Navigating through hydrogen

Policy Contribution
Issue n˚08/21 | April 2021

Navigating through hydrogen
Ben McWilliams and Georg Zachmann
Executive summary
BEN Mcwilliams (ben. Hydrogen is seen as a means to decarbonise sectors with greenhouse gas emissions that
mcwilliams@bruegel.org) are hard to reduce, as a medium for energy storage, and as a fallback in case halted fossil-fuel
is a Research Analyst at imports lead to energy shortages. Hydrogen is likely to play at least some role in the European
Bruegel Union's achievement by 2050 of a net-zero greenhouse gas emissions target.
Georg Zachmann (georg. However, production of hydrogen in the EU is currently emissions intensive. Hy-
zachmann@bruegel.org) is a drogen supply could be decarbonised if produced via electrolysis based on electricity from
Senior Fellow at Bruegel renewable sources, or produced from natural gas with carbon, capture, and storage. The the-
oretical production potential of low-carbon hydrogen is virtually unlimited and production
volumes will thus depend only on demand and supply cost.
Estimates of final hydrogen demand in 2050 range from levels similar to today's in
a low-demand scenario, to ten times today's level in a high-demand scenario. Hydrogen is
used as either a chemical feedstock or an energy source. A base level of 2050 demand can be
derived from looking at sectors that already consume hydrogen and others that are likely to
adopt hydrogen. The use of hydrogen in many sectors has been demonstrated. Whether use
will increase depends on the complex interplay between competing energy supplies, public
policy, technological and systems innovation, and consumer preferences.
Policymakers must address the need to displace carbon-intensive hydrogen with

low-carbon hydrogen, and incentivise the uptake of hydrogen as a means to decarbonise
sectors with hard-to-reduce emissions. Certain key principles can be followed without regret:
driving down supply costs of low-carbon hydrogen production; accelerating initial deploy-
ment with public support to test the economic viability and enable learning; and continued
strengthening of climate policies such as the EU emissions trading system to stimulate the
growth of hydrogen-based solutions in the areas for which hydrogen is most suitable.
Recommended citation
McWilliams, B. and G. Zachmann (2021) 'Navigating through hydrogen', Policy Contribution
08/2021, Bruegel
This content downloaded from 160.19.190.253 on Mon, 07 Oct 2024 12:06:50 UTC
All use subject to https://about.jstor.org/terms
1 Introduction
In the European Union’s decarbonisation drive, hydrogen is seen as a solution for sectors
with greenhouse gas emissions that are hard to reduce, as a means of energy storage, and as
a fallback in case halted fossil-fuel imports lead to energy shortages. The attractiveness of
hydrogen comes from the fact that no carbon dioxide is emitted when it is burned or used in a
fuel cell to produce electricity. In sectors where it could be applied, hydrogen could displace
fossil-fuel consumption and the associated carbon emissions.
Hydrogen is not a new fuel. Its ability to provide useful energy has been understood for
well over 100 years. As recently as the early 2000s, a wave of public interest focused on its
potential for powering automobiles (Lizza, 2003). Interest in hydrogen is now resurging in the
EU, linked to the bloc’s much more ambitious decarbonisation targets. On the demand side,
hydrogen could be a solution for particularly hard-to-abate sectors, such as steel, providing
a valuable argument that full decarbonisation is technically feasible. On the supply side, the
potential for large imports of low-carbon hydrogen is attractive when considered against
the argument that the EU’s clean energy potential might be too limited. Moreover, hydrogen
offers one solution to the seasonal storage issue that while renewable electricity generation
peaks in summer, demand peaks in winter.
Notwithstanding this technical promise, hydrogen remains prohibitively expensive. Its
use today in the European Union is thus far removed from the role optimists see it playing in
a net-zero EU in 2050. It is currently used almost exclusively as a chemical feedstock for the
production of ammonia and methanol and for crude oil refining. Furthermore, the domi-
nant production route for hydrogen – involving separation of hydrogen from methane – is
highly carbon-intensive. But hydrogen can also be produced from electricity via electrolysis.
The rapidly falling cost of electricity from renewables is creating excitement about low-cost,
low-carbon hydrogen production.
Figure 1: Estimated variation in hydrogen demand in 2050

Residential heat
Aviation
Steelmaking
Ammonia
Heavy-duty vehicles
Passenger vehicles
Shipping Residential
Oil refining
Transport
Industry
Light commercial veh.
Methanol
0 100 200 300 400 500 600 700

Annual demand (TWh)
Source: Bruegel. Note: Horizontal bars represent the range of annual hydrogen demand between our highest and lowest assumptions (see
section 3). The European Commission (2018) estimated total final energy demand in 2050 of 10,000 TWh. Some of the uses for hydrogen
shown in the figure are as a chemical feedstock, not energy consumption, but the 10,000 TWh figure still provides a sensible order of
magnitude. Our higher estimate (2,080 TWh) would see total hydrogen demand of approximately 20 percent of final energy demand in
2050, with the lower estimate (295 TWh) at 3 percent.
The future role hydrogen will play in the sectors where it could be deployed depends upon
the extent to which the necessary technologies reach commercial maturity. This will be driven
by the complex interplay of capital costs, consumer preferences, policy decisions, and the
2 Policy Contribution | Issue n˚08/21 | April 2021
relative performance of competing clean energy sources. Because of these uncertainties, we
estimate that in 2050, hydrogen could meet 20 percent of EU final energy demand – but it may
meet only 3 percent (Figure 1). This is in line with more sophisticated modelling studies.
This Policy Contribution examines the gap between 3 percent and 20 percent. Our analysis
supports the idea that decarbonisation will be driven mainly by electrification, while hydro-
gen will emerge to fill the niche for applications where electricity is either too expensive or
complex. We first explore the potential for hydrogen to evolve from today’s highly polluting
chemical feedstock to a key clean energy source in a decarbonised EU in 2050. The first fun-
damental step is the ability to produce significant volumes of clean hydrogen (section 2). We
then examine the main sectors in which hydrogen is currently being consumed or is consid-
ered an important pathway for future decarbonisation. To illustrate the uncertainty around
future hydrogen demand we assess what 2050 hydrogen demand might be in ten significant
sectors (section 3).
The difficulty for policymakers today lies in knowing exactly where the hydrogen niche
lies. It could cover whole sectors, such as aviation, or might cover sub-sectors, such as hydro-
gen fuel cells for heavy vehicles travelling long distances. Or hydrogen might find a temporary
niche, for example in heating of buildings. Despite the uncertainty, hydrogen’s current use
as a chemical feedstock and highly likely adoption in the steel sector mean that at least some
clean hydrogen will be required by 2050. Public policy, which we cover in section 4, should
therefore focus on stimulating cost reductions for the production of clean hydrogen.
2 Hydrogen supply
Large-scale production of hydrogen can be done using six very different inputs: natural
gas, electricity, biomass/waste, solar radiation, coal and oil. At least 16 different production
methods generate hydrogen from at least one of these inputs. Production methods differ
significantly in their associated greenhouse-gas emissions. For example, for production via
electrolysis (electricity is used to split water molecules into hydrogen and oxygen atoms), the
origin of the input electricity determines whether the hydrogen is carbon-neutral (eg when
produced from renewable or nuclear-generated electricity) or highly polluting (eg electricity
from lignite power plants). Figure 2 provides a schematic overview of the low-carbon produc-
tion pathways for hydrogen.
Figure 2: Low-carbon hydrogen production

Energy Concentrated Industrial
Natural gas Electricity Solar Waste Biomass
Resource solar byproduct
SMR with Thermal cracking Alkaline Photoelectro Waste gasification Biomass gasification
CCS methane electrolysis chemical with CCS (with CCS)
water
Production ATR with CCS Kvaerner process Solid oxide splitting Biomass pyrolysis
methods electrolysis
Water PEM Thermal (In)direct bio
thermolysis electrolysis electrolysis photolysis
Scientific research Prototype/demonstration Hydrogen

