2023 onwards. Super credits were also available to manufacturers when working towards the
2015 CO2 target.
Euro 6 Emission Limits Euro 6c/d: The Euro 6 standard was introduced by the European Commission in September 2015 and currently sets the emissions
limits with regards to carbon monoxide, NOx, hydrocarbons, and particulates. The standards vary for gasoline and diesel vehicles,
which are measured against Euro6c and Euro6d respectively. The most significant departure from Euro5 is in relation to NOx
emission limits for diesel vehicles, which were reduced from 0.18g/km to 0.08g/km, a 56% reduction. The rising incremental cost for
diesel makes this technology less attractive as a solution to the compliance challenge faced by European manufacturers.
Emissions Testing New European Driving Cycle (NEDC): Historically the Euro Emission Limits have been applied to vehicles tested using NEDC
testing - a laboratory-based drive cycle simulation. However NEDC testing has been criticized for the disparity in results compared
to real-world driving, and from September 2018 this will be replaced with the more strenuous WLTP testing.
Worldwide Harmonised Light-Vehicle Test Procedure (WLTP): From September 2017 all new models introduced to the market
in the EU have had to comply with WLTP testing, and from September 2018 WLTP will apply to all new car registrations, rather than
just new models. Unlike NEDC tests, WLTP is designed to better reflect real-driving on-road performance and therefore aims to
produce more accurate measurements of 15
fuel consumptions and emissions. Car manufacturers suggest the delta will be an
additional 15-20 g/km of CO2 emissions. .
Real Driving Emissions (RDE): Though the WLTP testing conditions are designed to better reflect on-road driving, it is still a
laboratory based procedure. Therefore, in order to enhance the accuracy of its results, vehicles will also be subject to Real Driving
Emissions (RDE) testing in conjunction with WLTP to come to a final conclusion on the vehicle's emissions and fuel consumption
from September 2018.
Key differences: Unlike NEDC, testing under WLTP is conducted over four more dynamic phases rather than just two phases as
part of a single test cycle. WLTP testing involves driving further (23.25km vs. 11km), for longer (30 vs. 20 minutes) and faster (max.
speed 131km/h vs. 120 km/h) than NEDC at a lower test temperate (23°C vs. 20-30°C).
Low Emissions Zones (LEZs) Europe: Over 200 Low Emission Zones have been implemented across Europe in an attempt to improve air quality in European
cities and meet the European Union’s Air Quality Standards. By restricting vehicle access, LEZs aim to reduce emission of fine
particles, NOx, and indirectly ozone. The most polluting vehicles (classified in accordance with the Euro standards) are regulated
within these areas; they are either banned or will have to pay a toll in order to access the area depending on local regulations.
London Ultra Low Emission Zone: Transport For London plan on launching an Ultra-Low Emission Zone based on its current
Congestion Charge Zone from April 2019. Vehicles entering the zone will have to meet minimum emission standards in order to
avoid paying penalty charges. Petrol vehicles will need to comply with Euro 4 standards and diesel will be measured against Euro 6.
Source: Citi Research, ICCT, European Commission, EU WLTP Facts, EU Urban Access Regulations
15
NEDC – WLTP comparative testing, TNO, 10th Oct 16 (link)
© 2018 Citigroup
Tough regulatory landscape increases significance of EV adoption. The

regulatory environment within which European OEMs operate is clearly a catalyst
for EV production. Figure 128 provides an overview of the European regulatory
environment that we argue is a key driver behind the OEMs’ focus on EV and the
growing pertinence of electric powertrain in Europe. Manufacturers are being tasked
both with significantly reducing their average fleet emissions (see Figure 129) and
simultaneously adapting to more stringent testing conditions under both WLTP and
RDE. Furthermore, Low Emission Zones provide an example of how European local
authorities are responding to the growing concern over air pollution levels and act
as a further incentive for consumers to switch to low-emission vehicles.
Figure 129. Fleet CO2 Emissions in 2016 vs. 2020 EU Target
2016A
124 120 123 120 120 125 121 110 110 119 106 102
100%
-11% -11%
-19% -17% -16% -16% -13%
-26% -24% -24% -23% -20%
80%
60%
93 92 104 94 91
40% 91 93 93 96 101 100
92
20%
0%
Hyundai
Volvo
VW
Opel
Ford
BMW
FCA
Daimler
PSA-OV
Ren-Nis
Toyota
PSA
2021 Target 2016
Source: Citi Research, ICCT
CO2 targets pose headwind to European carmakers. In order to meet EU CO2

targets OEMs will have to reduce average fleet emissions to 95g/km by 2020 with a
phase-in to 2021. Specific targets will vary depending on the average mass of the
fleet (heavier cars are permitted higher emissions). The headwind that
manufacturers face ranges between 11% and 26% (see Figure 129) and worryingly
the pace at which average CO2 emissions are declining appears to be slowing (see
Figure 130). Given the rising popularity of SUVs and declining penetration of diesel,
it’s a possibility that the CO2 count could get worse before it gets better. Increasingly
the OEMs are being compelled to produce low or zero emission vehicles.
© 2018 Citigroup
Figure 130. The Average Pace of CO2 Reduction by OEMs is Slowing
5%
4%
3%
2%
1%
0%
2010 2011 2012 2013 2014 2015 2016 2017
Average annual reduction in CO2 emissions
Source: Citi Research, ICCT, Company Data
At first glance targets set for 2025 and 2030 seem less strenuous… The EU’s
proposals of further CO2 limits in 2025 and 2030 will require passenger car fleets to
achieve a 15% reduction by 2025 and 30% by 2030 vs. 2021. This suggests a
slowdown in the pace of improvements required by the EU. In theory, to meet 2021
targets, fleet CO2 emissions need to improve 5.1% per year (2015–21), but under
these proposals that rate is 4% per year (2022–25) and 3.8% per year 2026–30.
…but WLTP testing will make the task more challenging. The targets for 2025
and 2030 will be defined in WLTP testing terms, and the 2020/21 base will be
converted from NEDC terms to WLTP. Therefore, although the annual rate of
reduction may be slower, the introduction of more stringent testing criteria (WLTP
and RDE) makes comparing the difficulty of meeting the 2015 CO2 targets with
2025/2030 tricky.
Diesel decline means OEMs relying increasingly on EV to meet CO2 targets.

