Economic,Trade and Employment Implications from EVs Deployment and Policies to Support Domestic Battery Manufacturing in the EU

Abstract

The decarbonization of the energy system requires the adoption of a mix of zero or low carbon intensive technological options, which depends on their cost-effectiveness, their potential to reduce emissions and on social acceptance issues. Transport electrification combined with renewable energy sources (RES) deployment in power generation is a key decarbonization option assessed in many recent studies that focus on national or international climate policies. The penetration of electric vehicles (EVs) together with a gradual retirement of conventional oil-fuelled vehicles implies that a new ‘trade ecosystem’ will be created characterized by different features (move from OPEX to CAPEX) and supply chains. A key component of the EVs are the Lithium-Ion batteries, the manufacturing of which is employment intensive and constitutes an essential element of the EVs that can act as a driver for establishing comparative advantages and increasing EV market shares. Our study focuses on the size of the EV market that can be established within ambitious global and EU decarbonization scenarios and investigates the economic, trade and employment implications considering the production chain of EVs (i.e., the regional production of batteries and vehicles). We use the large-scale global GEM-E3-FIT model to capture the trade dynamics of decarbonization scenarios. We find that under ambitious climate policies, the global size of the clean energy technologies will be US$44 trillion cumulatively over the 2020–2050 period. 44per cent of the market relates to EVs, which will mostly be produced outside EU. For the EU to capture a significant segment of the EV value chain, it needs to increase clean energy R&D and associated supportive policies so as to boost the domestic capacity to produce competitively batteries.

JEL: F11, F13, F16, F18, F62, F68

Keywords

GEM-E3-FIT clean energy trade battery manufacturing decarbonization

Introduction/Motivation

The decarbonization of the energy system, to achieve the Paris Agreement goals, requires the adoption of a mix of zero and low carbon intensive technological options. The deployment of clean energy technologies depends on their cost-effectiveness, their potential to reduce emissions and on social acceptance issues (Capros et al., 2014). Transport electrification combined with renewable energy sources (RES) deployment in power generation ranks high as an emission abatement option as identified in many recent studies that focus on national or international decarbonization strategies (including Fragkos, Tasios, Paroussos, Capros, & Tsani, 2017; Van Soest, Reis, Drouet, Van Vuuren, & Den Elzen, 2018). The rapid penetration of EVs in car markets coupled with a gradual retirement of conventional ICE vehicles implies that a new ‘trade ecosystem’ will be created characterized by different features (move from OPEX to CAPEX) and supply chains relative to the current paradigm. In particular, a key component of the electric vehicles (EVs) is the batteries that are quite employment intensive (IEA, 2018) and constitute an essential element of EVs that can act as a driver for establishing comparative advantages and increasing EVs market shares (Karkatsoulis, Capros, Fragkos, Paroussos, & Tsani, 2016). The objective of our study focuses on the size of the future global and EU specific market for EVs and batteries that can be established within global energy system decarbonization scenarios and investigates the economic, trade and employment implications, taking into account the production chain of EVs (i.e., the regional production of batteries). In particular, the study objective is to examine how the increased penetration of EVs would impact economic activity, trade patterns and employment in the presence of policy measures that would increase the EU share in the value chain of EVs in the low-carbon transition context.

With respect to batteries, decarbonization will lead to a substantial increase in EU demand, which, on current market and policy conditions, would largely be supplied from the Far East, as Japan, China and South Korea currently dominate the global market. The analysis considers two alternative scenarios with the aim to increase the share of EU-based suppliers in the battery market. The first scenario assumes that some form of EU regulation in battery imports leads to increasing share taken by EU battery suppliers, but at the cost of higher prices for batteries and thus EVs for customers as battery production cost in Far East is lower relative to the EU (CEMAC, 2016). The benefit, in terms of increased value-added and domestic jobs among battery producers in the EU, is estimated to be lower than the cost in terms of lost competitiveness for the producers of equipment (notably EVs) that incorporate batteries as a key component. In the second scenario, EU battery producers are assumed to receive support for R&D activities that leads to improved production cost and performance of EU batteries and makes them competitive with foreign suppliers. This scenario leads to increased production and employment in domestic EU battery manufacturing. The net benefits highly depend on the losses associated with the diversion of the support for R&D from other projects and on the extent to which the knowledge advantage gained from R&D can be prevented from spilling over quickly to foreign competitors: on the assumptions used in the scenario, the effect on EU GDP was marginally negative. However, the model-based analysis shows that if the EU wants to actively support domestic battery manufacturing, policies supporting R&D are preferable relative to the imposition of trade regulations.

The Policy Issue

The global market for clean energy technologies is highly competitive and fast growing driven by innovation dynamics, technological advancements and by the adoption of ambitious energy and climate policies and regulations. In 2015, the size of the global market is estimated to be about €250 billion and is dominated by the manufacture of solar PV panels and wind turbines (IEA, 2017). EVs registrations have followed an explosive path in recent years with China, the EU, Norway and the US being the key markets. In 2018, the global electric car fleet exceeded 5.1 million, up 2 million from the previous year and almost doubling the number of new electric car sales (IEA, 2019). Ambitious targets, fuel economy standards, policy support (i.e., fiscal incentives) and increased R&D investment have led to significant reductions in battery and EV costs, extended vehicle ranges and reduced consumer barriers in several countries. European electric car manufacturers are competitive in supplying EVs as they currently account for about 30 per cent of global production (IEA, 2019). The EU has the potential to remain an important player in the global electric car market both in manufacturing and in sales; this highly depends on the strengthening of demand-pull and supply-push policies, including emission standards and the establishment of ambitious policies to support the EU’s technology leadership in a highly competitive market. The lithium–ion (Li-ion) battery is the key component that will determine the development and eventual uptake of electric cars, as battery costs, availability and technical performance are key aspects for growth in e-mobility.