Commercial adoption
Source: Bruegel based on Hanley et al (2017), Nikolaidis and Poullikkas (2017), Piebalgs et al (2020) and IEA (2020). Notes: SMR = steam methane reforming. CCS = carbon capture and
storage. PEM = polymer electrolyte membrane. ATR= autothermal reforming.
The cost-competitiveness of different hydrogen production processes depend on the capi-
tal costs of the required installations, their technological efficiency in transforming input fuels
into hydrogen, the input fuel and carbon prices.
Hydrogen supply capacity in the EU is currently estimated at 339 terawatt hours per year1,
approximately 3 percent of EU final energy demand (FCH JU, 2019). Of this, over 95 percent
is hydrogen produced from fossil fuels, and less than 5 percent is produced via electrolysis
(Cihlar et al, 2020). Production of fossil hydrogen in Europe is mainly done by separation
of hydrogen from a stream of methane, a process that generates significant carbon dioxide
emissions. Box 1 compares these emissions to those from electrolysis, which depend on the
carbon intensity of electricity.
Box 1: Carbon emissions associated with hydrogen from methane and electrolysis
For hydrogen from methane without CCS, the carbon intensity of production is around 270g
CO2/kWh. For an electrolyser connected to the European electricity grid, average emissions
will be 430g CO2/kWh, based on current average electricity-related emissions of 285g CO2/
kWh. Therefore, electrolytic hydrogen will only result in better emissions performance than
SMR when the average emissions intensity of European electricity is reduced to significantly
below 200g CO2/ kWh. Extrapolation of current decarbonisation trends would see this
happening around 2025. Production from electrolysis will become even cleaner over time as
electricity is further decarbonised.
Very low carbon intensities can already today be achieved in many hours of the year,
such as on sunny and windy summer weekends, or in certain EU countries, such as France
and Denmark. But making hydrogen production ‘low-carbon’ by producing it from green
electricity has no economic justification because it would only imply that other consumers
would consume non-green electricity. To ensure that domestic hydrogen production does
not result in increasing emissions, the cap of the EU emissions trading system (which covers
hydrogen production from electricity and natural gas) should be tightened enough to meet
the EU climate targets.
The EU hydrogen strategy, published in July 2020, aims to set out a vision for “how the
EU can turn clean hydrogen into a viable solution to decarbonise different sectors” (European
Commission, 2020). It is centred on scaling up electrolysis production with renewable elec-
tricity input. An alternative option would be to apply carbon capture storage (CCS) tech-
nology in the production of hydrogen from methane, capturing up to 90 percent of the CO2
emissions generated2 (IEA, 2019a). The strategy sees a role for CCS in hydrogen production in
the short and medium terms, but not as a long-term priority.
The competitiveness of hydrogen from methane compared to electrolysis depends on the
price of the inputs (natural gas or electricity) and the carbon price (Figure 3). Our estimates
for methane production use a price of €20/MWh, while the range of electricity prices likely
available to industrial producers is the vertical shaded grey area, around €40-€50/MWh3.
Figure 3 shows that at current natural gas, electricity and carbon prices4, hydrogen pro-
duction from methane without CCS is significantly cheaper than hydrogen production from
1 In this paper we transfer all energy units (electricity, hydrogen, natural gas, etc) into terawatt hour (TWh) for easier
comparability. One TWh is about 0.03 million tonnes hydrogen or 92 million cubic meters of Russian natural gas.
2 90% is a technical maximum. The range of carbon captured is likely to be in the range of 60-90%. Capturing carbon
between 60 and 90% is relatively more expensive.
3 See https://www.eex.com/en/market-data/power/futures where futures at the Belgian energy exchange frequently
settle at around €50, and https://www.powernext.com/futures-market-data where an index of European natural gas
prices stays around €20.
4 In fact, we consider a carbon price of €50 a likely upper bound for the next three years.
electricity5. However, if electricity prices drop to about €20/MWh (for example because of
cost reductions related to renewables), hydrogen produced from electricity would become
cheaper than that produced from methane. On the other hand, increases in the carbon price
would also affect the cost. An increase in the carbon price to €200/tonne would mean that
electricity at current prices would become competitive with natural gas.
Figure 3: Hydrogen price for different electricity and carbon prices

140
Electrolysis
120
Hydrogen cost, €/MWh

100 Future high carbon price
Gas price: €20/MWh
80 CO2 price: €200/tonne
60 Current situation
Gas price: €20/MWh
40 CO2 price: €0-100/tonne
Current
20 price
range
0
0 10 20 30 40 50 60 70 80
Electricity price, €/MWh
Source: Bruegel based on IEA. Note: the graph shows the different hydrogen prices for different electricity prices using an electrolyser,
where the cost of carbon is already internalised through the carbon price paid by electricity generators. For natural gas production, we
assume a gas price of €30/MWh. The dashed lines represent different hydrogen costs for different carbon prices paid for a methane plant
without CCS. Calculations based on IEA assumptions: for electrolysis, CAPEX - $900/kWe, efficiency – 64%, Annual OPEX – 1.5% of CAPEX.
For natural gas reforming, CAPEX - $910/kWH2, Efficiency – 76%, Annual OPEX = 4.7% of CAPEX, emissions = 8.9kgCO2/kgH2.
Furthermore, the capital costs of electrolysers could fall significantly in the near term,
meaning that electrolysis would be competitive even if electricity prices do not drop to €25/
MWh. Wood Mackenzie (2020), for example, forecast electrolysis-produced hydrogen becom-
ing cost-competitive with methane-produced hydrogen between 2030 and 2040, depending
on the region, because of shifting cost dynamics. Commitments in hydrogen strategies pub-
lished so far by the EU, its member states and other countries to deploy electrolyser capacity
are set to stimulate cost reductions.
Moreover, our analysis is based on average EU values and cost assumptions. Differing tax
rates, network costs and wholesale prices drive significant regional electricity price differ-
ences. The competitiveness of electricity versus gas will therefore vary between regions.
And we base our analysis on an electrolysis plant behaving as a traditional baseload
consumer of electricity, ie demanding electricity with limited flexibility, which is the situation
with electrolysers today. However, developments in alternative electrolysis technology6, the
falling capital cost of electrolysers and the increasing variability of electricity prices (because
of increasing shares of renewable energy generation) could increase demand for electrolysis
as a source of flexible power demand. Therefore, electrolysis could emerge as a significantly
more competitive technology by: a) utilising close to zero (or occasionally even negative)
electricity prices for a substantial number of hours, and b) providing flexibility services to
the grid by consuming excess electricity at times of excess supply and helping to facilitate the
over-deployment of renewable electricity sources (see section 3.4).
5 Estimates using IEA data. This illustrative example assumes that the required installations for hydrogen production
from methane and from electrolysis are available.
6 The EU is in particular supporting the development of proton electron membrane (PEM) electrolysers. For
example the REFHYNE project (https://refhyne.eu/) will install and operate the world’s largest hydrogen PEM
electrolyser with 10MW capacity. This is important because PEM electrolysers are able to more quickly adjust
demand in response to fluctuating electricity supply compared to conventional electrolysers.
2.1 Alternative production pathways
Hydrogen production from natural gas and electricity are the most common methods, but
there are others. Table 1 lists them, along with some rough cost estimates.
Table 1: Additional low-carbon hydrogen production methods

Hydrogen cost
Production method Energy source Feedstock
est. (€/MWh)
Autothermal reforming
Fossil fuels Natural gas 50
with CCS
Methane pyrolysis/thermal Internally generated
Natural gas 54 - 57
cracking steam
Internally generated
Biomass pyrolysis Biomass 42 - 74
steam
Internally generated
Biomass gasification Biomass 60 – 69
steam
Direct bio-photolysis Solar Water and algae 72
Indirect bio-photolysis Solar Water and algae 48
Dark fermentation Organic biomass 87
Photo-fermentation Solar Organic biomass 96
Solar thermal electrolysis Solar Water 172 - 354
PEC process (photo-
Solar Water 350
electrolysis)
Nuclear thermolysis
Nuclear Water 73 - 89
(thermal cracking of water)
Solar thermolysis (thermal
Solar Water 269 - 284
cracking of water)
Source: Kayfeci et al (2019). Note: CCS = carbon capture and storage.
In sum, the global technical production potential of hydrogen exceeds demand by several
orders of magnitude7, meaning expansion of supply depends in principle only on the hydro-
gen production cost and demand at that cost level. National hydrogen production costs can
differ depending on the differing availability and cost of inputs and capital, the availability of
required infrastructure for transport, hydrogen storage and possibly carbon storage space.
2.2 Hydrogen imports

In optimistic scenarios, hydrogen could contribute a significant share of final energy demand
within the EU by 2050. The EU hydrogen strategy works with a projection of 13 percent to 14
percent by 2050 (European Commission, 2020). If hydrogen demand is to reach such levels,
imports of hydrogen might also develop. The European Commission hydrogen strategy aims
to develop 40 GW of hydrogen capacity in neighbourhood regions by 2030 – the same capacity
the EU aims for within its borders. From countries with an abundance of renewable energy
resources, green hydrogen could become an attractive export.
Installing renewables and electrolysers outside the EU and importing the hydrogen into
the EU only makes economic sense when the renewables conditions in the exporting coun-
tries are significantly better, while capital costs are not substantially higher than in the EU.
Furthermore, the cost advantage must exceed the costs of delivery of hydrogen as a gas via
pipelines. Alternatively, hydrogen can be transformed into, for example, ammonia, which can
be more easily stored in liquid form and transported by ship.
Based on IEA assumptions of current costs it seems hard to make a case for imports of
7 Solar potential of some 600,000 TWh/year from Korfiati et al (2016) alone would enable around 12,000 Mt of
hydrogen production.
hydrogen from solar energy from North Africa. If deployment of additional wind or solar units
in Germany becomes difficult because suitable/acceptable land is already utilised, while
investment costs in Africa decline, imports of hydrogen might become competitive. However,
consistent international rules would be needed to ensure that significant imports of hydrogen
do not directly or indirectly increase net emissions in the producing country, for example
through land-use change or replacement of renewable electricity for local populations by
fossil fuels.
Figure 4: Import vs domestic hydrogen (€/MWh)
North Africa
Investment cost
Electricity cost
Germany
Transport cost
0 20 40 60 80 100 120 140
Source: IEA (2019a). Note: Key assumptions: CAPEX electrolyser: $900/kW, electricity price in Germany: $47/MWh, electricity price in
North Africa: $23/MWh, interest rate in Germany: 5%, interest rate in North Africa: 10%, transport distance: 3,000Km, pipeline transport
cost of $2/Kg.
3 Hydrogen demand
The future evolution of demand for hydrogen in Europe is highly uncertain. Hydrogen has
historically had a limited role in influential global energy modelling studies (for example,
Quarton et al, 2020). In this section, we discuss the most likely sectors for future hydrogen
demand, with calculations for high, medium and low hydrogen demand scenarios. We thus
provide a broad overview of what hydrogen demand might look like in three scenarios: one
in which hydrogen technology and deployment is aggressively pursued by policymakers and
costs continue to fall, one in which the exact opposite occurs, and one in the middle. Our
numbers are not intended to be forecasts, but rather serve to highlight the significant uncer-
tainty surrounding future hydrogen demand8.
The evolution of competing or complementary decarbonisation options, including energy
efficiency, biomass, electrification and carbon capture, will be significant for determining the
role or niche for hydrogen. Hydrogen therefore cannot be considered in isolation but rather in
combination with the development of others fuels and energy carriers within complex energy
systems (Hanley et al, 2017).
Our assessment of hydrogen demand focuses on three broad sectors: transport (section
3.1), industrial applications (3.2) and residential heating (3.3). We also discuss the role hydro-
gen may play in the power sector (3.4). Table 2 provides an overview.
8 Our calculations are predominantly built on interpretations of the European Commission’s ‘Clean Planet for all’
strategy (European Commission, 2018). Additional sources are used to complement our analysis in many cases. A
footnote below each set of numbers briefly explains the underlying calculations.
Table 2: Sector scorecard
Sector Emissions (% Hydrogen Note
of EU total) potential
Already using hydrogen produced from natural gas/
Ammonia & methanol 14%* ‫ۻۻۻۻۻ‬ industry by-product
Already using hydrogen produced from natural gas/
Oil refining 2% ‫ۻۻۻۻ‬ industry by-product
Steelmaking 4% ‫ۻۻۻۻۻ‬ High potential to replace coal
Electric vehicles hold first-mover advantage in low