Diesel vehicles being ~20% more fuel-efficient than gasoline equivalents means
they play a vital role in reducing average fleet CO2 emissions (see here). However,
the falling diesel penetration (Western European diesel share of new car sales was
down 800bps in January 2017) has limited its dilutive effect on fleet CO2 emissions.
We assume this and the rising popularity of SUVs contributed to the slowdown in
CO2 reduction shown in Figure 130. Consequently, the speed of transition from ICE
to electric powertrain has become increasingly pertinent to OEM efforts to meet
forthcoming CO2 targets. Presumably manufacturers will rely increasingly on EV
volumes to reduce their average emissions.
© 2018 Citigroup
Figure 131. EU Super Credits: Extra Incentive to Produce Low Emission (<50g/km) Vehicles
First stage of emission reductions (2015 CO2 target) Second stage of emission reductions*
2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023
No. of vehicles each low-emitting 3.50 3.50 2.50 1.50 1.00 1.00 1.00 1.00 2.00 1.67 1.33 1.00
(<50g/km) car is counted as
*super credit scheme's contribution to the target will be capped at 7.5g/km per manufacturer over the three years during second stage of emission reductions
Source: Citi Research, European Commission
Super credits add extra incentive to produce electric. To help reach the 2015
CO2 target, the European Commission introduced a “super credit” scheme.
Between 2012 and 2015 every vehicle produced emitting below 50g/km CO2 earned
manufacturers “super credits” that could be used to reduce their average fleet
emissions (see Figure 131). Each low or zero-emission vehicle would be counted as
higher number of vehicles in accordance with the table shown in Figure 131. This
scheme is set to restart in 2020 to support OEM efforts to reach the 2020/21
targets. Though the scheme is currently set to phase out by 2023, the EU has
suggested a credit system will be available in relation to 2025 and 2030 targets. The
details of this system are not yet clear, but we assume it will be a continuation of the
super credit scheme.
Government Bans and Incentives

European cities and nations announce bans on ICE powertrain. Regulation of
the ICE powertrain is tightening across Europe. Bans on new gasoline and diesel
sales have been proposed in the Netherlands by 2030, plus in France and the U.K.
by 2040, and Norway wants all new cars sold by 2025 to be zero emission (see
Figure 133). Also, a number of European cities have already introduced restrictions
and bans on the use of diesel vehicles. Therefore, although the pace at which EV
penetration will grow is less certain, it appears to only be a matter of time before
almost all new cars sold come with an electrified powertrain of some description.
Incentives encourage EV adoption by reducing total cost of ownership. In

order to encourage the powertrain transition from ICE to electric and to support
demand for EVs, European governments and cities have introduced a swathe of
incentives for both individuals and companies (see Figure 132). Purchase subsidies
and tax benefits help reduce the total cost of ownership of an EV, which is a key
barrier to entry for many consumers. For example, the boom in EV stock in Norway
(from 5,380 in 2011 to 133,260 in 2016) has been largely driven by generous
subsidies. In Norway, a BEV is exempt from purchase tax, VAT, and toll road fees,
and in the capital city Oslo EVs enjoy free charging, free parking facilities, and
access to the bus lane.
Figure 132. Incentives for Europe’s Ten Largest EV Markets

Purchase Registration Ownership Tax Company Tax VAT Benefits Other Financial Local Infrastructure
Subsidies Tax Benefits Benefits Benefits Benefits Incentives Incentives
Austria      
Belgium    
France     
Germany     
Netherlands   
Norway       
Spain       
Sweden   
Switzerland  
UK      
Source: Citi Research, EAFO
© 2018 Citigroup
Figure 133. European Governments are Planning for Life After ICE and Setting Themselves Targets for EV Adoption
© 2018 Citigroup
EV demand appears to be highly sensitive to changes to incentives. This

raises the question of how EV demand will react when/if generous incentives
currently in place are withdrawn. Figure 134 below shows the impact increasing and
decreasing EV support incentives had on sales growth in 2016 and suggests
16
demand is highly sensitive to such changes.
Figure 134. EVs Highly Sensitive to Changes to Incentives Figure 135. Danish EV Market Shrinks Following Subsidy Withdrawal
200%
2016 Sales impact from Incentive Decrease 3,000
164%
2016 Sales impact from Incentive Increase
Denmark BEV & PHEV Sales (units)

150% Tax incentives scaled back
2,500
100%
2,000
Sales growth
50% 1,500
0% 1,000
-50% 500
-50%
-60%
-100% -
Denmark Netherlands Norway 4Q14 2Q15 4Q15 2Q16 4Q16 2Q17 4Q17
(BEV & PHEV) (PHEV) (PHEV)
Source: Citi Research, IEA Source: Citi Research, ACEA
We see an example of this in Denmark, where the scaling back of the Danish tax
incentives that began in the first quarter of 2016, led to a 60% decrease in EV sales
that year. As a result, the Danish state chose to stall its intended subsidy cuts in
order to prevent the market collapsing. Instead of 40% of the full registration tax in
2017, this rate has been capped at 20% until an additional 5,000 EVs are sold or 1
January 2019 is reached, at which point it will rise to 40%. The tax will increase to
65% in 2020, 90% in 2021, and 100% in 2022 (2 years later than originally
planned).
Announced in April 2017, plans to slow the pace of subsidy withdrawal have had a
mixed impact on the EV market in Denmark. Although Q2 and Q3 sales in 2017 are
up 94% and 112% month over month, respectively, they remain down 38% and 40%
year over year, and sales for the first three quarters are down 68% since 2015,
when the tax exemption still applied in full. This volatility is not helpful to carmakers
or those involved in the EV supply chain.
Norway’s EV market economics are being questioned. In contrast to Denmark’s

collapse and the 50% decline in the Netherlands following a rise in PHEV tax rates,
Norway’s PHEV sales more than doubled, growing 164%, after Norway increased
its their purchase subsidies and tax exemptions in 2016. This led to some
Norwegian politicians questioning the justification for the extent of their incentive
provisions: Despite the subsidized BEVs accounting for only 5% of daily commutes,
Oslo’s incentives for BEVs amounted to over half as much as the total city public
transport bill.
The dependence on temporary subsidies for the creation of demand creates

substantial uncertainty regarding activity post the expiry of these incentives.
While of course governments could sustain/increase subsidies to sustain growth,
the cost to the state as EV sales penetration grows would not in our view be
sustainable financially as a long-term solution.
16
In 2016 tax incentives were scaled back in Denmark, PHEV tax rates increased in the
Netherlands, and purchase subsidies and tax exemptions increased in Norway.
© 2018 Citigroup
Figure 136. European Incentives Are Key to Driving EV Adoption
© 2018 Citigroup
China
The boom in China, which started a few years ago, was largely triggered by
generous government subsidies on NEV for both manufacturers and customers.
Subsidy will continue to play a major role in stimulating NEV deployment in China.
Besides, China announced a dual credit management system in September 2017 to
encourage the development of NEV under a more market-oriented approach, which
we believe would give a strong push to NEV sales in 2019–2020E, with a
requirement for manufacturers to reach 10%/12% NEV credit based on their
conventional vehicle sales volume. Details of the latest 2018 NEV subsidy policy
can be found in Figure 137.
Figure 137. Details of 2018 Final EV Subsidy Policy