The world market for battery and EVs is growing rapidly. Currently, almost all Li-ion battery cells for EVs are produced by East Asian (Chinese, Japanese, and Korean) manufacturers. Meanwhile, the European automotive industry generates 4 per cent of European gross domestic product (GDP), and 12 million jobs (European Commission, 2018a). However, Europe has less than 1 per cent of the global Li-ion battery cell manufacturing capacity, and this production capability largely addresses niche markets (Beuse, Schmidt, & Wood, 2018). The limited EU production of batteries implies a continuing dependence on imported batteries that limits the creation of domestic employment that Europe can get from the deployment of EVs. In recent years, manufacturing of batteries for EVs lies at the centre of industrial and policy discussions at the European Commission (EC), with calls for ‘European sovereignty’ in Li-ion battery manufacturing (European Commission, 2018b).

Currently, battery amounts to more than 40 per cent of the full cost of an electric car and the cell makes up about 70 per cent of the cost of an automotive battery (IEA, 2018). East Asian manufacturers have achieved continuous improvements in the cost and performance of batteries induced by national industrial policies and the high demand for Li-ion batteries in consumer electronics (cell phones, tablets and laptops). Recently, several East Asian producers announced the construction of battery cell manufacturing Gigafactories in Europe with an output capacity higher than 1 GWh/year; for example, LG Chem in Poland, CATL in Germany, Samsung and SK Innovation in Hungary. In parallel, several policymakers in Europe have recognized the importance of developing the battery industry to ensure Europe’s competitiveness in the rapidly changing automotive sector. They have called for exclusively European-owned cell factories.¹

To support these ambitions, the EC has developed a Strategic Action Plan together with representatives from industry and academia, outlining potential actions and funding opportunities (potentially up to several billion €). In line with these ambitions, three consortia (led by Saft, TerraE, and Northvolt) have formed to build European-owned Gigafactories, supported by the European Investment Bank. However, the EU automotive industry (despite investing in battery R&D activities) has not invested in these consortia and has not yet indicated if and when they want to build large-scale battery manufacturing plants. Some automotive companies consider battery cells a commodity component, sourced from suppliers according to specification, and see added value primarily in downstream activities of the supply chain. Other companies consider ownership in cell manufacturing worthwhile; however, they have not yet entered the arena and discuss leapfrogging to ‘next generation’ technologies, such as solid-state batteries (Beuse et al, 2018).

Recent benchmark assessments by the IEA² show that the global EV market will expand considerably with annual EV sales projected to reach 43 million and EV stock exceeding 250 million by 2030, with further upscale beyond 2030. These developments are driven by targeted policies (i.e., CO₂ or fuel economy standards, subsidies), increased R&D expenditure, technology advances and expansion of EV and battery manufacturing capacity. The recent EC impact assessment for the ‘Clean Planet for all’ strategy³ confirms the very high importance of the massive upscale of EVs and batteries in the next decades to achieve ambitious targets towards climate neutrality. It shows that in ambitious decarbonization scenarios, the share of EVs in the EU car stock ranges between 60–75 per cent in 2050, with a projected EV stock of 200–250 million cars, thus indicating the large market potential for EVs and battery cells both at the EU and global level.

Given the current state of affairs and the projected uptake of EVs in the context of low-carbon transition, it is important for EU policy-makers, the automotive and the battery production industry to understand the economic, trade and employment impacts of developing battery manufacturing activities. The purpose of this article is to explore the trade, industrial and socio-economic impacts of the upscale of EV and battery manufacturing activities by 2030 and 2050 examining the EU in the global context. The article also examines the effectiveness of various policy measures to support domestic battery manufacturing and jobs, without adverse side impacts in other sectors of the economy. Apart from a cost-optimal decarbonization scenario, used as a benchmark for scenario comparison, the article presents three alternative policy-driven scenarios, which vary regarding the assumed policy frameworks used to support the domestic battery manufacturing in the EU. To perform rigorous quantitative assessments for the macroeconomic impacts of alternative policies, we use the well-established GEM-E3-FIT model, an advanced computable general equilibrium (CGE) model covering the whole world disaggregated into 46 countries/regions and 51 types of activity with endogenously derived bilateral trade flows.⁴

The remainder of this article is organized as follows. In the second section, the article provides a short review of the literature exploring the economic and trade impacts of EVs and battery upscale. The third section presents the GEM-E3-FIT model, focusing on the features that are relevant for the specific study. The fourth section describes the alternative policy-driven scenarios, while the fifth section discusses the results of the scenario projections. The sixth section draws concluding remarks and policy recommendations.

Literature Review

The EV industry is composed of a large number of firms dispersed worldwide, but the production of EVs and specifically of batteries takes place only in few countries. In 2015, it was estimated that the market for EV batteries is around €6bn and is heavily concentrated in the Far East, with Japan, S. Korea and China accounting for 95 per cent of the global market.⁵

China, Europe and the USA are the leading global markets in terms of EV sales and production. However, the picture in terms of battery packs manufacturing is different, as Asian producers (China, Japan and S. Korea) dominate the global market and export to EU countries and the USA. The EU electric car manufacturers outsource (import) 97 per cent of their Li-on battery requirements (Figure 1). Unlike conventional vehicles where the competition among manufacturers largely depends on the techno-economic features of their engines, the EV manufacturers compete based on the cost and performance of batteries. Recent announcements by car manufacturers and the EU point towards an increase in battery manufacturing activities in the EU. Volkswagen, as part of its new 2025 Strategy, has outlined plans for a €10bn battery factory in Salzgitter in Germany; Samsung and LG Chem plan to invest in EV battery factories in Hungary and Poland respectively to exploit the rapidly growing EU demand; Ford, BMW and Tesla are also considering building battery factories in Europe. In October 2017, the European Commission launched⁶ the European Battery Alliance with the target of developing battery manufacturing domestically in the EU. The EU industry and innovation community will drive this process, working in close partnership with the European Commission, the European Investment Bank and interested Member States, aiming to establish a competitive EV and battery manufacturing chain, capture sizeable market prospects and boost jobs, growth and investment across Europe.

Figure 1.

Source: Lutsey, Grant, Wappelhorst, and Zhou (2018).