Passenger vehicles 12% ‫ۻ‬ carbon market
Light commercial
vehicles
2% ‫ۻۻ‬ Electric vehicles likely to be strong competitors
Heavy duty vehicles 5% ‫ۻۻۻ‬ Hydrogen more suited to heavier vehicles
Shipping 7% ‫ۻۻۻۻ‬ Potential additional demand via use as ammonia
Aviation 4% ‫ۻۻۻ‬ Synthetic fuels; fuel cells
Residential heating 12% ‫ۻۻ‬ Competing with electricity
Source: Bruegel. Note: *emissions include all chemical sector, not only ammonia and methanol.
3.1 Transport
There are multiple options for hydrogen consumption in the road, rail, maritime and air
transport. Pure hydrogen can be consumed directly through a fuel cell to produce electricity,
or combusted. Alternatively, hydrogen can be transformed into ammonia before use in a fuel
cell or by combustion. Finally, hydrogen can also be used as a building block for renewable
synthetic fuels (e-fuels).
Figure 5: GHG emissions share in transport sector

Other 2.64%
Shipping
21.85% Passenger
vehicles
38.72%
Aviation
11.78%
Railways 0.35%
Heavy duty vehicles 17.01% Light commercial vehicles 7.65%
Source: Bruegel.
Transport: Road
Passenger vehicles
• Hydrogen potential: ‫ۻ‬
• Upper demand: 140 TWh. Medium demand: 50TWh. Lower demand: 0 TWh9
• 12 percent of EU greenhouse gas emissions
In road transport, hydrogen faces direct competition from electricity. Increasingly, the
decarbonised future of passenger vehicles looks to be one of battery electric vehicles (BEVs).
The price of batteries has rapidly dropped while range per charge is increasing. As a result,
the global stock of fuel cell (hydrogen) vehicles is just 11,200 compared to more than 5 million
BEVs (IEA, 2019). BEVs now enjoy a first-mover advantage as the conventional low-carbon
passenger vehicle. They attract significantly more government and private-sector funding,
particularly for charging infrastructure.
Nonetheless, there may be some scope for hydrogen if limitations arise because of raw
material shortages, technological limitations of batteries or excess strains on electricity
grids arising from too many poorly managed BEVs. Moreover, certain companies (Hyundai,
Honda) are still actively developing fuel-cell electric vehicles (FCEV), ie hydrogen passenger
vehicles. As markets grow and prices decrease, it is possible that FCEVs will one day compete
more seriously with BEVs. Large-scale deployment of hydrogen refuelling networks would
be fundamental to this but these currently still face the problem that while FCEV take-up is
low, investment in refuelling networks is not attractive. As other economic sectors begin to
demand more hydrogen, the roll-out of hydrogen refuelling networks may become economi-
cally more attractive.
Hydrogen offers quicker refuelling than battery charging, making it potentially more
suited to vehicles in constant use, such as taxis and buses.
Heavy-duty vehicles
• Hydrogen potential: ‫ۻۻۻ‬
• Upper demand: 200 TWh. Medium demand: 120 TWh. Lower demand: 10 TWh10
• 5.2 percent of EU greenhouse gas emissions (including buses)
Hydrogen appears to have greater potential for the heavy-duty road transport sector
because hydrogen is able to store more energy in a smaller space and at lower weight than
a lithium-ion battery. A challenge for manufacturers of battery electric vehicles has been
producing batteries which contain sufficient energy but are not too heavy. For example, to
provide the same range as a 1000 litre diesel truck, the battery of an electric truck would have
to weigh about 14 tonnes. As the capacity and range of lithium batteries has expanded, this
problem is gradually being overcome for small, passenger vehicles. However, hydrogen fuel
cells could be deployed in heavier vehicles for which greater range and higher power output
are required.
In this market segment, hydrogen would compete against biofuels and the use of electri-
cally-derived fuels (via hydrogen). The speed of battery improvements has been rapid so far,
and it is still very possible that innovations will allow battery-driven electrification to domi-
nate heavy-duty transport. Overhead transmission lines may also play a limited role.
The most optimistic EU 2050 scenarios see approximately a 15 percent share of hydrogen
9 Figures estimated using the growth rate in passenger vehicles assumed by European Commission (2018). Upper
demand based on 15 percent of the vehicle stock in 2050 being hydrogen fuel cell, 5 percent for medium, and 0
percent for lower.
10 Figures estimated using the growth rate in heavy duty trucks to 2050 from European Commission (2018). Upper
bound assumes 25 percent hydrogen fuel cell composition of 2050 heavy duty fleet, medium and lower bounds
assume 15 percent and 1 percent respectively.
FCEVs in the heavy goods vehicle stock (European Commission, 2018). Least optimistic sce-
narios would see 0-3 percent FCEV deployment. Some additional indirect hydrogen demand
might occur through electrically derived fuels.
Light-commercial vehicles
• Hydrogen potential: ‫ۻۻ‬
• Upper demand: 60 TWh. Medium: 15 TWh. Lower: 0 TWh11
• 2.3 percent of EU greenhouse gas emissions
Vans and light commercial vehicles occupy the middle ground between passenger
vehicles and heavy-duty vehicles. They tend to be slightly larger than passenger vehicles,
giving hydrogen an advantage because of its higher energy density, but not comparable to
heavy-duty vehicles, meaning it is still very possible that this market will be dominated by
BEVs. Currently, over 90 percent of light commercial vehicles in the EU are diesels (European
Commission, 2018). The deployment of hydrogen fuel cells in this sector may likely depend
on the initial success of hydrogen fuel cell deployment elsewhere (particularly in heavy-duty
vehicles). However, similarly to passenger vehicles, current market dynamics would still sug-
gest that BEVs will dominate this market.
Transport: Rail
• Hydrogen potential: ‫ۻ‬
• Demand: likely to be very close to zero
The strongest decarbonisation opportunities are in electrifying rail tracks, shifting away
from diesel consumption. Electrifying tracks implies significant upfront fixed costs. Tracks
electrified so far are those which are the most heavily used in order to increase the ratio of
returns to a fixed investment. For less-used tracks, the returns are not large enough to justify
the significant upfront capital costs of electrification. On these tracks, hydrogen fuel cells are
an attractive option (IEA, 2019).
The potential scope is still relatively small as approximately 50 percent of European tracks
have already been electrified (Donat, 2020). Take up of hydrogen for trains on non-electrified
tracks can be aided by falls in the costs of fuel cells, driven by deployment elsewhere. Battery
electric trains are another option.
Overall, rail is not likely to be a leading candidate sector for large volumes of hydrogen
consumption.
Transport: Shipping
• Hydrogen potential: ‫ۻۻۻۻ‬
• Upper demand: 120 TWh. Middle demand: 70 TWh. Lower demand: 20 TWh12
The maritime-fuel mix in the EU and globally is dominated by heavy fuel oil. Policy
restrictions on sulphur emissions and planned controls on greenhouse gas emissions mark an
attempt to move beyond heavy fuel oil. The European Commission is considering including
shipping in the EU emissions trading system.
11 Figures estimated using the growth rates in light commercial figures to 2050 from European Commission (2018).
Upper bound assumes 20 percent hydrogen fuel cell composition for 2050 light-duty fleet, medium and lower
bounds assume 5 percent and 0 percent respectively.
12 Figures estimated on the basis of hydrogen-optimistic and hydrogen-pessimistic scenarios for final energy
demand in the shipping sector from European Commission (2018). These figures exclude indirect demand for
hydrogen that would arise if ammonia were used as a fuel.
Hydrogen fuel cells may work for short-distance light shipping, for which power require-
ments are not too large. This is likely to be in competition with battery electric ships. Lique-
fied hydrogen, synthetic fuels derived from hydrogen and ammonia (Middlehurst, 2020),
have greater potential in terms of decarbonising longer distance shipping. Like hydrogen,
ammonia can be used to produce energy either by combustion within an internal combustion
engine, or by producing electricity through a fuel cell. Biofuels are likely to be another com-
petitor for hydrogen in the maritime sector.
A challenge will be to transform bunkering, or fuelling, facilities, which currently store
heavy fuel oils, so they can store hydrogen or hydrogen-derived fuels. Here, a global coordina-
tion problem arises as ships must refuel in multiple locations, normally in different coun-
tries. For this reason, it is quite likely that one or two fuels will become dominant. Hydrogen
might be boosted by other uses in port operations. Forklift trucks are already a big adopter of
hydrogen, with 25,000 deployed globally, for example. Port hydrogen storage and distribution
infrastructure will become economically more efficient with multiple end-use cases.
Transport: Aviation
• Hydrogen potential: ‫ۻۻۻ‬
• Upper demand: 340 TWh. Middle demand: 180 TWh. Lower demand: 0 TWh13
For short-distance • 3.60 percent of EU greenhouse gas emissions
flights, electricity and
pure hydrogen could For short-distance flights of less than 3,000 kilometres (encompassing most European
make a significant flights; Madrid to Helsinki is about 2950km, for example), electricity and pure hydrogen could
contribution make a significant contribution. This may be through battery or fuel cell (hydrogen) electric
planes, or through direct combustion of hydrogen. Hybrid options, combining the two (elec-
tricity and hydrogen combustion) are also possible.
Airbus has released three concept designs for hydrogen planes which they state could
enter service by 2035 (Airbus, 2020). The proposed planes are of a hybrid nature, combusting
hydrogen in modified gas-turbines and producing electricity through fuel cells.
Longer distance flights require fuels with higher energy densities. Advanced biofuels and
synthetic fuels14 derived from hydrogen are the most promising decarbonisation options.
Synthetic jet fuel can be a drop-in replacement for current jet fuel. However, options are today
far too expensive15. Significant policy support and cost reductions would be required for syn-
thetic fuels to be a realistic decarbonisation option.
For longer distance flights, the evolution of biofuels will be a key determinant for the
potential of hydrogen fuels. Biofuel production is constrained by land availability16 and any
constraints on biofuel production will provide a stimulus for investment into hydrogen. A fur-
ther influencing factor will be the extent to which biofuels are demanded by other economic
sectors.
Therefore, there are two separate considerations for future hydrogen demand in aviation:
directly through use in a fuel cell/combustion to power short-distance flights, or indirectly
producing synthetic jet fuels which are then combusted during flight. We estimate an upper
bound of 210 TWh of direct hydrogen use in aviation, and 130 TWh indirect hydrogen use for
13 Total energy demand for aviation sector in EU taken from European Commission (2018). Upper demand assumes
30 percent of demand met by direct hydrogen (ie fuel cell + combustion). Of the remaining 70 percent, jet fuel or
equivalent substitutes are used. Of this demand, 20 percent is assumed to be met by synthetic fuel production from
hydrogen. Lower demand is zero in the case that hydrogen technology does not develop. Medium is midpoint.
14 Synthetic fuel broadly refers to the concept of a chemical fuel synthesis in which hydrogen is reacted with carbon
from carbon dioxide in order to produce hydrocarbons with a significant commercial value (eg methane). When
hydrogen is produced from green electrolysis and carbon dioxide is captured from the air, this can theoretically be
a zero carbon emission fuel.
15 The implied mitigation cost of using power-to-liquid to produce synthetic jet fuel would be in the order of €800/
tonne CO2 (Pavlenko et al, 2019).
16 Biofuels from seaweed could address this issue but are not yet commercially proven (Bellona Europa, 2020).
producing synthetic fuels.
However, aviation remains firmly in the hard-to-decarbonise box, with technologies at a
very immature stage of development. It will take many years of research and development
before the potential of hydrogen relative to alternatives is clarified. Moreover, as one of the
hardest sectors to decarbonise, aviation is a strong contender for residual emissions in a net-
zero 2050 scenario that involves significant use of negative emissions technologies. Aviation
may therefore to some extent carry on burning conventional fossil fuels and emitting green-
house gases.
3.2 Industry
Currently, over 90 percent of hydrogen produced in Europe is used as a feedstock in oil
refining, ammonia and methanol production (Cihlar et al, 2020). The possibility of substitut-
ing hydrogen for fossil fuels used in steel production is one of the most commonly discussed
future uses for hydrogen. These four sectors together account for up to 41 percent of the EU’s
industrial emissions17.
Figure 6: EU industrial greenhouse gas emissions
Chemicals 14%
Other 59%
Steel 16%
Oil refining 11%
Source: Bruegel.
Chemical sector: ammonia and methanol