2017 2018 Draft 2018 Draft 2018 Final vs. Draft
version 1 Version 2 Policy Policy
Passenger Vehicles
Cap subsidy amount (Rmb/kwh) 1,100 1,100 J
Subsidy per vehicle (Rmb/unit)
By driving range:
100-150km 20,000 - - - --
150-200km 36,000 10,000 15,000 15,000 --
200-250km 36,000 25,000 25,000 25,000 L
250-300km 44,000 34,000 34,000 34,000 --
300-400km 44,000 45,000 45,000 45,000 --
≥ 400 km 44,000 50,000 50,000 50,000 --
Subsidy multiplier
Battery energy density (wh/kg)
<90 -- -- -- -- --
90-105 1.0x -- -- -- --
105-120 1.0x 0.5x 0.5x 0.6x J
120-140 1.1x 1.0x 1.0x 1.0x --
140-160 1.1x 1.1x 1.1x 1.1x --
>60 1.1x 1.1x 1.1x 1.2x J
Electric bus
Subsidy amount (Rmb/kwh) 1,800 1,100 1,440 1,200 middle
Cap subsidy per vehicle (Rmb/unit)
6-8m 90,000 50,000 72,000 55,000 middle
8-10m 200,000 120,000 120,000 120,000 --
>10 300,000 180,000 240,000 180,000 --
Subsidy multiplier
Battery energy density (wh/kg)
<85 -- -- -- -- --
85-95 0.8x -- -- -- --
95-110 1.0x -- -- -- --
110-115 1.0x 0.8x 0.8x -- L
115-120 1.2x 0.8x 0.8x 1.0x J
120-135 1.2x 1.0x 1.0x 1.0x --
135-140 1.2x 1.0x 1.0x 1.1x J
>140 1.2x 1.2x 1.2x 1.1x L
Special Vehicle
Cap subsidy amount (Rmb/unit) 150,000 100,000 100,000 100,000 --
Subsidy per battery capacity (Rmb/kwh)
≤ 30kwh 1,500 900 850 850 --
30-50kwh 1,200 750 750 750 --
>50kwh 1,000 650 650 650 --
Source: Ministry of Finance, Ofweek, Citi Research; Note: EV subsidy amount = base subsidy amount x energy
density multiplier x energy efficiency adjustment factor
© 2018 Citigroup
Phasing Out Subsidy Accelerate Industry Consolidation
The Chinese government started subsidizing NEVs in 2009. By implementing a

series of actions to plug subsidy leakages and by introducing a set of requirements
on battery size, energy density, pure electric range, etc., the government now
focuses on subsidies for high-quality NEVs.
According to news reports over the past two months, the potential 2018 NEV
subsidy adjustment appears to be much stricter on pure-electricity range (R) and
battery-energy density than the previous ones, potentially with -100% to +14%
changes for BEV PV, -8% changes for PHEV PV, and -100% to -40% changes for
BEV bus. This, in our view, will further enlarge the gap between high- and low-
quality NEV makers’ subsidy, thus accelerating industry consolidation. We think this
also shows the strong government determination to make the NEV industry lean —
with no “fat” but “muscle.” The Chinese government started subsidizing NEVs in 2009.
By implementing a series of actions to plug subsidy leakages and by introducing a set
of requirements on battery size, energy density, pure electric range, etc., the
government now focuses on subsidies for high-quality NEVs.
We show the latest 2018 NEV subsidy policy in Figure 137. The magnitude of
subsidy cut is generally in line with draft policies circulated in the market previously.
Key interpretations from the materials demand point of view:
Higher-energy density requirement to support NMC battery and cobalt

demand. Government adopts differentiated treatment in the passenger vehicle sub-
sector: incentive is cut for short-driving-range cars (<400km) and low-energy-density
batteries (<140wh/kg), but subsidies for high-end ones are increased. During 2017,
there is no incentive for producers to produce driving-range >400km & energy-density
>160wh/kg batteries because they could already enjoy the highest subsidies once
they reach 300km and 120wh/kg. The new policy encourages producers to develop
even further.
Magnitude of cut is lower than the market expectation. Final 2018 subsidy per
kwh reached Rmb1,200/kwh vs. Rmb1,100/kwh in the draft policies for both e-bus
and e-PV. Subsidy policy of SPV is generally in line.
More measures introduced to resolve cash flow pressure resulting from

lengthy time to collect subsidy. Required accumulated driving mileage for non-
private NEVs is reduced from 30,000km to 20,000km. A portion of the subsidy will be
granted when the vehicle is sold; the rest is paid when the mileage meets
requirement. This could lead to healthier cash flow along the EV supply chain.
The new policy will be implemented from June 12. From Feb 12 to June 11, a
temporary subsidy policy will be applied for the transition period: ePV and e-bus will
get 0.7x 2017 subsidy amount and special vehicles will get 0.4x 2017 subsidy
amount. Most ePv and eBus will enjoy a higher subsidy in transition period, compared
with after June. Therefore, we think a moderate level of rushing sales and installation
will continue to take place, to support demand in the next few months.
Dual Credit System Stimulates NEV Production
China introduced the dual credit system in September 2017. This management
system consists of corporate average fuel consumption (CAFC) credit and NEV credit,
and its aim is to push down average fuel consumption for gasoline cars and push up
the production for NEVs. The policy came into effect on April 1, 2018, and is
practically applicable to all nameable passenger vehicle OEMs in China.
© 2018 Citigroup
CAFC credit is calculated by the OEM’s annual production volume times the
difference between its actual and target CAFC. Positive credits can be carried forward
for no more than three years at a conversion ratio of 80% for credits earned in 2018
and before or 90% thereafter. Credit can also be transferred within related companies,
defined by shareholding at or more than 25%. Negative credits should be offset by
those from previous years or from related corporates, and, if not enough, by NEV
points on a 1:1 basis. CAFC credit must be at a non-negative balance for each year;
otherwise, the government will suspend the OEM from launching new models or
implement a partial production suspension of its models with high fuel consumption.
NEV credit is calculated by the OEM’s annual production volume times the respective
point for each model; PHEVs get 2 points, and BEVs can earn up to 5 points based
on their range per charge under the formula 0.012R+0.8 (i.e. BEV with range at
350km or above can earn full score at 5). NEV credit cannot be carried forward
(except 2016 when a 1-year carry forward is allowed), and cannot be transferred with
related companies; it can, however, be traded freely through MIIT’s platform. NEV
insufficient credit can only be balanced by NEV credits purchased from others.
Depending on the number of gasoline vehicles produced, OEMs are required to
generate NEV credit up to 10%/12% of that volume in 2019/20E. Failure in achieving
the requirement can result in production suspension of ICE vehicles.
In short, the big producers of conventional vehicles will be compelled to subsidize

NEV manufacturers if they aren’t able to ramp up their NEV products in time, thus
effectively forcing the auto industry to take on the financial burden for NEV to reduce
the Chinese treasury’s involvement in order to foster long-term development for the
sector. In answer to this, many conventional car makers have set up partnership with
NEV producers to leverage their NEV credits as an answer to the NEV credit
requirement over the past year.
Figure 138. Duel-credit Management System