The rapid expansion of EVs is also driven by energy and climate policies like the recent CO₂ standards imposed on car manufacturers and the ambitious decarbonization targets of the EU and its Member States. Additional elements that drive the deployment of EVs relate to unilateral tariff and non-tariff barriers (ICTSD, 2017) but also on restrictions regarding the availability of critical raw materials (Olivetti, Ceder, Gaustad, & Fu, 2017).

A study by (De Cian et al., 2013) found that the EU would bear costs in case of unilateral climate action, but renewable technology costs would decline faster due to increased R&D and innovation. Alexandri et al. (2018) explored the impacts of early EU climate action and found that there is a modest potential for the EU to establish a first mover advantage and increased competitiveness in EV manufacturing driven by ambitious climate policies. The first mover advantage highly depends on whether batteries are produced internally in the EU and whether EU battery manufacturers achieve cost parity to Asian producers in a timely manner by 2030.

Battery production costs in leading Asian manufacturers are currently around 250 €/kWh. Costs have fallen substantially over the last decade (a 14% annual reduction between 2007 and 2016, from more than US$1,000 per kWh to around US$273 per kWh). However, the production cost for US and EU battery manufacturers is currently 10–30 per cent higher relative to Far East producers (Donald Chung et al., 2015).

Figure 2.

Source: Curry (2017).

The cost of battery packs needs to fall to ‘below US$150 per kWh in order for BEVs to become cost-competitive on par with internal combustion vehicles’, as confirmed by Björn and Nilsson (2015). The key drivers for the steep cost reductions in battery manufacturing are:

Economies of scale and learning by doing effects (Global sales of EVs and batteries have significantly expanded and the associated learning rate is estimated to be 16–17%)

Research and innovation, as in recent years large developed and developing economies direct a substantial amount of R&D to EV batteries

Figure 2 presents the reduction in Li-ion battery prices from 2010 to 2016 as a result of both learning by doing and innovation.

Methodology

Model

General Features of the GEM-E3-FIT Model

The GEM-E3-FIT model (developed and operated by E3Modelling) is a multi-regional, multi-sectoral, recursive dynamic CGE model, which provides details on the macro-economy and its interactions with the environment and the energy system. GEM-E3-FIT simultaneously represents 46 regions (all EU28 member states are represented separately) and 51 sectors linked through endogenous bilateral trade flows and runs until 2050 with a 5-year time step. It is a comprehensive model of the global economy, covering the complex interlinkages between productive sectors, consumption, price formation of commodities, labour and capital, bilateral trade and investment dynamics. The model is dynamic, recursive over time, driven by accumulation of capital and equipment. The model features alternative market regimes, equilibrium unemployment, energy efficiency standards, carbon pricing and emission permits for Greenhouse Gas (GHG) emissions and can quantify the macro-economic, employment and distributional impacts of policies, both in the short and long-term. GEM-E3-FIT allows for a consistent comparative analysis of policy scenarios since it ensures that in all scenarios, the economic system remains in general equilibrium. In addition, it incorporates micro-economic mechanisms and institutional features within a consistent macro-economic framework. GEM-E3-FIT is calibrated to three base years (2004, 2007 and 2011) using the GTAP 9 dataset, whereas a new advanced calibration methodology is applied that ensures maximum consistency between input output tables, energy volumes and GHG emissions.

Industries operate within a perfect competition market regime and maximize profits. Production functions consider the possibilities of substitution between capital, labour, energy and materials in each production sector. The model identifies one representative firm for each sector. Households demand, savings and labour supply are derived from utility maximization using a linear expenditure system (LES) formulation, assuming exogenous population. Households receive income from labour supply and from holding shares in companies. Investment by sector is dynamic depending on adaptive anticipation of capital return and activity growth by sector.

The GEM-E3-FIT model includes several features that go beyond a classical/conventional CGE approach, allowing for an improved representation of the impacts of policies on the economy and the society (Figure 3). In this respect, the model incorporates: a detailed and explicit representation of the financial sector (and its linkages to low-carbon investment), endogenous growth through R&D, detailed modelling of the energy system and related technologies (especially in power supply, transport and buildings) and disaggregated representation of employment by skill. The GEM-E3-FIT model is extensively used as a tool for policy analysis and impact assessments, especially in the energy and climate policy fields by the European Commission and national governments (Capros et al., 2016; Fragkos et al., 2017; Alexandri et al., 2018).

Representation of Trade in GEM-E3-FIT

All regions and sectors are linked through endogenous bilateral trade flows. Total demand (final and intermediate) in each country is optimally allocated between domestic and imported goods, under the hypothesis that they are imperfect substitutes (Armington assumption) (Armington, 1969). The supply mix is represented as a multi-level nested constant elasticity of substitution dual cost function: at the upper level, firms decide on the optimal mix between domestically produced and imported goods; at the next level, the demand for imports is split by country of origin.

Figure 3.

Source: Capros et al. (2017).

The optimal demand for domestic {1} and imported goods is obtained by the cost minimization problem of purchasing the composite good (1st level):

Q D_{j, r, t} = Q Y_{j, r, t} . A C_{j, r, t}^{σ x_{j, r, t} - 1} {(1 - δ_{j, r, t})}^{σ x_{j, r, t}} . {(\frac{P Y_{j, r, t}}{P D_{j, r, t}})}^{σ_{x_{j, r, t}}}

(1)

Q I_{j, r, t} = Q Y_{j, r, t} . A C_{j, r, t}^{σ x_{j, r, t} - 1} . δ_{j, r, t}^{σ x_{j, r, t}} . {(\frac{P Y_{j, r, t}}{P I_{j, r, t}})}^{σ x_{j, r, t}}

(2)

where

j: sectors, r: countries, t: time,

QY_j,r,t: composite goods volume index, PY_j,r,t: composite goods price index,

QD_j,r,t: domestic goods volume index, PD_j,r,t: domestic goods price index,

QI_j,r,t: imported goods volume index, PI_j,r,t: imported goods price index,

AC_j,r,t: scale parameter in the Armington function,

δ_j,r,t: share parameter estimated from the base year data related with the value shares of QD_j,r,t and QI_j,r,t in the demand for composite good QY_j,r,t,

σx_j,r,t: the Armington elasticity between imported and domestically produced goods.