• 3.2 percent of EU greenhouse gas emissions18
The ammonia and methanol sectors both require hydrogen as a feedstock. The most con-
venient and cost-effective source is fossil-fuel derived hydrogen.
Ammonia production
• Hydrogen potential: ‫ۻۻۻۻۻ‬
• Upper demand: 240 TWh. Medium: 160 TWh. Lower: 100 TWh19
• 2015 demand: 129TWh
Over 80 percent of ammonia produced worldwide is for the manufacture of fertilisers

(Bazzanella and Ausfelder, 2017). Other uses are for nitric acid, pharmaceuticals and cleaning
products.
In Europe, natural gas is the most important feedstock. Hydrogen is extracted from natural
gas (methane) before being combined with nitrogen from the air to produce ammonia, or
NH3. Green hydrogen would therefore be able to directly reduce emissions from ammonia
17 Up to 41 percent with chemical sector emissions used to represent methanol and ammonia.
18 For the whole chemical sector, not just ammonia and methanol.
19 Based on the assumption of a 178 kilogramme hydrogen requirement per tonne of ammonia. The variation arises
from different final demands for ammonia in the EU in 2050.
production by eliminating the need for production of hydrogen from methane20. Such green
ammonia projects are already underway21.
Europe currently produces 17 million tonnes of ammonia annually and the future evolu-
tion of demand is uncertain. As the global population increases, demand for ammonia-based
fertilisers will increase; food production must become more efficient to feed an increasing
number of mouths from the same amount of land. However, public policy may drive out
ammonia in favour of biological fertilisers or higher levels of organic production. The EU in
2019 updated fertiliser rules to promote fertilisers based on organic materials rather than
chemicals22.
Our analysis is based on traditional uses of ammonia, but ammonia demand could rise
significantly if ammonia becomes a significant future energy carrier. Ammonia could be a
preferable option for transporting the energy contained in a hydrogen atom (ammonia’s
physical properties make it easier to transport than hydrogen). Ammonia could help in
transporting energy from areas of renewable energy abundance to areas of demand. Moreo-
ver, non-traditional demands for ammonia may arise in shipping (section 3.1) and potentially
even in the power sector23. Such a scenario would significantly increase hydrogen demand for
ammonia production.
Methanol production
• Upper demand: 30TWh. Medium demand: 25TWh. Lower demand: 15TWh24
• 2015 demand: 27 TWh
Similarly to ammonia, demand for hydrogen in methanol production is predominantly