Regulatory authority Ministry of Industry and Information Technology (MIIT)
Two parallel system CAFC credit NEV credit
Calculation method CAFC credit = (Target CAFC - Actual CAFC) x # of vehicles NEV credit = NEV point / vehicle x # of vehicles
BEV with R>50km has multiplier impact of 5x/3x/2x in 2016-17/2018-19/2020 NEV point per BEV = R x 0.012 + 0.8 (cap at 5)
when calculating CAFC NEV point per PHEV = 2
PHEV with fuel consumption < 2.8L/100km has multiplier impact of
3.5x/2.5x/1.5x in 2016-17/2018-19/2020 when calculating CAFC
Management method - CAFC negative credit can be offset by CAFC positive credit earned from - NEV negative credit can only be offset by NEV positive
previous year, transferred from related corporates, or by NEV positive credit points via purchases from other manufacturers
- CAFC credit is allowed to be carried forward for at most 3 years (with 80% - NEV credit can be trade freely on MIIT's platform, but
conversion ratio in 2018 and 90% in 2019 onwards) and can be transferred cannot be re-sold.
within related corporates (shareholding at or more than 25%) - NEV credit is not allowed to be carried forward
- The 2019/20 NEV balance will be examined together
Assessment companies All PV OEMs selling in China (including import) All PV OEMs with annual production volume or import
volume greater than 30k units in China
Assessment criteria A positive balance under GB 27999-2014 2019/2020 NEV point to # of non-NEV vehicles ratio at
10%/12%
Assessment period 2016+2017, 2018, 2019, 2020 2019+2020
Penalty measures Suspend application of car models that do not meet GB27999 standard and Suspend partial production of ICE models
suspend partial production of high fuel consumption models
In the table below, we present our calculation on the industry’s positioning under the
dual credit system. Based on our forecast for NEV sales volume reaching 1.5m
units by 2020E from 0.3 million units in 2016, and assuming average range for BEV
to progressively increase to 300km in 2020E from 150km in 2016, the total NEV
© 2018 Citigroup
points earned for all the NEV makers in China will be 3.8/5.5m points in 2019/20E.
Compared with the 2.6/3.4m points NEV credit requirement in 2019/20E, derived
based on the 10%/12% NEV point to conventional vehicle volume ratio as required
by the MIIT, NEV credit surplus for the auto industry as a whole will be 1.1/2.1m
points in 2019/20E, according to our calculation.
Figure 139. Estimation of NEV Credit and CAFC Credit for the Auto Industry in 2017-20E
2016 2017E 2018E 2019E 2020E
NEV credit (mn points) 0.83 1.23 2.11 3.76 5.50
BEV 0.67 1.04 1.78 3.19 4.56
PHEV 0.16 0.20 0.33 0.57 0.94
NEV credit to conventional vehicle ratio - - - 10% 12%
NEV credit requirement (mn points) - - - 2.63 3.39
NEV credit surplus (mn points) 0.83 1.23 2.11 1.13 2.11
CAFC deficit after intra-group offset (mn points) (0.70) (0.72) (0.76) (0.80) (0.87)
PV sales volume (mn units) 23.7 24.4 25.9 27.3 29.8

Conventional vehicle 23.3 23.9 25.2 26.3 28.3
New energy vehicle 0.3 0.5 0.7 1.0 1.5
BEV 0.3 0.4 0.6 0.7 1.0
PHEV 0.1 0.1 0.2 0.3 0.5
Proportion(as % of total volume)

Conventional vehicle 99% 98% 97% 96% 95%
NEV 1.4% 2.0% 2.8% 3.7% 5.1%
PHEV 1.1% 1.6% 2.2% 2.7% 3.5%
BEV 0.3% 0.4% 0.6% 1.0% 1.6%
Source: Citi Research estimates
Intensified CAFC pressure under the industry’s upsizing trend. According to

MIIT’s release and based on our calculation, among the 42 auto groups in China,
29% of them or 12 groups recorded a CAFC point deficit in 2016 (after considering
intra-group point off-set), topped by GWM with -0.3m points and Changan with -
0.1m points. Total negative CAFC balance for the industry was at -1.37m points,
and after intra-group off-set at -0.7m points.
Top performers are BYD with +1.7m points, SAIC group with +1.5m points, and
BAIC group with +1.4m points, followed by Geely and GAC group. For most of
them, the high credit is thanks primarily to their engagement in NEV, which helped
lower its CAFC (as BEV is considered to have 0 fuel consumption and have a
multiplier impact of 5x/3x/2x in 2016–17/2018–19/2020 and PHEV’s fuel
consumption is also substantially lower than that of a gasoline vehicle with similar
multiplier impact of 3.5x/2.5x/1.5x in 2016–17/2018–19/2020), especially given most
of them recorded negative balance back in 2015.
While CAFC credit is not a problem for most of the OEMs as of now, it is important
to note that (1) when NEV credit assessment commences in 2019–20, a company
that has no other way but to use NEV points to offset CAFC negative credit will face
higher pressure in achieving the 10/12% NEV credit requirement, and (2) OEMs
with abundant CAFC credit will have a much larger room to seize the vehicle
upsizing trend in the coming few years (from class A or below to class B or above,
and from sedan to SUV, as we discussed in the previous sections) to gain market
share. In other words, we think NEV is not only essentially for the OEMs in meeting
the NEV credit requirement, but also helpful in lowering the CAFC to give room for
gasoline vehicle upsizing.
© 2018 Citigroup
Figure 140. Comparison of CAFC Credit for Leading Auto OEM Groups vs. Others in 2016
(CAFC credit, mn points)
2.0
BYD
1.6
SAIC
BAIC
1.2
0.8
Geely
GAC
0.4 DFM
Brilliance
0.0
Chang'an
GWM
(PV sales volume, mn units)
(0.4)
0 1 2 3 4 5 6
Figure 141. CAFC Balance for Auto OEMs in 2016 vs. 2015
(mn CAFC points)

2.0 1.7
1.5
1.5 1.3
1.0 0.8
0.6
0.4 0.3
0.5
0.0
(0.5) -0.2 -0.1 -0.2
-0.3 -0.3
(1.0) -0.5
-0.9
(1.5) -1.1 -1.1
-1.3
(2.0)
(2.5) -2.3
(3.0)
GAC
SAIC
BAIC
Geely
Brilliance
GWM
BYD
DFM
Changan
2016 2015
In addition to the sector top-down analysis on the dual credit system, in this section,
we present a more detailed bottom-up analysis. Our core conclusion is that the NEV
credit requirement, as stretched as it may seem, is easily achievable for the majority
of the OEMs.
© 2018 Citigroup
In conjunction with the core conclusion, given possible a NEV credit surplus in
2019/20E, we find that (1) BEV (that can earn as much as 5 points/ vehicle if the
range is longer than 350km) is not a must, that our expectation for PHEV to surge
faster than BEV in 2019–2020E is reasonable, and (2) should vehicle upsizing
result in a negative CAFC balance (which will happen according to our estimation),
NEV credit can offset it.
To measure each OEM’s positioning, we assume a conservative case where OEMs

either maintain their NEV volume proportion status quo or ramp up to a minimal
level to meet the requirement. Under this methodology, the NEV volume CAGR for
the OEMs will hugely diversify, ranging from 10% for BYD to over 200% for GAC
and GWM. NEV’s share of total vehicle sales ranges from 2.7% for DFM to 32% for
BYD, after taking into account a different mix for BEV/PHEV. The majority of the
OEMs can easily achieve the NEV point requirement by 2020E based on our
forecasts, the volume of which is lower than some of the internal targets, according
to news reports.
Figure 142. NEV Volume CAGR in 2017-20E for Major OEMs Figure 143. NEV’s Share of Total Vehicle Sales Volume in 2020E for
Major OEMs (status quo or raised to meet the minimum credit requirement)
240% 225% 60%
218%
200% 50% 48.0%
160% 40%
120% 107% 30%