At the 2nd level, the buyer seeks to minimize the cost of imported goods by choosing the optimal mix of imports by origin (Paroussos et al., 2015):

Q I M_{j, r, s, t =} Q I_{j, r, t} . {(β_{j, r, s, t} . \frac{P I_{j, r, t}}{P I M_{j, r, s, t}})}^{σ m_{j, r, t}}

(3)

where

QIM_j,r,s,t: imported goods of country r from the country s volume index,

PIM_j,r,s,t: imported goods of country r from the country s price index,

β_j,r,s,t: share parameter estimated from the base year data related with the value shares of imported goods by origin.

σm_j,r,t:the Armington elasticity among imported goods by origin.

Table 1 contains the upper-level Armington elasticity values used in GEM-E3-FIT. The Armington elasticities differ among sectors, but are identical for all countries/regions. Homogeneous products, like the crude oil, are assumed to have higher elasticity values.

Endogenous Representation of Clean Energy Markets

GEM-E3-FIT includes the manufacturing of clean energy products and equipment as separate production sectors in order to consistently derive the evolution of their manufacturing and trade under alternative policy assumptions. The model database has been extended to allow for a separate representation of the clean energy producers, namely for solar PV, wind turbines, EVs, Li-Ion batteries and biofuels, in economic terms. Current market shares and bilateral trade flows have been consistently incorporated into the model building on various data sources, including ISE (2018), Navigant (2017), CEMAC (2016) and transport and environment (2017). This feature is extremely important to capture growth effects driven by clean energy industries and innovation, competitiveness, industrial impacts and changes in trade flows induced by ambitious decarbonization measures and/or clean energy innovation policies.

Table 1.

Armington Elasticities

	σx	σm		σx	σm
Transport	1.90	3.80	Coal	3.05	6.10
Construction	1.90	3.80	Consumer goods	3.21	6.43
Non-market services	1.90	3.80	Chemical products	3.30	6.60
Market services	2.03	4.06	Transport equipment	3.55	7.10
Oil products	2.10	4.20	Electric vehicles	3.55	7.10
Gas	2.80	5.60	Non-ferrous metals	3.98	7.95
Power supply	2.80	5.60	Equipment goods	4.05	8.10
Ferrous metals	2.95	5.90	Batteries	4.05	8.10
Paper products	2.95	5.90	Electronic goods	4.08	8.15
Agriculture	3.03	6.07	Crude oil	5.20	10.40

Source: GTAP 10 Data Base Documentation. https://www.gtap.agecon.purdue.edu/resources/download/4184.pdf

As the GTAP database does not include the manufacturing of EVs and batteries as separate sectors, supplementary data sources have been used to provide reasonable estimates for the size, structure and trade transactions of these sectors. First, the demand (sales) and manufacturing volumes of EVs for each country has been derived using the (IEA, 2018) data and the ‘Electric cars’ report (Transport and Environment, 2016) respectively. The manufacturing volumes of batteries by country are based on (IEA, 2018). The cost structure assumed at the base year for EVs and batteries are derived from (Fragkiadakis, Paroussos, & Fragkos, 2019). Combining these costs with the manufacturing volumes, the production and net imports of EVs and batteries are estimated for each country (in economic terms). Then, we used the study by Fries et al. (2017)⁷ to determine the inputs required from other sectors (i.e., equipment, metals, plastics, etc.) to produce EVs and batteries, which are different relative to the cost structure of conventional cars. To determine the bilateral trade in these sectors, we developed a RAS routine with weighting derived from the GTAP sector 43 ‘Manufacture of motor vehicles, trailers and semi-trailers’. Finally, a process was developed to ensure that the production of conventional and EVs sums up to the GTAP sector 43. All data sources related to EVs have been consistently integrated in the GEM-E3-FIT framework to produce balanced Input–Output tables for all countries that incorporate explicitly the manufacturing and trade of batteries and EVs.

The implementation of ambitious energy and climate policies in GEM-E3-FIT would lead to emission reductions, which are mostly driven by improvements in energy efficiency, expansion of RES and electrification. The GEM-E3-FIT model explicitly includes several emission reduction options (details can be found in model manual (Capros et al., 2017)), that is:

a variety of RES technologies to produce electricity (wind, solar PV, hydro, biomass);

increased uptake of EVs in the road transport sector and in heating uses (electrification);

technologies to capture and store carbon dioxide emitted from power plants;

investment in energy savings in buildings, transport and industries;

fuel substitution away from carbon-intensive fuels (oil, coal) and towards natural gas and/or clean energy forms (electricity, biomass, RES);

increased uptake of biofuels in the transport sector; and

uptake of advanced, energy efficient equipment in buildings (e.g., for electric and heating appliances).

In GEM-E3-FIT, the future deployment of EVs strongly depends on the availability of other emission abatement options, that is, competition between biofuels and EVs to decarbonize the transport sector. The uptake of EVs depends also on the inter-linkages with the power generation sector in particular with regard to the evolution of electricity price that is determined by RES uptake and technology progress. The GEM-E3-FIT model can capture all these interlinkages between the alternative emission reduction options and the complex dynamics of the EV uptake combined with RES expansion, electrification and efficiency improvements.