met by hydrogen from methane. The production of green hydrogen would reduce demand for
hydrogen from natural gas and its significant carbon emissions.
Currently, EU methanol production (1.5Mt/annum) as a share of global production is
much lower than for ammonia. Assuming similar trends, final demand for hydrogen in this
sector is likely to be significantly lower than in the ammonia sector within the EU.
Oil refining
• Hydrogen potential: ‫ۻۻۻۻ‬
• Upper demand: 110TWh. Medium demand: 90TWh. Lower demand: 50TWh25
• 2015 demand: 153 TWh
A major use of hydrogen today is in oil refining: turning crude oil into commercially attrac-
tive end-use products. Hydrogen is used in hydrotreating and hydrocracking. Hydrotreating
refers to the removal of sulphur impurities from crude oil, necessary because sulphur is an
air pollutant. Hydrocracking is used to transform heavier residual oils into lighter and more
commercially attractive fuels.
20 Production of hydrogen from methane emits 1.83 tonnes of CO2 per tonne of ammonia (Bazzanella and Ausfelder,
2017).
21 For example, a 100MW wind-powered renewable hydrogen production plant in the Netherlands developed by
power company Ørsted and fertiliser company Yara (Durakovic, 2020).
22 See https://www.consilium.europa.eu/en/press/press-releases/2019/05/21/eu-adopts-new-rules-on-fertilisers/.
23 Ammonia can be combusted to produce electricity. On a small scale, it is currently co-fired in coal plants to
produce electricity with lower emissions.
24 Based on assumption of 189 kilogrammes hydrogen per tonne of methanol. Variation arises from differences in EU
final methanol demand in 2050.
25 Our estimations first take IEA trends for a slight decrease in hydrogen requirements in oil refining from 2020 to
2030 while assuming constant demand for oil refining. The scenarios then differ depending on assumptions on the
decrease in demand for oil refining.
Future demand for hydrogen in this sector will be determined by future demand for crude
oil products, which in Europe is set to decrease. Meanwhile, sulphur restrictions are progres-
sively being tightened, increasing the hydrogen demand per barrel of crude oil26. Ironically,
sulphur restrictions on crude oil products such as jet fuel have in recent years likely increased
the sector’s greenhouse emissions because of the current carbon intensity of hydrogen
(Catalá et al, 2013, Figure 4.5.5). In 2050, there will likely still be demand in the oil refining
sector because of the use of hydrocarbons in certain chemical products.
Steelmaking
• Upper range: 240TWh. Middle range: 150TWh. Lower range: 100TWh27
The EU produces 177 million tonnes of steel a year, 11 percent of global output28. Signifi-
cant emissions are associated with the steel sector and hydrogen is widely regarded as funda-
mental to decarbonising the sector.
Most steelmaking greenhouse gas emissions are associated with the turning iron ore into
iron prior to its processing into steel. Steel can be produced in blast oxygen furnaces (BOF)
(60 percent of EU production; European Commission, 2018) and electric arc furnaces (EAF).
The BOF route produces steel using coal and has little future in a decarbonised world, though
efforts are being made to reduce emissions by improving efficiency, replacing some coal with
hydrogen and retrofitting plants with carbon capture technology. However, unless carbon
capture can be done at levels of emissions far above capabilities today, there will always be
significant emissions associated with BOF.
Decarbonisation of steel production therefore relies on switching to the EAF (currently 40
percent of EU production). Here, the primary energy input is electricity29, making green steel
possible if the electricity is decarbonised. Two different feedstocks can be used with EAF:
scrap steel and direct reduced iron (DRI), or a combination.
Globally, scrap steel contributes to about 25 percent of steel production. Increasing the
use of scrap steel would be a welcome shift toward the circular economy30, but is limited by
availability of high-quality scrap31. Meanwhile, producing DRI for use in EAF involves reacting
iron ore with a reducing agent, currently a mixture of hydrogen and carbon monoxide. This
is already a technologically proven route, with deployment particularly in the Middle East
where industry has access to low-cost natural gas, which is used for producing the stream of
hydrogen and carbon gases for reduction.
All major European steelmakers are currently building or testing hydrogen-based
reduction for use in EAF32. The target is to use pure hydrogen rather than a hydrogen/carbon
mixture for reduction of iron ore. Using both scrap steel and DRI produced using hydrogen
in electric arc furnaces is considered the most viable decarbonisation option for the sector
within the EU (Hoffmann et al, 2020). A related question is whether the move to DRI-EAF
26 See for example the International Maritime Organisation’s IMO 2020 rule: https://www.imo.org/en/MediaCentre/
HotTopics/Pages/Sulphur-2020.aspx.
27 Demand assumed constant at today’s level. Upper assumes all demand met from electric arc furnaces (EAF)
with 60 percent direct reduced iron (DRI)/40 percent scrap steel feedstock. Lower assumes 50 percent EAF, and
feedstock 50 percent DRI/50 percent scrap steel. Medium assumes 75 percent EAF and feedstock of 50 percent
DRI/50 percent scrap steel.
28 See https://ec.europa.eu/growth/sectors/raw-materials/industries/metals/steel_en.
29 Electric arc furnaces can also be rapidly started and stopped. A shift in steel production towards electricity could
therefore have positive spillover effects for demand response in electricity grids with lots of variable renewable
power.
30 As well as removing the majority of emissions which are associated with the reduction of iron ore.
31 Recycled steel can be contaminated with other elements, most commonly copper. This reduces the quality of steel.
32 HYBRIT in Sweden is a well-known example. See https://www.ssab.com/company/sustainability/sustainable-
operations/hybrit.
will affect the location of steel production from close to coal/iron resources to close to cheap
green-energy resources.
One issue is the long lifespan of steel plants – approximately 35 years. The production of
steel through DRI-EAF using hydrogen is not yet economically mature. However the indus-
try must be wary of locking in any further BOF capacity, with such facilities likely to become
stranded assets by 2050.
3.3 Residential heating

• Hydrogen potential: ‫ۻۻ‬
• Upper demand: 600 TWh. Medium demand: 300 TWh. Lower demand: 0 TWh33
Natural gas is currently the most common primary fuel used for household heating in
the EU, accounting for 44 percent of demand. Coal, oil, and biomass are the other significant
contributors (Bertelsen and Vad Mathiesen, 2020).
Energy efficiency is the main tool for ensuring decarbonisation of the buildings sector. The
EU’s long-term roadmap sees energy demand for residential heat halving in a baseline sce-
nario (European Commission, 2018, Figure 39). Demand reductions will be achieved through
a combination of rules for new build and existing households. From 2021, new buildings must
comply with requirements in the Energy Performance of Buildings Directive 2010/31/EU: new
buildings must be nearly zero energy consumption. Old buildings must be renovated, and
heating demand reduced through better insulation. The EU’s 2020 Renovation Wave strategy
is intended to address exactly this issue34.
Domestic heating can also become more electrified. Electrically powered heat pumps,
with an efficiency of 300 percent, are able to draw three times more heat energy from outside
air than they consume in terms of electric energy. Currently, the share of electricity in final
residential heating demand is approximately 5 percent but European Commission scenarios
forecast a growth in this share to between 22 percent and 44 percent by 2050 (European Com-
mission, 2018, Figure 43).
Nonetheless, as a temporary solution, blending natural gas with hydrogen in gas grids is
being discussed. Technically, this can be done up to a certain proportion (roughly 5 percent
to 20 percent). In the short run, the blending of hydrogen into gas grids allows for incremental
reductions in emissions while creating an early demand market for green hydrogen.
To achieve concentrations of hydrogen in gas grids above 20 percent, pipes and grid
appliances must be retrofitted. This is not an impossible task; grids in the United Kingdom
were retrofitted in the 1960s to move away from town gas (a mixture with a high hydrogen
concentration) to natural gas. Switching a gas distribution grid to hydrogen would be organ-
ised top-down and would require less significant investments on the user side to decarbonise
residential heating, while moving to electric heat pumps will in principle be more efficient
and allow for gradual switching of users at their convenience. But it will be more difficult to
push individual users to make the necessary substantial investments – one might consider
this in light of the difficulties faced with smart meter roll-outs across Europe35. Moreover, the
required strengthening of electricity distribution grids would also have to be financed.
Under a scenario in which electrification is pursued as the primary residential heat-
33 Based on modelling results from European Commission (2018). Upper demand is taken from the H2 scenario –
this scenario achieved an 80 percent reduction in emissions. We assumed a slightly increased hydrogen demand to
reach a 100 percent reduction. Medium and lower linearly extrapolated to zero.
34 See https://ec.europa.eu/energy/topics/energy-efficiency/energy-efficient-buildings/renovation-wave_en.
35 In 2014, it was estimated that the penetration of electricity smart meters in the EU in 2020 would be 72 percent.
Most recent estimates suggest that the actual figure is about 43 percent. Lack of consumer acceptance, often for
privacy reasons, has been a main reason for delay (Tounquet and Alaton, 2020). While heat pumps should not
present privacy issues, the example clearly illustrates the challenges associated with a policy that requires the
agreement of individual households.
ing technology, hydrogen may still play a complementary role. Decentralised provision of
hydrogen (ie gas bottles) could supplement residential heating on the coldest days to prevent
excessive strain on local electricity distribution grids.
A final option involves keeping the natural gas network much as it is today but injecting
biomethane36 or synthetic methane produced by combining hydrogen with carbon dioxide.
An obvious advantage is minimal disruption to the grid. However current levels of supply of
biogas fall far short of demand, and synthetic methane is an inefficient source of energy and is
very expensive.
3.4 Hydrogen as an enabler of renewable electricity deployment

In addition to deployment in end-use sectors, hydrogen could be used for energy storage,
enabling the integration of increasing shares of variable renewable generation into electricity
systems.
Historically, electricity grids have operated on the basis of volatile aggregate demand from
end-users being met by a mix of inflexible base-load (nuclear, lignite, run-of-river) and peak-
load power that is dispatched on demand (for example gas or hard coal), with relatively little
storage. Increased adoption of variable renewable electricity sources is changing this model.
A challenge for grid operators is to maximise the uptake of renewable electricity that is pro-
duced when the sun is shining and the wind blowing. A number of options, beyond the scope
of this Policy Contribution, are under consideration, including the use of hydrogen produced
from electrolysis.
Short-term flexible demand

Hydrogen production via electrolysis could be increased during times of excessive renewable
power generation and reduced when supply is weak, allowing more efficient balancing of
the electricity market. Kopp et al (2017) showed that already in 2016, a 6 MW electrolyser in
Mainz, Germany was deployed with economic benefit to the German control reserve market.
Whether electrolysers can be competitive as providers of grid-balancing services will
depend on technological and regulatory developments in the next few years. In particular,
battery storage systems that already feature much lower storage losses than hydrogen will
likely see their capacity costs drop dramatically as more batteries are produced and deployed.
They may therefore be better suited than electrolysis to managing intra-daily or even intra-
weekly fluctuations on electricity grids.
Long-term seasonal storage