96%
Industry 78% 80% Industry average
80% CAGR: 45% 20% penetration:5.1%
10% 8.6%
40% 21% 6.0%
23% 4.8% 4.3% 4.0% 3.9% 3.6% 3.3%
12%
0% 0%
GAC
GAC
Geeky
BAIC
GWM
Brilliance
SAIC
DFM
BYD
Chang'an
BAIC
Geely
SAIC
Brilliance
BYD
Chang'an
DFM
GWM
Source: Citi Research estimates Source: Citi Research estimates
© 2018 Citigroup
U.S.
U.S. Regulatory Landscape: Agnostic at the Core
Looking back historically at U.S. vehicle regulations highlights a proactive approach

to achieving road safety and environmental benefits for all stakeholders involved.
Regulations help these stakeholders by: setting standards for a vehicle to be
considered road worthy (FMVSS); introduction of emission and fuel efficiency
targets; and the passing of new safety requirements, to just name a few. A
commonality across these regulations, from what we’ve noticed historically, is that
they are structured to be agnostic from a technological perspective. This means that
many different types of technology may be used to achieve the desired result.
Explanatory Example
Let’s take rear-vision safety as an example. In 2007 the U.S. Congress passed
legislation requiring vehicles to have rearward visibility – meant to detect areas
behind the motor vehicle to reduce death and injury resulting from backing
incidents, particularly incidents involving small children and disabled persons. The
law stated that this may be achieved by the provision of additional mirrors, sensors,
cameras, or other technology to expand the driver’s field of view. This law was
completely technology agnostic and as such OEMs were at liberty to use whatever
technology they deemed most successful.
CAFE Standards
U.S. fuel efficiency regulations (CAFE standards) have been in a state of political
flux as of late. Regulatory targets for 2022+, which were previously set following the
mid-term review under the Obama administration, are now once again being
reviewed by the current administration and it seems likely these will be rolled back.
The Impact on Propulsion Technology in the U.S.
However, we don’t believe a scrapping or reduction in fuel efficiency targets really

slows down the powertrain electrification (hybrids & EVs) push that OEMs have in
place, in order to help them reach the prior or potentially new CAFE targets. In our
discussions with suppliers we hear that OEMs simply aren’t waiting for regulations
to be set, rather they are proactively moving at electrified powertrains and other fuel
efficiency technologies (i.e., lightweighting) in order to reach targets. This
reinforces our prior comment on regulations being technology agnostic. In this
instance we can see OEMs deploying different powertrain technology (more fuel
efficient ICEs, hybrids, EVs) and other technologies. While some technologies, such
as EVs, do offer greater flexibility based on MPG/range calculations (as it pertains
to the vehicles footprint), we believe that this flexibility isn’t primary driver of EV
penetration/adoption here in the US. Rather, we believe that the driver of EV
demand will stem more from consumer pull vs. a regulatory push
© 2018 Citigroup
Japan
In 2015, Japan submitted an action plan ahead of the Paris Agreement which
targeted a 26% reduction in greenhouse gas emissions relative to 2013 by 2030. As
part of this it set a fuel efficiency target of 20.3km/L by 2020, compared with
17.0km/L in 2015.
The Japanese government has announced a number of policies to promote uptake

of next-generation vehicles over the past several years.
2009 – Next-generation vehicles, including hybrids, are exempted from the

acquisition and weight taxes (the latter is determined by vehicle weight) from 2009.
The tax rate was decreased for traditional vehicles as well that meet stipulated
emission reduction standards.
2010 – METI (Ministry of Economy, Trade and Industry) announces its “Next-
generation Vehicles Strategy 2010”, comprising six main policies (Figure 144).
Figure 144. Six Key Policies in the “Next-generation Vehicles Strategy 2010”
Targets Action Plan
Overall Plan Next Gen vehicle development and production Next gen vehicles account for up to 50% in 2020
Promote the use of various fuels
Promote the siting of low-emission industries
Batteries Secure battery technology by R&D Improve performance of LiB
Develop post-LiB
Achieve economies of scale by promoting EVs
Rare Metals Secure rare metals and build recycling systems Strategically secure rare metals
Develop batteries and motors free of rare metals
Establish battery recycling system
Infrastructure Install 2 million normal chargers and 5,000 quick chargers Build infrastructure during market preparation phase
Collaborate with the private sector for further penetration
Systems systemize the vehicles Establish new business model
Verify the system in the social demonstration program
Promote global standardization
Global Standards strategic global standardization Establish global standards for battery performance and safety level
Set a global standards for charging connectors/systems
Promote collaboration of public-private organization
Source: Citi Research.
2014 – The “Automobile Strategy 2014” shares similar goals to the “Next-generation
Vehicles Strategy 2010,” and added more detailed plans regarding the global
market and further development of the market and industry. The country is aiming to
increase the next-generation vehicle rate to 70%, compared with less than 30% as
of 2015. We think this is an attainable target for HEVs in 2030. We note that BEVs
and PHEVs were not broken out in the plan.
Figure 145. Next-generation Vehicles Promotion Plan in the “Automobile Strategy 2014”
2015 (actual) 2020 (goal) 2030 (goal)
Traditional gasoline 73.5% 50~80% 30~50%
HEV 22.2% 20~30% 30~40%
BEV/PHEV 0.6% 15~20% 20~30%
FCV 0.01% ~1% ~3%
CDV 3.6% ~5% 5~10%
Source: METI, Citi Research.
© 2018 Citigroup
2015 – METI allocates funds to EV-related projects: ¥30 billion for EV charging
facilities (2014), ¥30 billion for EV purchase subsidies (2015), ¥2.5 billion for lithium-
ion battery R&D (2015), and ¥3.1 billion for fundamental new battery research
(2015).
Figure 146 shows what subsidies a buyer gets when purchasing different EV
models. These amounts are not significant relative to the total price of the vehicle in
question, so we do not think this will do a lot to stimulate EV uptake.
Figure 146. Subsidy Amount by Model

Subsidy Subsidy
Category Maker Model
( '000 Yen) ($, as of 1/19/2018)
BEV Tesla Model S/ ModelX 400 3,611
Nissan Leaf 228-400 2,058-3,611
BMW i3 390 3,521
VW e-golf 301 2,718
Mitsubishi i-MiEV 120-172 1,083-1,553
PHEV Toyota Prius PHV 200 1,806
BMW 330e/530e/740e 200 1,806
VW Passat GTE 200 1,806
Porche Panamera4 E-hybrid 200 1,806
Volvo V90/XC60/XC90 200 1,806
FCV Toyota MIRAI 2,020 18,238
Honda CLARITY 2,080 18,779
Source: Citi Research.
FCVs are another category of next-generation vehicles that the Japanese