Representation of R&D and Technology Progress in GEM-E3-FIT

Technology progress is explicitly represented in GEM-E3-FIT depending on R&D expenditure by private and public sector and spillover effects. The technical progress is represented through two-factor learning curves (learning by doing and learning by research). The learning by doing component corresponds to the productivity gained through cumulative production (i.e., learning from experience and economies of scale). The learning by doing effect is introduced only in the new clean energy technologies (i.e., PV, Wind, electric cars, batteries and biofuels). Learning rates for clean energy technologies are taken from the upper and lower bounds available in literature (Paroussos et al., 2019). The R&D learning rate indicates the reduction in unit costs of energy technologies for each doubling of the cumulative R&D expenditure, while learning by doing rate indicates the reduction in unit costs for each doubling of the cumulative capacity of technologies.⁸

GEM-E3-FIT incorporates a semi-endogenous representation of the R&D sector separating public from private R&D expenditures. Each firm decides upon the optimal R&D spending so as to maximize its profits, whereas public R&D is set exogenously. R&D expenditures generate a stock of knowledge that in turn is linked to productivity growth. The link between R&D expenditure and productivity is provided in the following equations, where $R D_{j}^{\Pr i v a t e}$ is the optimal demand of the firms for R&D, $θ_{j}^{r d}$ is the value share of R&D expenditures in total production costs, Q_j represents the total sales of the firm, PQ_j is selling price and PRD is the unit cost of R&D.

R D_{j}^{\Pr i v a t e} = θ_{j}^{r d} . Q_{j} . {(\frac{P Q_{j}}{P R D_{j}})}^{r h o}

(4)

R D_{j}^{P u b l i c} = (E x o g e n o u s)

(5)

R D_{j}^{T o t a l} = R D_{j}^{\Pr i v a t e} + R D_{j}^{P u b l i c}

(6)

Learning by doing and learning by research increase total factor productivity in the clean energy producing industrial sectors as presented in Equation (7), where Z_j,t is the production in the case of learning by doing and the R&D expenditures of the firms in the case of learning by research and bb_tec the corresponding learning by doing or learning by research rate of the technology tec that is linked one-to-one with the firm j.

T F P_{j} = {(\frac{{\sum^{​}}_{t} Z_{j, t}}{{\bar{Z}}_{j}})}^{- b b_{t e c}}

(7)

Productivity that is generated through R&D is diffused into other sectors and countries according to a patent citation⁹ matrix approach. The productivity spillover to other sectors and countries is calculated as in Equation (8).

T F P_S P I L L_{i, j, r, s} = T F P_{i} . s p i l l o v e r_{i, j, r, s}

(8)

Additional details on the assumed learning rates and the representation of technical progress in GEM-E3-FIT can be found in Paroussos et al. (2019).

Scenario Design

The scenario design is based on the implementation of various exogenous assumptions with regard to GDP growth, population, technology costs and climate policies; Alexandri et al. (2018) includes details on these assumptions and their evolution until 2050 with main data sources including EC, reference scenario (2016), International Energy Agency (2017), World Energy Outlook (2017) and EUCO scenarios (2017).¹⁰ The GEM-E3-FIT model has been used to quantify the macroeconomic implications of a future in which the production of batteries for EU-manufactured EVs are manufactured in the EU rather than imported. To this end, three core policy scenarios have been designed:

C-REG-Battery: A regulation-based scenario in which imports of batteries from non-EU countries are not allowed after 2025. In this scenario, 100 per cent of EU demand for batteries is produced in the EU irrespective of their cost compared with imports. Hence, in this scenario, EU battery production is supported by regulations and standards and is not price competitive.

C-R&D-Battery-ETS: An R&D based scenario in which part of the ETS auction revenues is directed to R&D for batteries until EU-produced batteries are competitive in international markets, that is, EU production costs reach those of Asian leading manufacturers. This scenario is based on the suggestion of Lebedeva et al. JRC (2016)¹¹ that ‘To become competitive in this time frame however, consolidated R&I action at a European level is required in the short term to secure Europe’s strong industrial position in pack manufacturing in the future’.

C-R&D-Battery-RES: An R&D based scenario in which EU-28 public R&D is redirected from other clean energy technologies to batteries until European production becomes competitive with leading battery manufacturers in international markets.

All scenarios assume that the EU and the world adopt ambitious climate policies with the target to limit average global warming to CO₂ relative to pre-industrial levels.¹² The DECARB scenario represents the cost-optimal global mitigation pathway, as a uniform carbon price is applied across regions and sectors to meet the global carbon budget constraint by 2050. The same climate policy intensity (i.e., same carbon price) is imposed to EU and non-EU-regions. No explicit policies to limit battery imports to the EU are imposed in the DECARB scenario.

Model-based Results

In the global decarbonization context, the world market for clean energy technologies is projected to amount to €43.6 trillion in cumulative terms over 2020–2050. The manufacture of electric cars accounts for 44 per cent of the global clean energy market. The EU is projected to account for about 18 per cent of the global clean energy market, but has very small shares in the production of batteries and PV modules, reflecting the current low competitiveness of EU-based producers. On the other hand, EU manufacturers are projected to remain competitive in the global market for electric cars (27% share in global production) and wind turbines (20% share in global production). The size of the market for clean energy technologies depends greatly on climate, energy, industrial, trade and innovation policies. Figure 4 shows the size of the global market for batteries for EVs in the Business-As-Usual and DECARB scenarios.

Figure 4.

Source: GEM-E3-FIT.

In the decarbonization context, the global battery market reaches almost €300bn in 2050 (the respective figure for EVs is €1.8 trillion). As the DECARB scenario does not assume explicit policies to support EU domestic battery production, EV manufacturers continue to outsource the batteries mostly from China, Japan and South Korea; the average share of the EU in global battery production is projected to remain close to current levels of about 1 per cent over 2020–2050. Thus, a substantial part of the value-added and employment associated with the production of electric cars is not located in Europe.

C-REG-Battery Scenario

In the scenario assuming that battery imports in the EU are not allowed by regulation, the EU GDP in 2050 is reduced by 0.19 per cent compared with the DECARB. The learning and economies of scale effects are not sufficient to make the European battery production competitive in international markets and thus the production cost of EU-manufactured EVs increases and sales decrease compared with DECARB. The EU is assumed to produce batteries at a price 30 per cent higher than its main supplier in 2020,¹³ falling to 10 per cent in 2050 due to economies of scale induced by increased domestic battery production relative to the DECARB scenario. In the C-REG-BATTERY scenario, EU battery manufacturing is higher than in DECARB, as battery imports from non-EU countries are not allowed. This induces increased learning by doing and economies of scale effects, which drive down the costs of EU-produced batteries, but is not sufficient to lead to price parity with leading Asian manufacturers (Table 2).