Hydrogen could be a more useful option for managing fluctuations in renewable electricity
produced in different seasons. Hydrogen could be produced during months of excess renewa-
ble electricity production, stored geologically, and then converted back into electricity during
months of lower renewable electricity supply. Compared to batteries, hydrogen is a more
plausible solution for seasonal storage because investment costs are almost independent of
storage volume37 and ‘self-discharge’ is low (Parra et al, 2019).
From an economically efficient perspective, whether hydrogen emerges as a seasonal stor-
age mechanism will depend on the relationship between seasonal price differentials and the
capital costs of deploying electrolysers along with storage. German electricity price differen-
tials show that currently only for 5 percent of the time does the price differential (arbitrage
gain) exceed €50/MWh. The evolution of this potential for arbitrage gain will inter alia depend
on the deployment of renewable electricity generation sources and on the deployment of
flexible demand side resources.
36 Refined from biogas, which is produced through anaerobic digestion of waste or organic matter from a variety of
sources.
37 Most of the investment cost is related to the capacity (ie the MW) of the appliances that transform electricity into
hydrogen and back – while the size of the storage tanks/aquifers (ie the MWh) does not drive cost that much.
Figure 7: Price differential, lowest vs highest hourly prices in Germany, 2019
250
200
150
100
An intertemporal arbitrage
In about 75% of the hours, the maximum
of more than €50/MWh
intertemporal arbitrage was €15/MWh or less
50
0
0% 5% 9% 14% 18% 23% 27% 32% 37% 41% 46%
Source: Bruegel based on SMARD.
4 Overview of market dynamics

The current cost structure of hydrogen is based on its production from natural gas (methane).
But, as we have discussed, this supply is expected to be considerably transformed. The market
consensus is that the price of low-carbon hydrogen will decrease over the coming years,
largely driven by falling electrolyser costs (which can be reinforced by deployment policies
that allow economies of scale and learning). The extent of this cost decrease will determine
the competitiveness of low-carbon hydrogen for each end-use sector (Figure 8).
Figure 8: Stylised hydrogen demand scenarios

Low demand High demand
Supply at current
Supply at current cost structure
cost structure
Price of hydrogen
Price of hydrogen
Supply at improved
cost structure
Ammonia/methanol
Ammonia/methanol
Supply at improved
cost structure
Shipping
Shipping
Aviation
Aviation
Heating
Heating
Trucks
Trucks
Steel
Steel
Volume of hydrogen Volume of hydrogen
Source: Bruegel. Note: left panel shows a scenario with limited technological/commercial development on the demand side; right panel
shows a scenario with significant breakthrough in all demand sectors.
A further fundamental driver of supply costs will be the cost of fuel inputs – electricity
in the case of electrolysis. Expectations are that, unless there are breakthroughs in terms of
more flexible power demand/storage, there might be more hours with very high and very low
prices (Bossmann et al, 2018). Thus, there is potential for hydrogen costs to be further lowered
by running electrolysers only in hours when – thanks to abundant power from renewables
and/or low demand – electricity is particularly cheap. However, while the investment cost of
electrolysers remains high, they will have to run most of the year to justify their fixed costs.
Only when the fixed costs of electrolysers reduce enough will their use be economic in part-
load. But then they will start to push up electricity prices during exactly those hours where it
is economic to operate them. This will make additional renewables investments economically
viable and an equilibrium could develop. Electrolyser capacity in this equilibrium will be
determined not only by the cost of renewables and electrolysers, but also by the cost of com-
peting flexibility providers (eg batteries, demand-response). Thus, if batteries continue their
rapid pace of technological advancement, and/or innovation sees electricity demand become
increasingly flexible, it is still possible that the capital costs of electrolysers will be too high to
justify their part-load operation.
Non-EU countries are also investing in hydrogen production capacity. In some cases, this
involves cooperation with Europe, such as between Germany and Morocco (BMZ, 2020).
In other cases there is no European cooperation and hydrogen will potentially be traded on
international markets. The ability of third countries to produce hydrogen under more favour-
able conditions may exert downward pressure on European prices, although transport costs
would have to be factored in, as discussed in section 2.
The evolution of hydrogen demand within Europe is highly uncertain (section 3). But
whatever happens, a certain level of hydrogen demand is almost certain to remain, the
extent of which will depend on the demand for the end products: ammonia, methanol, crude
oil-derived products. In other sectors, demand for hydrogen will depend if hydrogen-utilising
technologies reach commercial maturity.
5 Policy options
The future prospects of hydrogen are highly uncertain. Currently, it is a chemical feedstock
but significant breakthroughs in production and end use could mean hydrogen might even
contribute 20 percent of the EU’s final energy demand in 2050. The challenge for policymak-
ers today is to assess the correct level and type of policy support in the context of this un-
certainty. We conclude with a discussion of some of the policy measures that could support
hydrogen deployment currently being debated.
Meaningful price on all greenhouse gas emissions

Tightening/extending the EU emissions trading system and re-thinking the design of energy
taxation systems
Higher prices on the use of fossil fuels help the competitiveness of all low-carbon tech-
nologies relative to fossil-fuel alternatives. In 2021 the European Commission will propose
to extend and tighten the EU emissions trading system (ETS) in line with tougher emission
reduction targets. Addressing the current taxation discrepancy between electricity and nat-
ural gas prices would be another no-regret option. From a carbon emissions standpoint, the
European taxation system currently biases consumption away from electricity and toward
natural gas38. The European Commission can address such discrepancies by reforming the EU
Energy Taxation Directive (2003/96/EC), which is also scheduled for 2021.
Supporting low-carbon hydrogen production

State support for the production of hydrogen with low carbon emissions
We classify this as a no-regret policy option. Decarbonising the production of current
hydrogen demand would already avoid approximately 100 Mt of CO2 emissions in the EU
per year. The wide range of sectors which could potentially use clean hydrogen suggests that
the benefits of decarbonising hydrogen production are likely to exceed those from current
demand only.
38 This DG ENER factsheet shows the discrepancies in taxation rates: https://ec.europa.eu/energy/sites/ener/files/

qmv_factsheet_on_taxes.pdf.
Moreover, until a low-carbon hydrogen source at scale is secured for Europe, there is
limited value in stimulating a massive ramp up in additional hydrogen demand, which would
be met by carbon-intensive production methods39. Supporting low-carbon hydrogen should
therefore be a policy priority.
The deployment of a significant volume of electrolysers should be supported to reduce
their cost. This could be done using tools that proved successful for wind and solar technology
(auctioning of feed-in premia). Policies to support the deployment of renewable electricity
generation to fuel growing demand from electrolysers would also be a no-regret option. The
deployment of other low-carbon hydrogen production should also be phased in when indus-
try is willing to share some of the remaining technology risk.
From a geopolitical standpoint, developing commercial know-how in technologies used
to produce clean hydrogen is likely to make Europe’s exports more competitive in a decar-
bonising world.
Supporting green products

State support for the production of low-carbon products, particularly in markets currently dom-
inated by emissions-intensive production
Focusing public support to the demand for low-carbon products and intermediate goods
(such as low-carbon steel) has the advantage of being technologically neutral. Markets would
be allowed to decide the most cost-efficient manner for production. Public revenue would
be spent only for products for which a clear carbon-emissions reduction has been achieved.
This would allow policymakers to adopt a neutral standpoint regarding the applicability of
hydrogen technologies, and to avoid public money being spent on projects that eventually do
not significantly reduce emissions.
The EU already has a tool for defining low-carbon benchmarks in the ETS product bench-
marks40. A challenge would be choosing which products to support, and how much to support
each product.
One drawback to this solution may be that one or two technologies are over-supported,
while other options are ignored. The question then arises of whether the state is able to
predict accurately which products and technologies should be supported. This is because
an explicit focus on decarbonising one sector prioritises technologies that are suitable for
that sector while not necessarily taking into account that support for a different technology
may have wider benefits for the rest of the economy. For example, a focus on decarbonising
heavy transport today might boost the competitiveness of new fuel cells and hydrogen tanks
that then could be used in light vehicles, trains and aircraft, while a focus on decarbonising
light vehicles today might instead extend the head start batteries have to all other modes of
transportation.
Supporting R&D
Support for hydrogen research and development
Europe invests too little into R&D in general (D’Andria et al, 2017). Public support for
low-carbon R&D is a no-regret option. However, prioritising support for different areas is
more controversial.
Many potential hydrogen applications would benefit from R&D investment. On the supply
side, a range of potential production pathways could be explored. Public support for increas-
39 There is an argument that it is still worthwhile pursuing demand cases today and that clean hydrogen supply
will eventually ‘catch up’. There is clear reason to this argument, but we believe that a clearer route must first be
established for the decarbonisation of hydrogen supply within Europe. Supplying the volume of clean hydrogen
suggested by our highest demand case would currently be very difficult for Europe.
40 Product benchmarks have been calculated for a range of emission-intensive products under the ETS. This
benchmark is based on the average greenhouse gas emissions associated with the best performing 10 percent
of installations. They can therefore be thought of as the best-practice emissions associated with production of a
particular product.
ing the number of potentially viable decarbonisation options would make the low-carbon
transition more resilient (eg if other technologies fail unexpectedly). Increased technology
competition is also important to exert pressure on dominant technologies (eg electric vehi-
cles) to invest in innovation based on specific criteria where alternative technologies still have
a lead (eg limited range).
There is a strong case for Europe to significantly scale up R&D for all decarbonisation
options. But, in a scenario of limited R&D budgets the value of hydrogen R&D must be
weighed against R&D in competing technologies or energy carriers.
A consistent and predictable support mechanism at the European level would be benefi-
cial. It could periodically allocate R&D funding to areas that appear most attractive according
to decarbonisation criteria and priorities. The mechanism would adapt to technological
evolution in order to avoid institutional lock-in41. It could take the form of an independent
public body, a European Energy Agency, which could provide policy advice to the European
Commission and interested member states. For example, future bottlenecks in the shift to
a low-carbon economy could be identified as a basis for today’s public R&D support. Such
a mechanism would help identify which hydrogen technological applications justify public
R&D.
Finally, in the context of the current economic crisis, a focus on creating jobs and high
multipliers might lead to an underappreciation of the merits of R&D for long-term economic
development.
Retrofitting natural gas networks