government is interested in developing. The government targets FCV ownership of
40,000 units by 2020 and 800,000 by 2030, which looks a high hurdle. FCVs are
more energy-efficient than EVs and do not need charging, and they also do not
release any harmful emissions. Moreover, the government expects to generate
synergies with fuel cell supply chain companies in Japan. On the other hand, the
lack of hydrogen infrastructure is likely the biggest obstacle to FCV popularization.
We think there is still a chance FCVs could gain market share if improvement in
lithium-ion batteries and next-generation batteries take longer than expected (or
their efficiency reaches its limits).
Japan is not the one of the most aggressive participants in BEV market. The
government does have a plan for BEV promotion, but it lacks teeth. We think
Toyota, one of the world’s largest automakers, may be one reason for this. The
Toyota Prius is the most successful HEV, having sold more than 10 million units
worldwide (and cut a total of 77 million MT of carbon dioxide emissions thus far).
There seems little motivation to promote BEVs, due to Toyota’s dominant position in
HEVs.
© 2018 Citigroup
Rest of World
Korea: No Exception for Car Electrification via Stringent Regulations
Just as in other countries, Korea is setting more stringent standards for fuel-
efficiency and gas-emissions; automakers in Korea (for their domestic sales) should
improve average fuel-economy to 24.3km / liter by 2020 (from 17km/ liter in 2015),
while CO2 emission standards will be tightened to 97g/ km by 2020 (from 140g/ km
in 2015). That said, the Korean government has introduced various subsidy
programs in order to stimulate demand for NEV or xEV; both national and local
governments provide subsidies in the form of tax-benefits and cash incentives.
Figure 147. Fuel-Efficiency Standards in Key Markets

Region Fuel economy Standard
Fuel efficiency 17.9km/l, CO2 emission 130g/km by 2015
Europe
U.S.
China
Fuel efficiency 20km/l, CO2 emission 117g/km by 2020
Japan
Fuel efficiency 17km/l, CO2 emission 140g/km by 2015
Korea
Source: ICCT, Citi Research
Figure 148. Korea: Government Subsidies for NEV

Frequency Area Tax/ fee Type Regular Tax Incentive for BEV
Individual consumption (a) 5% of vehicle base price Max W2mn reduction
National Education (b) 30% of base price Max W0.6mn reduction
One Time VAT 10% of (base price+a+b) n.a.
Acquisition 7% of (base price+a+b) Max W1.2mn reduction
Local
Public Bond 9-20% of (base price+a+b) Max W0.7mn reduction
Vehicle KRW80-200/cc Exempt
Annual Local
Education 30% of annual vehicle tax Exempt
*BEV: Battery Electric Vehicles
Source: MOLIT, Citi Research
© 2018 Citigroup
Appendix 5
Product Pipeline by Manufacturer
Figure 149. Germany’s ‘Big Three’ BEV Product Pipeline by Segment
A Segment B Segment C Segment and SUV Unclassified

above
i8 (2022)
3 Series (2020)
BMW X3 (2019) Other (2020)
I5 (2021)
iNext (2021)
Mini Mini One (2019)
Phantom (2020)
Rolls Royce
Ghost (2021)
BMW Group 11,656 9,689 3,638

-
2021e BEV Volumes (18%) (15%) (6%)
EQ D Sedan (2019) EQ D SUV (2018)

EQ E Sedan (2019) EQ E SUV (2019)
Mercedes EQ A (2020) Other (2020)
EQ F Sedan (2021) EQ C SUV (2022)
EQ C (2022) EQ F SUV (2021)
Smart forfour (2018)

Smart
Smart BSUV (2022)
Daimler 15,094 19,535 24,528 22,873 11,856

2021e BEV Volumes (16%) (21%) (26%) (24%) (13%)
Q E-Tron (2018)
A9 E-Tron (2020) Q5 BEV (2021)
Compact EV (2020) Q6 BEV (2018)
Audi Q2 (2019) D SUV (2021) & D Other (2019)
A5 (2021)
SUV Sportback
R8 (2021) (2022)
Q8 BEV (2022)
Other (2020) Bentley SUV (2021)

Bentley
Speed 6e (2022) Bentayga (2022)
Lamborghini Urus (2022)
D-SUV (2021)
Porsche Mission E (2019) Other (2020)
Macan (2021)
SUV EV (2020)
Seat Mii (2019) Other (2020)
Compact EV (2021)
Octavia (2019)
Skoda Citigo (2019) Coupe SUV (2020) Other (2020)
Compact EV (2021)
Golf (2018)
New Bora (2018)
I.D. Crozz (2020)
ID (2019)
VW Brand I.D. Lounge (2021) Other (2019)
Lavida (2020)
X BEV (2021)
ID AEROe (2021)
T-Roc (2021)
VW Group 10,385 1,477 179,867 85,796 160,040

2021e BEV Volumes (2%) (0%) (41%) (20%) (37%)
© 2018 Citigroup
Figure 150. PSA and Renault BEV Product Pipeline by Segment

above
Citroen C3 (2020) C4 Cactus (2021) Other (2019)
DS DS3 SUV (2019) DS7 Crossback Other (2020)

(2021)
City EV (2021)
Opel Other (2020)
Corsa (2019)
208 (2019)
Peugeot Other (2020)
2008 (2020)
PSA Group 2,683 45,848 9,174 6,420

-
2021e BEV Volumes (4%) (71%) (14%) (10%)
Dacia Duster (2022) LE (2019) Other (2019)
LE (2019)
Infiniti Other (2019)
Infiniti EV (2021)
Mitsubishi eK Wagon (2020) Xpander (2022) SUV EV (2020) ASX (2019)
Nissan DAYZ (2020) SUV EV (2020) Other (2019)
Zoe (2019)
Renault SUV EV (2020) Other (2020)
Kwid (2020)
Renault-Nissan 38,636 7,172 129,043 2,245 63,853

2021e BEV Volumes (16%) (3%) (54%) (1%) (27%)
Figure 151. U.S. Manufacturers BEV Product Pipeline by Segment

above
C-SUV (2020)
Ford Small SUV (2020) Other (2021)
Focus (2020)
Ford 4,195 21,042 5,962

- -
2021e BEV Volumes (13%) (67%) (19%)
Compact MPV
Buick Encore (2019) (2019) Other (2020)
Midsize Car (2019)
Cadillac XT4 (2020)
Chevrolet Bolt (2020) Other (2020)
GM 41,139 11,092 47,120

- -
2021e BEV Volumes (41%) (11%) (47%)
Model 3 (2018)
Tesla Model Y (2020) Model X (2018)
Roadster (2020)
Tesla 210,547 33,265

- - -
2021e BEV Volumes (86%) (14%)
© 2018 Citigroup
Figure 152. Japanese Manufacturers BEV Product Pipeline by Segment

above
Clarity (2018)
DF HR-V (2018)
Honda Urban EV (2019) Other (2020)
Fit/Jazz (2020)
Sports EV (2020)
Honda 11,997 32,973 41,652

- -
2021 BEV Volumes (14%) (38%) (48%)
Compact EV (2022)
Genesis Other (2021)
G80 (2020)
Hyundai Kona (2018) Elantra/Avante Other (2020)

(2020)
Niro (2018)
Kia Stonic (2019) Other (2019)
K3 (2020)
Hyundai Group 32,384 32,973 41,652