Table 2.

Macroeconomic and Employment Implications of C-REG-BATTERY Scenario

				2020	2030	2040	2050
GEM-E3-FIT
	GDP		2020=100
		REF		100.0	114.7	133.6	154.8
		DECARB		100.0	115.1	133.6	154.0
		C-REG-BATTERY		100.0	115.0	133.5	153.7
		Difference from DECARB	%	0.0	–0.03	–0.10	–0.19
	Employment		millions
		REF		219.0	216.4	212.9	208.4
		DECARB		219.1	217.0	212.7	208.6
		C-REG-BATTERY		219.1	217.0	212.6	208.4
		Difference from DECARB	%	0.0	–0.01	–0.04	–0.06

Source: GEM-E3-FIT.

The EU benefits from lower battery imports (relative to the DECARB scenario) but the loss of EV sales is dominant in the EU’s trade balance: in 2050 EU imports of batteries are reduced by €58bn, but exports of EVs are reduced by €148bn driven by increased EV production costs and reduced competitiveness of the EU in the international markets (Table 3).

Table 3.

Trade Implications of C-REG-BATTERY Scenario

Absolute change from DECARB (in €bn)	2025	2030	2040	2050
Exports of electric vehicles	–0.8	–8.1	–52.6	–148.7
Imports of batteries	–3.4	–7.6	–32.0	–58.3
Net benefit/loss for the EU	2.6	–0.4	–20.6	–90.4

Source: GEM-E3-FIT.

In terms of employment, EU production of batteries increases the labour intensity of the economy. It is estimated that for each additional €1m of value of batteries produced in the EU, three full-time equivalent jobs are generated. If the income effect of the scenario is not taken into account (i.e., if the EU economy had the same GDP as in DECARB scenario), then employment would increase by 140,000 jobs in 2050. However, lower sales and production of the EV industry drive GDP reductions and hence employment is 0.06 per cent lower than in DECARB in 2050.

C-R&D-Battery-ETS

In this scenario, it is assumed that the EU increases its public R&D expenditure so as to reduce the production costs of EU-manufactured batteries through increased innovation and learning by research. As the relationship between R&D expenditure and total factor productivity (TFP) improvements is highly uncertain (Diaz Anadon, Baker, Bosetti, & Reis, 2016), three variants have been developed. In the variants, the relationship between R&D expenditure and TFP improvements is varied (i.e., different learning rates used). Table 4 presents the main assumptions for the three variants. Learning rates are taken from the upper and lower bounds available in literature. The R&D learning rate indicates the reduction in unit costs for each doubling of the cumulative R&D expenditure.

In the efficient variant, R&D expenditure for batteries is calculated so that the EU produces batteries at a competitive price (relative to leading Asian manufacturers) by 2035. In the unsuccessful R&D scenario, the learning rate is set at the lower bound and the EU only achieves competitive battery pricing in 2050. In the ambitious R&D variant, the EU achieves price parity in 2035 and produces batteries at a lower cost than its competitors thereafter. In the efficient scenario, it has been estimated that €26bn would need to be spent on battery R&D over the 2020–2050 period. The resulting spending and price parity are presented in Figure 5.

In order to simulate a budget neutral policy, it is assumed that a part of ETS revenues is directed to R&D for batteries. A second variant regarding the financing of R&D expenditures is examined that assumes that public R&D is redirected from other clean energy technologies to batteries (this variant has been developed only for the efficient R&D scenario). It has been assumed that any patents on batteries that are generated in the EU do not spillover to non-EU countries in the first 5 years.

Table 4.

Variants of the C-R&D-BATTERY-ETS Scenario

Scenario Name	Cumulative R&D Expenditure (2020–2050)	Average Annual Expenditure	TFP Improvement	Learning Rate*
Efficient R&D	€26 billion	€860 million	3.5	24%
Unsuccessful R&D	€45 billion	€1.5 billion	2.8	16%
Ambitious R&D	€160 billion	€5.3 billion	4	16%

Source: GEM-E3-FIT.

Note: * Learning rates are taken from the upper and lower bounds available in literature. The R&D learning rate indicates the reduction in unit costs for each doubling of the cumulative R&D expenditure.

Figure 5.

Source: GEM-E3-FIT.

The net impact of the ETS-financed R&D variants on EU GDP and employment is slightly negative when compared with the DECARB scenario, but positive when compared to the C-REG-BATTERY scenario. The C-R&D-BATTERY-RES scenario has a slight negative impact on EU GDP and employment relative to the ETS-financed R&D cases (other than the unsuccessful R&D case), due to the reduction of R&D expenditures in other clean energy technologies (Table 5).

Table 5.

Macroeconomic and Employment Implications of Battery R&D Scenarios

		2020	2030	2040	2050
GEM-E3-FIT
GDP	2020=100
REF		100.0	114.7	133.6	154.8
DECARB		100.0	115.1	133.6	154.0
C-R&D-BATTERY-RES		100.0	115.1	133.5	153.9
Efficient R&D-ETS		100.0	115.1	133.5	153.9
Unsuccessful R&D - ETS		100.0	115.1	133.5	153.8
Ambitious R&D - ETS		100.0	115.1	133.6	154.1
Difference from 2DEG	%
C-R&D-BATTERY-RES		0.00	–0.02	–0.05	–0.05
Efficient R&D-ETS		0.00	–0.02	–0.05	–0.05
Unsuccessful R&D-ETS		0.00	–0.02	–0.07	–0.13

Source: GEM-E3-FIT.

Table 6.