Public infrastructure investment to adapt the natural gas grid to make it suitable for
transporting hydrogen
The natural gas grid constitutes a significant infrastructure asset, capable of holding large
volumes of energy. At reasonable cost it could be repurposed for a low-carbon economy.
The necessity of repurposing the gas grid depends upon the size and geographic disper-
sion of demand clusters. If households are to consume significant volumes of hydrogen,
clearly repurposing is necessary and investment should slowly begin. However, it is not clear
if household-level hydrogen demand will ever materialise (section 3). Instead, our demand
analysis points to the likelihood of relatively significant hydrogen demand emerging in a
series of large industrial clusters (‘hydrogen valleys’) across Europe. In each cluster, hydro-
gen-using industries (eg ammonia, steel, carbon storage) would co-locate and share the costs
of hydrogen production or transmission. Therefore, a planning perspective would not place
much importance on building out a hydrogen distribution grid to the scale of anything like
that resembling the current natural gas infrastructure. Instead, a few transmission pipelines
connecting large demand and supply sources would be the priority investment.
Therefore, while it is technologically possible that hydrogen could satisfy household
energy demand, it is not a first-best solution. The challenge in the transition between two
network-based systems (eg gas-based heating to electricity-based heating) is, that at one stage
in the transition, the incumbent network will lose so many subscribers that its remaining sub-
scribers will bear too much of its fixed cost, leading them to unsubscribe at increasing speed.
Such an disorderly transition (which was seen for some central heating networks in eastern
Europe) can be inefficient and might need to be publicly managed.
In addition, policymakers should focus short-term planning (5-10 years) and regula-
tory activity on enabling industry to build the infrastructure necessary for a system of large
industrial hydrogen clusters. Decisions over whether to retrofit natural gas grids at a more
granular level should be postponed until clearer evidence emerges of the capacity of electrical
solutions to fully satisfy household energy demand.
Current market dynamics do not yet suggest that retrofitting natural gas grids to carry
hydrogen is a sensible public policy.
41 See Zachmann et al (2012) p 99, for further discussion.
Roll-out of hydrogen vehicle charging stations
State support for the deployment of hydrogen vehicle charging stations
Hydrogen vehicle charging stations are an enabling infrastructure. Providing the means to
refuel and operate hydrogen vehicles should stimulate private investment in the production
and purchase of hydrogen vehicles. Some pilots have already been supported (fewer than 100
in Germany).
However, significant public support for hydrogen charging stations would likely not be
sensible. As discussed in section 3, the case for a transition of most transport sectors to hydro-
gen appears weak when compared to the case for battery electric technology. There is a risk
that public support for hydrogen refuelling stations would be at the expense of public support
for electric charging stations.
European policymakers should continue to increase the stringency of decarbonisation
policies for the transport sector. As discussed, with higher carbon prices or tougher poli-
cies, hydrogen solutions may be viable for heavy vehicles. In such a future scenario, private
investment could cover the required charging stations (at either end of a trucking route, for
example). If private consortia come forward with co-financing options for publicly available
hydrogen charging stations, policymakers might consider offering small incentives, but this
should not be a landmark policy.
Hydrogen vehicle charging stations are not today a priority for public support.
Certification scheme for low-carbon hydrogen

Developing a system for robust classification of the carbon content for each MWh of hydrogen
Knowing the carbon emissions associated with the production of each MWh of hydrogen
will be an issue for future hydrogen consumption. Within Europe, calculations should not
be necessary because hydrogen production falls under the ETS, and so carbon emissions are
already priced in. But certification may be necessary for certifying the ‘greenness’ of hydrogen
imports.
Designing a robust classification system will be difficult. For electrolysis, this would
involve certifying the electricity input. When electricity for electrolysis is taken from the
public grid its carbon content is more a matter of definition/accounting, than an objective
value42. But even certifications of dedicated supplies from renewable electricity often do
not pass the additionality test: has new renewable electricity capacity been built exclusively
for hydrogen purposes, or has existing or already planned renewable capacity simply been
‘assigned’ to hydrogen production?
While a difficult task, European policymakers should think about designing a framework
for the international trade in clean hydrogen. The extent to which hydrogen will become an
internationally traded commodity remains to be seen, but if such a scenario emerges, Europe
is likely to be a significant net importer. It would be wise, therefore, to start the conversation
about how Europe can be sure its hydrogen imports are low carbon.
Competition policy/regulation holidays

Providing breaks from the rules of competition policy or regulation to encourage targeted
investment
Providing some temporary exemptions from strict competition/network regulation rules
designed for mature markets can be a tool for encouraging private sector buy-in. Horizontal
and vertical coordination are both crucial during the earlier stages of building a new net-
work. For example, initial investments in the production, transmission and consumption of
42 For electrolysis, it would require defining the carbon content of the used electricity and three very different values
can be used for each hour in which the electrolyser was used: cleanest power plant; average power plant; dirtiest
power plant or last (marginal) power plant required to meet the demand. In the short-term, the last option seems
most plausible, but in the longer term, additional demand from electrolysis might be met be increasing supply,
potentially from renewable sources.
hydrogen need to be well synchronised. Without the ability to ensure the provisioning of the
complementary elements of the hydrogen value chain (through vertical integration or bind-
ing agreements), private investment may be discouraged in some areas. There is a coordina-
tion problem, with investment into all elements needing to be synchronised because each
individual investment (eg an electrolyser) is only worthwhile if all other parts of the new value
chain (eg a hydrogen pipeline, storage or steel-plant) also work.
Additionally, regulatory breaks can help encourage breakthrough R&D and investment.
This is particularly the case for testing new technologies, such as the correct protocols for
using hydrogen in households.
In both cases, such exemptions must be temporary and well targeted so they encourage
investment in areas with high benefits.
References
Airbus (2020) ‘Airbus reveals new zero-emission concept aircraft’, press release, 21 September, available at
https://www.airbus.com/newsroom/press-releases/en/2020/09/airbus-reveals-new-zeroemission-
concept-aircraft.html
Bazzanella, A. and F. Ausfelder (2017) Low carbon energy and feedstock for the European chemical
industry, DECHEMA, available at https://dechema.de/dechema_media/Downloads/
Positionspapiere/Technology_study_Low_carbon_energy_and_feedstock_for_the_European_
chemical_industry.pdf
Bellona Europa (2020) ‘Factsheet: Pros and Cons Seaweed for Biofuel’, available at: https://bellona.org/
assets/sites/3/2017/03/FACTSHEET-seaweed-for-energy.pdf
Bertelsen, N. and B. Vad Mathiesen (2020) ‘EU-28 Residential Heat Supply and Consumption: Historical
Development and Status’, Energies, 13(8):1894, available at https://doi.org/10.3390/en13081894
BMZ (2020) ‘Securing a global leadership role on hydrogen technologies: Federal Government adopts
National Hydrogen Strategy and establishes National Hydrogen Council’, press release, 10 June,
German Federal Ministry for Economic Cooperation and Development, available at http://www.bmz.
de/20200610-1en
Bossmann, T., L. Fournié and G.P. Verrier (2018) ‘Wholesale market prices, revenues and risks for
producers with high shares of variable RES in the power system’, METIS Studies S14, European
Commission, available at https://ec.europa.eu/energy/sites/ener/files/documents/metis_s14_
electricity_prices_and_investor_revenue_risks_in_a_high_res_2050.pdf
Catalá, F., R. Flores De La Fuente, W. Gardzinski and J. Kawula (2013) Oil refining in the EU in 2020, with
perspectives to 2030, CONCAWE report 1/13R, available at https://www.concawe.eu/wp-content/
uploads/2017/01/rpt_13-1r-2013-01142-01-e.pdf
Cihlar, J., A. Villar Lejarreta, A. Wang, F. Melgar, J. Jens, and P. Rio (2020) Hydrogen generation in Europe:
Overview of key costs and benefits, European Commission, available at https://op.europa.eu/en/
publication-detail/-/publication/7e4afa7d-d077-11ea-adf7-01aa75ed71a1/language-en
D’Andria, D., D. Pontikakis and A. Skonieczna (2017) ‘Towards a European R&D Incentive? An
assessment of R&D Provisions under a Common Corporate Tax Base’, JRC Working Papers on Taxation
and Structural Reforms No. 03/2017, European Commission, Joint Research Centre (JRC), Seville,
available at https://www.econstor.eu/bitstream/10419/202249/1/jrc-wptsr201703.pdf
Donat, L. (2020) ‘Connecting Europe with a Rail Renaissance: Eight measures to revive the European
rail system’, GermanWatch, available at https://germanwatch.org/sites/default/files/Connecting%20
Europe%20with%20a%20Rail%20Renaissance_2.pdf
Durakovic, A. (2020) ‘Ørsted and Yara Form Green Ammonia Pact’, offshoreWIND.biz, 5 October,
available at https://www.offshorewind.biz/2020/10/05/orsted-and-yara-form-green-ammonia-pact/
European Commission (2018) In-Depth Analysis in Support of the Communication COM(2018) 773: A
Clean Planet for All. A European long-term strategic vision for a prosperous, modern, competitive
and climate neutral economy, European Commission, available at https://ec.europa.eu/clima/sites/
clima/files/docs/pages/com_2018_733_analysis_in_support_en_0.pdf
European Commission (2020) ‘A hydrogen strategy for a climate-neutral Europe’, COM(2020) 301 final,
available at https://ec.europa.eu/energy/sites/ener/files/hydrogen_strategy.pdf
FCH JU (2019) Hydrogen Roadmap Europe, Fuel Cells and Hydrogen Joint Undertaking, available at
https://www.fch.europa.eu/sites/default/files/Hydrogen%20Roadmap%20Europe_Report.pdf
Hanley, E., J.P. Deane and B.O. Gallachóir (2017) ‘The role of hydrogen in low carbon energy futures - A
review of existing perspectives’, Renewable and Sustainable Energy Reviews 82(3): 3027-3045, available
at https://www.sciencedirect.com/science/article/abs/pii/S1364032117314089
Hoffmann, C., M. Van Hoey and B. Zeumer (2020) ‘Decarbonization challenge for steel’, McKinsey,
3 June, available at https://www.mckinsey.com/industries/metals-and-mining/our-insights/
decarbonization-challenge-for-steel
IEA (2019) The Future of Hydrogen, International Energy Agency, available at https://www.iea.org/
reports/the-future-of-hydrogen
IEA (2019a) IEA G20 Hydrogen report: Assumptions, International Energy Agency, available at https://
iea.blob.core.windows.net/assets/a02a0c80-77b2-462e-a9d5-1099e0e572ce/IEA-The-Future-of-
Hydrogen-Assumptions-Annex.pdf
IEA (2020) ETP Clean Energy Technology Guide, International Energy Agency, available at https://www.
iea.org/articles/etp-clean-energy-technology-guide
Kayfeci, M., A. Kecebas and B. Mutlucan (2019) ‘Chapter 3 - Hydrogen production’, in F. Calise, M. Dentice
D’Accadia, M. Santarelli, A. Lanzini and D. Ferrero (eds) Solar Hydrogen Production, Processes,
Systems and Technologies, Elsevier, available at https://www.sciencedirect.com/science/article/pii/
B9780128148532000035?via%3Dihub
Kopp, M., D. Coleman, C. Stiller, K. Scheffer, A. Aichinger and B. Scheppat (2017) ‘Energiepark
Mainz: Technical and economic analysis of the worldwide largest Power-to-Gas plant with PEM
electrolysis’, International Journal of Hydrogen Energy 42(19): 13311-13320, available at https://www.
sciencedirect.com/science/article/abs/pii/S0360319917300083
Korfiati, A., C. Gkonos, F. Veronesi, A. Gaki, S. Grassi, R. Scenkeln, S. Volkwein, M. Raubal and L. Hurni
(2016) ‘Estimation of the Global Solar Energy Potential and Photovoltaic Cost with the use of Open
Data’, International Journal of Sustainable Energy Planning and Management 9: 17-30, available at
https://doi.org/10.5278/ijsepm.2016.9.3
Lizza, R. (2003) ‘The Nation: The Hydrogen Economy; A Green Car That the Energy Industry Loves’, The
New York Times, 2 February, available at https://www.nytimes.com/2003/02/02/weekinreview/the-
nation-the-hydrogen-economy-a-green-car-that-the-energy-industry-loves.html
Middlehurst, C. (2020) ‘Ammonia flagged as green shipping fuel of the future’, Financial Times, 29 March
Nikolaidis, P. and A. Poullikkas (2017) ‘A comparative overview of hydrogen production processes’,