- -
2021e BEV Volumes (21%) (38%) (48%)
© 2018 Citigroup
Appendix 6
Navigating the Shift from NEDC to WLTP
The existing lab-test for vehicles, NEDC (New European Driving Cycle), which was
designed in the 1980s, was superseded by a new test; namely WLTP (Worldwide-
harmonized Light-vehicle Test Procedure) in September 2017. The rationale for the
move to WLTP was the previous test was seen as outdated, given it had not
evolved with the technological advancement of vehicles. WLTP provides a more
accurate representation of vehicle emissions and fuel-efficiency.
Figure 153. Summary of What Changes from NEDC to WLTP
Source: wltpfacts.eu
© 2018 Citigroup
Timeline for Transition from NEDC to WLTP
The new test will be enforced gradually; it started with all ‘new-type’ cars (i.e. newly
introduced models) that were tested as of September 2017. From September 2018
WLTP will apply to all newly registered cars (i.e., cars physically on the road). Until
the end of 2018, the NEDC-CO2 value will be used to calculate the tax-band for the
vehicle, however from the January 1, 2019 all cars in dealerships will quote WLTP-
CO2 values (there will be a year’s grace for outgoing models i.e. those approved
under the NEDC test). It is worth noting the CO2 targets for 2021 will continue
to be based on the NEDC test. Given the increased rigor of the WLTP test it will
result in a higher CO2 emission figure than under NEDC (c15% more as a rule-of-
thumb).
Real Driving Emissions Testing Is Also Being Conducted
On top of the change from NEDC to WLTP, an additional test, RDE (Real Driving
Emissions), was implemented from September 2017. This does exactly what it says
on the tin; that is to measure car emissions/ pollutants (e.g. NOx) of a car when
driven on a road. The automotive industry has come under increased scrutiny in
recent years, this heightened after VW’s use of ‘cheating-devices’ came to light in
September 2015, and the objective of RDE is to ensure vehicles are as efficient in
real-driving conditions (i.e. on the road) as when being tested in a lab. An RDE test
uses a ‘portable emissions measurement system’ (PEMS), which is attached to the
vehicle during an on-road test. As of September 2017, all new car models must
comply with a ‘not to exceed’ (NTE) limit of 2.1x permissible NOx emissions (which
under Euro 6 is 80mg/km). From September 2019 the same conformity factor (2.1x)
will apply to all newly registered vehicles. The conformity factor reduces to 1.5x
from January 2021 on all newly registered vehicles.
© 2018 Citigroup
Citi Global Perspectives & Solutions (Citi GPS) is designed to help our clients navigate
the global economy’s most demanding challenges, identify future themes and trends, and
help our clients profit in a fast-changing and interconnected world. Citi GPS accesses the
best elements of our global conversation and harvests the thought leadership of a wide
range of senior professionals across the firm.
All Citi GPS reports are available on our website www.citi.com/citigps
ePrivacy and Data Disruptors at the Gate

Protection Strategic M&A for Managing
Privacy Matters: Navigating Disruptive Innovation
the new World of Data April 2018
Protection
May 2018
Sustainable Cities The Public Wealth of Cities

Beacons of Light Against the How to Turn Around Cities
Shadow of Unplanned Fortunes by Unlocking Public
Urbanization Assets
April 2018 March 2018
The Bank of the Future Investment Themes in 2018

The ABC’s of Digital Disruption How Much Longer Can the
in Finance Cycle Run?
March 2018 January 2018
Securing India's Growth China Entering a New

Over the Next Decade Political Economy Cycle
Twin Pillars of Investment & The World According to Xi
Productivity Jinping Thought
February 2018 December 2017
2018 Corporate Finance Disruptive Innovations V

Priorities Ten More Things to Stop and
January 2018 Think About
November 2017
Women in the Economy II Technology at Work v3.0

How Implementing a Women’s Automating e-Commerce from
Economic Empowerment Click to Pick to Door
Agenda Can Shape the Global August 2017
Economy
November 2017
Inequality and Prosperity in Solutions for The Global
the Industrialized World Water Crisis
Addressing a Growing The End of ‘Free and Cheap’
Challenge Water
September 2017 April 2017
Education: Back to Basics Digital Disruption - Revisited

Is Education Fit for the Future What FinTech VC Investments
July 2017 Tells Us About a Changing
Industry
January 2017
ePrivacy & Data Protection 2017 Corporate Finance
Who Watches the Watchers? Priorities
– How Regulation Could Alter January 2017
the Path of Innovation
March 2017
2017 Investment Themes Infrastructure for Growth

A Wind of Change The dawn of a new multi-trillion
January 2017 dollar asset class
October 2016
Car of the Future v3.0 Re-Birth of Telecoms into a

Mobility 2030 New Digital Industry
November 2016 Time to Dump the Dumb Pipe
October 2016
Virtual & Augmented Reality Digital Disruption

Are you sure it isn’t real? How FinTech is Forcing
October 2016 Banking to a Tipping Point
March 2016
Disruptive Innovations IV Technology at Work v2.0

Ten More Things to Stop and The Future is Not What It Used
Think About To be
July 2016 January 2016
The Coming Pensions Crisis Investment Themes in 2016

Recommendations for Keeping New Normal or No Normal
the Global Pensions System January 2016
Afloat
March 2016
Global Political Risk Energy 2030

The New Convergence Financing A Greener Future
between Geopolitical and Vox November 2015
Populi Risks
January 2016
2016 Corporate Finance The Curtain Falls

Priorities How Silicon Valley is
January 2016 Challenging Hollywood
October 2015
The Global Art Market Disruptive Innovations III