Trade Implications of C-R&D-BATTERY Scenario

Absolute change from DECARB (in €bn)	2025	2030	2035	2040	2045	2050
C-R&D-BATTERY-RES
Exports of electric vehicles	–0.5	–4.1	–10.8	–24.4	–57.3	–65.2
Imports of batteries	–3.4	–7.6	–14.9	–31.9	–55.5	–58.1
Net benefit for the EU	2.9	3.5	4.1	7.5	–1.8	–7.1
% of EU GDP	0.01	0.01	0.02	0.02	–0.01	–0.02
Unsuccessful R&D-ETS
Exports of electric vehicles	–0.7	–6.7	–17.9	–41.1	–96.4	–108.0
Imports of batteries	–3.5	–7.4	–13.6	–28.0	–46.3	–46.0
Net benefit for the EU	2.8	0.7	–4.3	–13.1	–50.1	–62.0
% of EU GDP	0.01	0.00	–0.02	–0.04	–0.15	–0.17

Source: GEM-E3-FIT.

As other clean energy technologies, mainly solar PV and wind, have lower potential for cost reductions through learning and innovation, the reallocation of R&D funds to battery production offers higher multiplier effects in the EU economy. However, the time needed for the EU to achieve price parity with its international competitors is crucial for the derivation of positive outcomes on economic growth and job creation.

The GEM-E3-FIT results (Table 6) show that in the short-term, the benefit from reduced battery imports is larger than the loss from lower exports of EVs, leading to a net trade benefit for the EU. However, in the long-term, where the global market of electric cars becomes larger, the losses from reduced exports of electric cars are not compensated by reduced battery imports. The proportion of the battery cost to the total cost of EVs is projected to decline from 35 per cent in 2020 to 20 per cent in 2050 as battery costs drop faster relative to the total cost of EVs. This means that (in terms of quantity) a larger reduction in battery imports is required to compensate for lower exports of electric cars in 2050 compared to 2020. Thus, the EU balance of trade deteriorates slightly after 2040 relative to the DECARB scenario.

Conclusions and Policy Recommendations

Currently, the EU has a very small share in the global battery production, as most European EV manufacturers outsource the required batteries from leading Far East producers (China, Japan, South Korea). Following current trends in battery market and assuming limited policies targeting domestic battery manufacturing, the EU share in global battery manufacturing is estimated to remain at about 1 per cent over 2020–2050. This means that through to 2050 the EU electric car manufacturers will still outsource the automotive batteries, mostly from the Far East. In order for the EU to become a competitive battery supplier, extensive public R&I is required to support private firms in the production of batteries and in the development of domestic value chains; alternatively, economic incentives to private firms can take the form of large tax exemptions/subsidies.

The economic, employment and trade impacts, expected to arise from a policy supporting the domestic EU battery manufacturing, highly depend on the type of policy measures implemented. On the one hand, the imposition of regulations and standards that do not allow battery imports to the EU leads the EU electric car producers to purchase higher-cost batteries from EU producers, thus increasing the production cost of EVs. On the other hand, R&D support to domestic battery manufacturing may lower the costs of EU producers to match those of international competitors. Both cases bring benefits in the form of reduced imports of batteries and reduced dependence on non-EU battery manufacturers located in China, Japan and South Korea. However, in the regulation case, EU electric car producers face higher prices for batteries and hence increased production costs that reduce their competitiveness in international markets. The benefit, in terms of increased value added and jobs among EU battery producers is found to be lower than the cost in terms of lost competitiveness for the producers of equipment (notably EVs) that incorporate batteries as a key component. In case that EU producers are assumed to receive support for R&D that makes them competitive with foreign suppliers, the net benefits depend on the losses associated with the diversion of the support for R&D from other clean energy projects and on the extent to which the knowledge advantage gained from R&D can be prevented from spilling over quickly to foreign competitors: on the assumptions used in the scenario, the effect of the scenario assuming increased R&D support to batteries on EU GDP and employment was positive relative to the regulation scenario.

The analysis estimates the potential impact of measures to increase the EU share in the battery value chain, that is, through a ‘battery initiative’ implemented either through restricting imports by means of regulation and standards or by supporting European R&D for battery production. Currently, the EU has a very small share in the battery value chain and so the scale of public R&D support to assure the production of competitively priced batteries is likely to be substantial. In the modelling analysis, it was assumed that R&D support was redirected from other kinds of clean energy R&D support (to assure budget neutrality). The relocation of battery manufacture to the EU brings benefits in the form of reduced imports of batteries and reduced dependence on non-EU manufacturers. If the initiative is implemented by restricting battery imports, the consequence is higher production costs for the EU EV industry, and the loss of EV export revenues outweighs the reduction in the value of battery imports. The analysis shows that the implementation of well-planned innovation policies targeting battery manufacturing would lead to the development of a low-cost battery manufacturing capability in the EU boosting domestic activity both through the production of batteries and through higher employment and the associated impact on household income.

The study points to the need of a technology-smart policy strategy, considering battery cells’ complexity, based on insights from battery research and innovation. The incentivization of R&D directed to companies along the battery supply chain is crucial to ensure that EU-based manufacturing becomes competitive in international markets. Collaborative innovation including both EU and non-EU companies should be strengthened and be backed up with measures to educate engineers and researchers and to secure access to raw materials and low-cost capital. In addition, a clear, consistent and ambitious EV strategy is needed to create an attractive European EV market and provide well-anticipated price signals and planning security to European investors, industries and innovators. This strategy should integrate carefully both demand (i.e., though ambitious emission standards, EV subsidies or phase-out mandates of oil-fuelled ICE cars) and manufacturing aspects—largely focused on R&D and low-cost capital provision—for the EV and battery market.

Footnotes

Acknowledgements

We acknowledge funding for this research by the European Commission, Directorate-General for Energy, contract no. ENER/A4/2015–436/SER/S12.716128 and by European Union’s Horizon 2020 research and innovation programme under grant agreement No 727114 (‘MONROE^b’ project). The information and views set out in this article are those of the author(s) and do not reflect the official opinion of the Commission.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

Notes

References

Armington

P. S.

(1969). A theory of demand for products distinguished by place of production. IMF Staff Papers, 16(1), 159–178.