Renewable and Sustainable Energy Reviews, 67: 597-611, available at https://www.sciencedirect.com/
science/article/abs/pii/S1364032116305366
Parra, D., L. Valverde, F. Javier Pino and M. Patel (2019) ‘A review of the role, cost and value of hydrogen
energy systems for deep decarbonisation’, Renewable and Sustainable Energy Reviews 101(3): 279-294,
available at https://www.sciencedirect.com/science/article/abs/pii/S1364032118307421
Pavlenko, N., S. Searle and A. Christensen (2019) ‘The cost of supporting alternative jet fuels in the
European Union’, Working Paper 2019-05, The International Council on Clean Transportation,
available at: https://theicct.org/sites/default/files/publications/Alternative_jet_fuels_cost_
EU_20190320.pdf
Piebalgs, A., C. Jones, P. Carlo Dos Reis, G. Soroush and J.M. Glachant (2020) Cost-Effective
Decarbonisation Study, Florence School of Regulation, available at https://fsr.eui.eu/
publications/?handle=1814%2F68977
Quarton, C., O. Tlili, L. Welder, C. Mansilla, H. Blano, H. Heinrichs … S. Samsatli (2020) ‘The curious
case of the conflicting roles of hydrogen in global energy scenarios’, Sustainable Energy Fuels 4(80),
available at: https://doi.org/10.1039/C9SE00833K
Tounquet, F. and C. Alaton (2020) Benchmarking smart metering deployment in the EU-28, Tractabel
Impact and European Commission, available at https://www.buildup.eu/sites/default/files/content/
mj0220176enn.en_.pdf
Wood Makenzie (2020) ‘Hydrogen production costs: is a tipping point near?’ available at: https://www.
woodmac.com/our-expertise/focus/transition/hydrogen-production-costs-to-2040-is-a-tipping-
point-on-the-horizon/

Navigating through hydrogen

Uploaded by

Copyright:

Available Formats

Navigating through hydrogen

Uploaded by

Document Information

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

Navigating through hydrogen

Uploaded by

Copyright:

Available Formats

Policy Contribution

Issue n˚08/21 | April 2021

Ben McWilliams and Georg Zachmann

Policymakers must address the need to displace carbon-intensive hydrogen with

Figure 1: Estimated variation in hydrogen demand in 2050

0 100 200 300 400 500 600 700

2 Policy Contribution | Issue n˚08/21 | April 2021

Figure 2: Low-carbon hydrogen production

Scientiﬁc research Prototype/demonstration Hydrogen

3 Policy Contribution | Issue n˚08/21 | April 2021

4 Policy Contribution | Issue n˚08/21 | April 2021

Figure 3: Hydrogen price for different electricity and carbon prices

Hydrogen cost, €/MWh

5 Policy Contribution | Issue n˚08/21 | April 2021

Table 1: Additional low-carbon hydrogen production methods

2.2 Hydrogen imports

6 Policy Contribution | Issue n˚08/21 | April 2021

Figure 4: Import vs domestic hydrogen (€/MWh)

0 20 40 60 80 100 120 140

7 Policy Contribution | Issue n˚08/21 | April 2021

Steelmaking 4% ‫ۻۻۻۻۻ‬ High potential to replace coal

Electric vehicles hold first-mover advantage in low

Heavy duty vehicles 5% ‫ۻۻۻ‬ Hydrogen more suited to heavier vehicles

Shipping 7% ‫ۻۻۻۻ‬ Potential additional demand via use as ammonia

Aviation 4% ‫ۻۻۻ‬ Synthetic fuels; fuel cells

Residential heating 12% ‫ۻۻ‬ Competing with electricity

Figure 5: GHG emissions share in transport sector

Heavy duty vehicles 17.01% Light commercial vehicles 7.65%

8 Policy Contribution | Issue n˚08/21 | April 2021

9 Policy Contribution | Issue n˚08/21 | April 2021

10 Policy Contribution | Issue n˚08/21 | April 2021

11 Policy Contribution | Issue n˚08/21 | April 2021

Figure 6: EU industrial greenhouse gas emissions

Oil reﬁning 11%

Chemical sector: ammonia and methanol

Over 80 percent of ammonia produced worldwide is for the manufacture of fertilisers

12 Policy Contribution | Issue n˚08/21 | April 2021

Similarly to ammonia, demand for hydrogen in methanol production is predominantly

13 Policy Contribution | Issue n˚08/21 | April 2021

14 Policy Contribution | Issue n˚08/21 | April 2021

3.3 Residential heating

15 Policy Contribution | Issue n˚08/21 | April 2021

3.4 Hydrogen as an enabler of renewable electricity deployment

Short-term ﬂexible demand

Long-term seasonal storage

16 Policy Contribution | Issue n˚08/21 | April 2021

Source: Bruegel based on SMARD.

4 Overview of market dynamics

Figure 8: Stylised hydrogen demand scenarios

Volume of hydrogen Volume of hydrogen

17 Policy Contribution | Issue n˚08/21 | April 2021

Meaningful price on all greenhouse gas emissions

Supporting low-carbon hydrogen production

38 This DG ENER factsheet shows the discrepancies in taxation rates: https://ec.europa.eu/energy/sites/ener/files/

18 Policy Contribution | Issue n˚08/21 | April 2021

Supporting green products

19 Policy Contribution | Issue n˚08/21 | April 2021

Retroﬁtting natural gas networks

41 See Zachmann et al (2012) p 99, for further discussion.

20 Policy Contribution | Issue n˚08/21 | April 2021