Perspectives on Current Ten More Things to Stop and
Drivers & Future trends Think About
November 2015 July 2015
© 2018 Citigroup
© 2018 Citigroup
© 2018 Citigroup
IMPORTANT DISCLOSURES
This communication has been prepared by Citigroup Global Markets Inc. and is distributed by or through its locally authorised affiliates (collectively, the "Firm")
[E6GYB6412478]. This communication is not intended to constitute "research" as that term is defined by applicable regulations. Unless otherwise indicated, any reference to a
research report or research recommendation is not intended to represent the whole report and is not in itself considered a recommendation or research report. The views
expressed by each author herein are his/ her personal views and do not necessarily reflect the views of his/ her employer or any affiliated entity or the other authors, may differ
from the views of other personnel at such entities, and may change without notice.
You should assume the following: The Firm may be the issuer of, or may trade as principal in, the financial instruments referred to in this communication or other related
financial instruments. The author of this communication may have discussed the information contained herein with others within the Firm and the author and such other Firm
personnel may have already acted on the basis of this information (including by trading for the Firm's proprietary accounts or communicating the information contained herein to
other customers of the Firm). The Firm performs or seeks to perform investment banking and other services for the issuer of any such financial instruments. The Firm, the Firm's
personnel (including those with whom the author may have consulted in the preparation of this communication), and other customers of the Firm may be long or short the
financial instruments referred to herein, may have acquired such positions at prices and market conditions that are no longer available, and may have interests different or
adverse to your interests.
This communication is provided for information and discussion purposes only. It does not constitute an offer or solicitation to purchase or sell any financial instruments. The
information contained in this communication is based on generally available information and, although obtained from sources believed by the Firm to be reliable, its accuracy
and completeness is not guaranteed. Certain personnel or business areas of the Firm may have access to or have acquired material non-public information that may have an
impact (positive or negative) on the information contained herein, but that is not available to or known by the author of this communication.
The Firm shall have no liability to the user or to third parties, for the quality, accuracy, timeliness, continued availability or completeness of the data nor for any special, direct,
indirect, incidental or consequential loss or damage which may be sustained because of the use of the information in this communication or otherwise arising in connection with
this communication, provided that this exclusion of liability shall not exclude or limit any liability under any law or regulation applicable to the Firm that may not be excluded or
restricted.
The provision of information is not based on your individual circumstances and should not be relied upon as an assessment of suitability for you of a particular product or
transaction. Even if we possess information as to your objectives in relation to any transaction, series of transactions or trading strategy, this will not be deemed sufficient for
any assessment of suitability for you of any transaction, series of transactions or trading strategy.
The Firm is not acting as your advisor, fiduciary or agent and is not managing your account. The information herein does not constitute investment advice and the Firm makes
no recommendation as to the suitability of any of the products or transactions mentioned. Any trading or investment decisions you take are in reliance on your own analysis and
judgment and/or that of your advisors and not in reliance on us. Therefore, prior to entering into any transaction, you should determine, without reliance on the Firm, the
economic risks or merits, as well as the legal, tax and accounting characteristics and consequences of the transaction and that you are able to assume these risks.
Financial instruments denominated in a foreign currency are subject to exchange rate fluctuations, which may have an adverse effect on the price or value of an investment in
such products. Investments in financial instruments carry significant risk, including the possible loss of the principal amount invested. Investors should obtain advice from their
own tax, financial, legal and other advisors, and only make investment decisions on the basis of the investor's own objectives, experience and resources.
This communication is not intended to forecast or predict future events. Past performance is not a guarantee or indication of future results. Any prices provided herein (other
than those that are identified as being historical) are indicative only and do not represent firm quotes as to either price or size. You should contact your local representative
directly if you are interested in buying or selling any financial instrument, or pursuing any trading strategy, mentioned herein. No liability is accepted by the Firm for any loss
(whether direct, indirect or consequential) that may arise from any use of the information contained herein or derived herefrom.
Although the Firm is affiliated with Citibank, N.A. (together with its subsidiaries and branches worldwide, "Citibank"), you should be aware that none of the other financial
instruments mentioned in this communication (unless expressly stated otherwise) are (i) insured by the Federal Deposit Insurance Corporation or any other governmental
authority, or (ii) deposits or other obligations of, or guaranteed by, Citibank or any other insured depository institution. This communication contains data compilations, writings
and information that are proprietary to the Firm and protected under copyright and other intellectual property laws, and may not be redistributed or otherwise transmitted by you
to any other person for any purpose.
IRS Circular 230 Disclosure: Citi and its employees are not in the business of providing, and do not provide, tax or legal advice to any taxpayer outside of Citi. Any statements
in this Communication to tax matters were not intended or written to be used, and cannot be used or relied upon, by any taxpayer for the purpose of avoiding tax penalties. Any
such taxpayer should seek advice based on the taxpayer’s particular circumstances from an independent tax advisor.
© 2018 Citigroup Global Markets Inc. Member SIPC. All rights reserved. Citi and Citi and Arc Design are trademarks and service marks of Citigroup Inc. or its affiliates and are
used and registered throughout the world.
© 2018 Citigroup
NOW / NEXT
Key Insights regarding the future of Electric Vehicles
REGULATION In the near term, we expect EV demand to be largely propelled by a combination of

regulatory and government incentives to force auto manufacturers to comply with
emission targets. / Regulation to reduce CO emissions continue to evolve and will
2
be a driving force behind OEM R&D and capital decisions going forward.
INFRASTRUCTURE Charging infrastructure is consistently seen as a barrier to broader adoption of EVs

as the availability of public charge points for EVs still significantly lags. / Global
charging points are expected to rise to 13 million units in 2020 from just 2 million in
2020, a 51% CAGR.
TECHNOLOGY Battery degradation continues to weigh on consumer confidence while battery costs
continue to weigh on vehicle range ad total cost of ownership. / Battery costs are
continuing to fall and solid-state batteries are a leading candidate for mainstream
next-generation battery technology.
© 2018 Citigroup
Citi GPS: Global Perspectives & Solutions © 2018 Citigroup
www.citi.com/citigps

Electric Vehicles

Uploaded by

Copyright:

Available Formats

Electric Vehicles

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Electric Vehicles

Uploaded by

Copyright:

Available Formats

What are the main barriers currently preventing widespread adoption of electric vehicles?

What are the main barriers currently preventing widespread adoption of electric vehicles?

What factors are driving the increasing demand for electric vehicles?

What factors are driving the increasing demand for electric vehicles?

ELECTRIC VEHICLES

Ready(ing) for Adoption

Citi GPS: Global Perspectives & Solutions

Dr. Menahem Anderman PhD. Jeff Chung

Itay Michaeli Arifumi Yoshida

Contributors Kota Ezawa Charlie Grieg

...but battery 2017 2014

CURRENT DEMAND FOR EVs IS DRIVEN PRIMARILY BY ‘PUSH’

Source (Number of respondents)

2014 2015 2016 2017

25% 30m 27m

BEV ICE* Hydrogen Fuel Cell

*Based on latest Toyota 2.0 litre gasoline engine

Regulatory pressures are reducing EV’s cost disadvantage. Transportation is

Where Are We Now?

Source: Citi Research, Google Trends Source: Citi Research, LMC

Average R&D per unit (€)

What Does an EV Look Like Today?

Where Are We Going To?

acea LOW 2017

Roland Berger Scenario B 2016

Source: Published forecasts, Ricardo

What Will Drive Adoption?

Source: Citi Research, Autodata Source: Citi Research, Autodata

Mass Adoption Requires the Stars to Align

Source: Citi Research

A combination of carrots and sticks from local, national, and supranational

Figure 16. Summary of Regulatory Drivers for EV Adoption in Top 3 Markets

U.S. fuel-efficiency regulation is in a state of flux as the 2021/25 targets

The Supply Outlook for EVs

This leads us on nicely to what we dub as ‘The CFO’s Conundrum.’ What do we

The CFO’s Conundrum Matrix

We think ‘The CFO’s Conundrum’ is best illustrated by the following matrix:

Figure 17. The CFO’s Conundrum: Where Are We Most Comfortable?

High Follower Leader

Carmakers Will Dictate the Pace of Adoption

Who is Leading Today, and What Have the OEMs Announced?

Cost Parity Is a Critical Milestone for an Inflection of Supply

Battery size (kWh) - - 0.2 12.5 50

OEMs have been readying themselves for EV-adoption

Current Barriers to Mass Adoption

1. Range: “Will It Get Me There?”

* respondents from California * Rapid charging to 80%

Successive BEV generations have seen an improvement in range. While BEV

As battery costs come down. vehicle range is being extended. In Figure 28 we

Why Not Just Add Bigger Batteries?

Figure 30. Citi’s Battery Challenges Diagram

Source: Citi Research

Range (km) Tesla X 100D 210 Range (km) 300Ah 260Ah

320 BYD e6 130

240 Ford Focus 90

2. Infrastructure: “Where Do I Plug It In?”

Gas pumps per 100km of road

Outlook for Charging Points by Region

At the national level, there appears to be little correlation between charging