Alexandri

, Fragkiadakis

, Fragkos

, . (2018). A technical analysis on decarbonization scenarios - constraints, economic implications and policies. Technical report on the Macroeconomics of Climate and Energy Policies, European Commission, DG ENER. Retrieved from https://ec.europa.eu/energy/sites/ener/files/documents/technical_analysis_decarbonisation_scenarios.pdf

Beuse

, Schmidt

T. S.

, Wood

(2018, September 14). A ‘technology-smart’ battery policy strategy for Europe, Science, 361(6407), 1075–1077. doi: 10.1126/science.aau2516.

Capros

, Paroussos

, Fragkos

, Tsani

, Boitier

, Wagner

, …, & Bollen

(2014). European decarbonisation pathways under alternative technological and policy choices: A multi-model analysis. Energy Strategy Review 2(3–4), 231–245.

Capros

, De Vita

, Tasios

, . (2016). EU reference scenario 2016 – energy, transport and GHG emissions Trends to 2050. European Commission Directorate – General for Energy, Directorate – General for Climate Action and Directorate – General for Mobility and Transport.

Capros

, Van Regemorter

, & Paroussos

, . (2017). GEM-E3 model manual. Retrieved from https://e3modelling.gr/modelling-tools/gem-e3/.

Clean Energy Manufacturing Analysis Center (CEMAC) (2016). Benchmarks of Global Clean Energy Manufacturing. USA: Clean Energy Manufacturing Analysis Center, Joint Institute for Strategic Energy Analysis, DOE’s National Renewable Energy Laboratory.

Curry

(2017). Lithium-ion battery: Costs and market. New York, NY: Bloomberg New Energy Finance.

De Cian

, Keppo

, Bollen

, . (2013). European-led climate policy versus global mitigation action: Implications on trade, technology, and energy. Climate Change Economics 4. Retrieved from https://www.worldscientific.com/doi/abs/10.1142/s2010007813400022

10.

Diaz Anadon

, Baker

, Bosetti

, & Reis

(2016). Expert views—and disagreements—about the potential of energy technology R&D. Climatic Change. Retrieved from https://doi.org/10.17863/CAM.30693.

11.

Donald

, Elgqvist

, & Santhanagopalan

(2015). CEMAC, Automotive lithium-ion battery (LIB) supply chain and U.S. competitiveness considerations. Retrieved from https://www.nrel.gov/docs/fy15osti/63354.pdf

12.

EC (2018a). Automotive industry – Internal market, industry, entrepreneurship, and SMEs. Retrieved from https://ec.europa.eu/growth/sectors/automotive.

13.

EC (2018b). Strategic action plan on batteries. Retrieved from https://ec.europa.eu/transport/sites/transport/files/3rd-mobility-pack/com20180293-annex2_en.pdf

14.

Fragkos

, Tasios

, Paroussos

, Capros

, & Tsani

(2017). Energy system impacts and policy implications of the European Intended Nationally Determined Contribution and low-carbon pathway to 2050. Energy Policy, 100, 216–226.

15.

Fraunhofer Institute for Solar Energy Systems ISE, Photovoltaics Report, Freiburg, 2018

16.

ICTSD (2017). International trade governance and sustainable transport: The expansion of electric vehicles. Retrieved from https://www.ictsd.org/sites/default/files/research/international-trade-governance-and-sustainable-transport-mahesh-sugathan.pdf

17.

IEA (2017). World Energy Outlook 2017. France: International Energy Agency.

18.

IEA (2018). Global EV Outlook. France: International Energy Agency.

19.

IEA (2019). Global EV Outlook 2019. France: International Energy Agency.

20.

Kane

(2017). EV battery makers 2016: Panasonic and BYD combine to hold majority of market. Retrieved from https://insideevs.com/ev-battery-makers-2016-panasonic-and-byd-combine-to-hold-majority-of-market/.

21.

Karkatsoulis

, Capros

, Fragkos

, Paroussos

, & Tsani

(2016). First-mover advantages of the European Union’s climate change mitigation strategy. International Journal of Energy Research, 40(6), 814–830.

22.

Lutsey

, Grant

, Wappelhorst

, & Zhou

(2018). Power play: How governments are spurring the electric vehicle industry (White Paper). International Council on Clean Transportation. Retrieved from https://www.theicct.org/publications/global-electric-vehicle-industry

23.

Navigant Research (2017). World wind energy market update 2017. Navigant Germany.

24.

Nykvist

& Nilsson

(2015). Rapidly falling costs of battery packs for electric vehicles. Nature, Climate Change, 5, 329–332.

25.

Olivetti

, Ceder

, Gaustad

, & Fu

(2017). Lithium-ion battery supply chain considerations: Analysis of potential bottlenecks in critical metals. Joule, 1, 229–243. Doi: 10.1016/j.joule.2017.08.019.

26.

Paroussos

, Fragkos

, Capros

, & Fragkiadakis

(2015, January). Assessment of carbon leakage through the industry channel: The EU perspective. Technological Forecasting and Social Change, 90, Part A, 204–219.

27.

Paroussos

, Fragkiadakis

, & Fragkos

(2019). Macro-economic analysis of green growth policies: The role of finance and technical progress in Italian green growth. Climatic Change, 1–18.

28.

Paroussos

, Fragkiadakis

, Fragkos

, Hinkel

, Mandel

, & Vrontisi

(2019). Climate clubs and the macro-economic benefits of international cooperation on climate policy. Nature, Climate Change, 9(7), 542–546.

29.

Transport and Environment (2016). Electric vehicles 2016. Retrieved from https://www.transportenvironment.org/sites/te/files/publications/TE%20EV%20Report%202016%20FINAL.pdf

30.

Van Soest

H. L.

, Reis

L. A.

, Drouet

, Van Vuuren

D. P.

, & Den Elzen

M. G. J.

(2018). Low-emission pathways in 11 major economies: Comparison of cost-optimal pathways and Paris climate proposals. Climatic Change, 142(3–4), 491–504.