EYE ON THE MARKET | 14TH ANNUAL ENERGY PAPER

Electravision
Electravision. The predominant vision for the future involves the electrification of everything, powered
by solar, wind, transmission and distributed energy storage. This vision primarily relies upon the
greater efficiency of electric motors and heat pumps vs their fossil fuel counterparts. While the grid is
getting greener, electrification is advancing at a much slower pace for reasons related to chemistry,
physics, cost, politics and human behavior. Our 14th annual energy paper takes a closer look, and also
includes sections on nuclear power, China, hydrogen, “net zero oil” and Gaza’s energy future.

By Michael Cembalest | Chairman of Market and Investment Strategy for J.P. Morgan Asset & Wealth Management
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper

Electravision: 14th Annual Eye on the Market Energy Paper

The fossil fuel share of global energy use is falling at ~0.40% per year as the renewable transition progresses.
That’s almost exactly the same pace of decarbonization that occurred from 1973 to 1988 during the nuclear
buildout. To be clear, global CO2 emissions have not declined since energy consumption keeps rising; what’s
falling is the share of primary energy from fossil fuels, not their level.
The fossil-renewable gap in the second chart1 should close at a faster pace given growing global initiatives to
decarbonize; global transition spending has exceeded fossil fuel spending for the fourth year in a row and the
gap is widening2. Looking ahead, the fiscal costs of the US energy bill could reach $900 bn by 2030 and $1.1
trillion by 20353. But as things stand now, renewable energy is used almost exclusively to decarbonize the grid
and its main purposes: space cooling, lighting, refrigeration, data centers, electronics and some space heating.
The grid’s use for industrial production and transport is still small.
The consensus path forward is Electravision, the electrification of everything. The reason: if something can be
electrified, it can eventually be decarbonized via wind, solar and energy storage. While this transition is
underway, it will take time due to chemistry, physics, cost, human behavior and politics. As a result, current
human prosperity is difficult to imagine without substantial contributions from natural gas. This gas ecosystem
needs sufficient investment to avoid electricity and natural gas outages, and its methane footprint needs greater
attention (a topic covered last year). This year’s paper gets into the details of Electravision along with sections
on nuclear power, China, “decarbonized oil”, levelized costs, hydrogen, bio-oil, EV emissions, the latest from
Vaclav Smil and concluding thoughts on solar power and Gaza’s energy future. The appendices include data on
global oil and LNG markets, and a travelogue of EV misadventures with Electrify America, Waymo and Jeep.
Falling fossil fuel shares mask reality of rising emissions Global fossil fuel and renewable energy use
Percent Billion tonnes of CO2 Exajoules
94% 40 500
Global fossil fuel share
92% of primary energy 35 450
Fossil fuels,
400 primary energy
30
90% 350
Global emissions
from energy 25
88% 300
20 250
86% Fossil fuels,
15 200 post combustion
84% 150 energy
10
100
82% 5 Renewable energy
50
80% 0 0
1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020
Source: EI Statistical Review of World Energy, JPMAM, 2023 Source: EI Statistical Review of World Energy, IEA, JPMAM, 2024

1
The standard convention in energy analysis (IEA, EI, EIA) is to show renewable and nuclear power on an “input-
equivalent” or “thermal-equivalent” basis, grossing these electricity sources up to make their primary energy
figures more comparable to fossil fuels which experience ~62% thermal losses during combustion. The EIA is
moving to a new standard: they will show renewable and nuclear power without any gross-up, directly reflecting
their electrical output whether in TWh, exajoules or BTUs.
The first chart uses the standard primary energy figures. The second chart shows primary energy of global fossil
fuel use, and another series we compute on “post-combustion fossil fuel energy”. This is derived based on the
actual use of oil, natural gas and coal for transport, electricity generation and industrial/building heat and reflect
their combustion efficiencies (30%, 38% and 90% respectively). This latter series is more comparable to
renewable and nuclear energy since it gets closer to the amount of “work” that electrical energy can perform.
2
“Energy Transition Investment Trends 2024”, BNEF, January 2024, Page 7
3
“Economic Implications of the Climate Provisions of the Inflation Reduction Act”, Bistline et al, citing EPRI’s
REGEN model, March 30, 2023
1
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper

The foundations of this piece are rooted in the science of the transition, rather than in the imaginary ways that
the transition might work. Anyone who supports a clear-eyed analysis should be disturbed when science is
distorted for political reasons:
• According to the “Silencing Science” tracker maintained by Columbia University, 346 anti-science actions
were taken by the federal government during Trump’s administration. Categories include gov’t censorship,
self-censorship, budget cuts, personnel changes, research hindrance and bias/misrepresentation
• Nearly 400 EPA scientists said they observed violations of the agency’s scientific integrity policy in the
second half of 2018 but did not report them due to “fear of retaliation, belief that reporting would make no
difference, perceived suppression or interference by Agency leadership, and belief that politics and policy
outweigh science”4
• Trump has pledged to revive a plan to reclassify thousands of federal employees5. These include scientists
who are currently shielded from politics in permanent professional positions. The plan would allow his
administration to more easily fire “rogue bureaucrats” that oppose his agenda. The administration could
appoint replacements who are aligned with Trump politically regardless of scientific or technical expertise
Sorting out fact from fiction regarding the energy transition is hard enough without this kind of thing. There are
legitimate debates about the right energy transition policies. As one example, the EPA will finalize standards
this spring for existing coal and new natural gas plants that might require carbon capture as a precondition for
approval to operate. One can debate the cost and technological readiness of this constraint. But the scientists
that work on empirical analysis of energy and the environment should be left alone.
Michael Cembalest
JP Morgan Asset Management
Acronyms
BEV battery electric vehicle; BNEF Bloomberg New Energy Finance; BTU British thermal unit; CHP combined heat and
power; CCS carbon capture and storage; CF capacity factor; DACC direct air carbon capture; DOE Department of Energy;
DRI direct reduced iron; EI Energy Institute; EIA Energy Information Administration; EOR enhanced oil recovery; EPA
Environmental Protection Agency; EV electric vehicle; FERC Federal Energy Regulatory Commission; FF fossil fuel; GHG
greenhouse gas; GW gigawatt; HGB hydropower, geothermal and biomass; HVAC heating, ventilation and air conditioning;
ICE internal combustion engine; IEA International Energy Agency; IRENA International Renewable Energy Agency; kg
kilogram; km kilometer; kW kilowatt; kWh kilowatt-hour; LBNL Lawrence Berkeley National Laboratory; LCOE levelized cost
of energy; Li-Ion lithium-ion; LNG liquefied natural gas; MJ megajoule; MPG miles per gallon; mtpa million tons per annum;
MW megawatt; MWh megawatt-hour; NERC North American Electric Reliability Corporation; NRC Nuclear Regulatory
Commission; NREL National Renewable Energy Laboratory; PHEV plug-in hybrid electric vehicle; Quad quadrillion BTU; REE
rare-earth element; SMR small modular reactor; TTF Title Transfer Facility; TWh terawatt-hour; UCS Union of Concerned
Scientists; UNCTAD United Nations Conference on Trade and Development; USGS US Geological Survey. [Version Mar 26].

4
“How President Trump’s war on science impacted public health and environmental regulation”, Webb and
Kurtz, in “Progress in Molecular Biology and Translational Science”, 2020
5
“Trump’s presidential push renews fears for US science”, Nature, Jeff Tollefson, January 2024
2
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper

Table of contents
Electravision: the contours of an electrified future ................................................................................................4
The latest from Vaclav Smil ................................................................................................................................ 15
All Eyes on China: the world’s largest energy user is building a lot of everything .................................................. 16
Essential energy transition charts ....................................................................................................................... 18
Investment returns in oil & gas, renewable energy and nuclear-exposed shares .................................................. 22
Nuclear power: Elusive measures of true cost, German decommissioning/deindustrialization and NY State ......... 24
Lazard’s epiphany does not go far enough: the inadequacies of levelized cost, Part 2 .......................................... 29
Advanced Electravision topics: timing, temperature, transmission and turbines .................................................. 30
Why-drogen Epilog: A tough year for the green hydrogen industry ...................................................................... 35
“Ignore the PUNIs”: The distraction of small countries with unique natural energy resources .............................. 38
On “Net Zero” Oil: worth trying if it can be done close to very ambitious pro-forma projections .......................... 40
A proper burial: can sequestration of bio-oil gain more traction than CCS? .......................................................... 44
What if? A thought experiment on the reconstruction of Gaza and the role of distributed solar power ................ 46
Appendix I: on US and European energy supply/demand and global LNG markets ............................................... 50
Appendix II: EV misadventures with Electrify America, Waymo and Jeep ............................................................. 52

Prior year energy topics

On our Eye on the Market energy archive page, you can read about topics we covered last year:
• the US electricity transmission quagmire and the growing interconnection queue of wind/solar projects
• the flawed concept of Levelized Cost when applied to renewable energy
• a deep dive on critical minerals, geological abundance and China’s role in energy supply chains
• the high cost of biofuel replacements for traditional jet fuel
• benefits and limitations of municipal solid waste, and the controversy in Europe over solid biomass
• the slow pace of carbon sequestration and the related issue of flue gas CO2 concentrations
• the dreamlike reveries of electric planes, nuclear fusion, space-based solar power, direct air carbon capture
and fully autonomous passenger car networks
• field studies on US methane emissions showing higher levels than those reported to the EPA

3
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Electravision
Executive Summary
14th Annual Energy Paper

Electravision: the contours of an electrified future

There are two main reasons for electrification: (a) if you can electrify it, you can decarbonize it; and (b) electric
motors and electric heat pumps are more efficient than their fossil fuel counterparts. In most countries,
electrification accounts for 20%-30% of total energy consumption. There’s a lot of press coverage on countries
with higher shares; but more people talk about Norway than there are people living in Norway, and the three
countries shown on the right are not good proxies for the rest of the world. If there’s something notable going
on, it’s rising electrification in China.
Major countries: electricity share of final energy consumption Most electrified: electricity share of final energy consumption
Percent Percent
32% 70%
Japan Iceland
28% France 60% Norway
China Sweden
24% 50%
India Global
20% Brazil 40%
Europe
16% 30%
US
12% Russia 20%

8% 10%
1990 1995 2000 2005 2010 2015 2020 1990 1995 2000 2005 2010 2015 2020
Source: EI Statistical Review of World Energy, JPMAM, 2023 Source: EI Statistical Review of World Energy, JPMAM, 2023

What is electricity used for? Mostly for residential-commercial-industrial space cooling, lighting, data centers,
computer equipment, ventilation and some space heating. The use of electricity in transportation and industrial
sectors is generally very small, and thermal heat is still the dominant form of space heating. That explains the
chart on the right: grid decarbonization (green line) is happening faster than electrification (blue line). In
other words, the grid is getting greener but the purposes for which it’s used are expanding much more slowly.
US electricity uses: primarily HVAC Grid decarbonization outpaces electrification of energy
Quadrillion BTUs use, global, Percent
35%
5 Renewable share of global
electricity generation
30%
4

25%
3
Space heating, water heating, motors and
2
process heat 20%

Space cooling, lighting, ventilation,

1 refrigeration, computers and other 15% Electricity share of global
electronic equipment final energy consumption
0 10%
Residential Commercial Industrial 1990 1995 2000 2005 2010 2015 2020
Source: EIA, JPMAM. 2023. Transport too small to plot at 0.06 quads. Source: EI Statistical Review of World Energy, IEA, JPMAM, 2023

4
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Electravision
14th Annual Energy Paper Executive Summary

How Electravision would work. Each country has a different roadmap but many details are similar to the US
version outlined below6: the grid is decarbonized further while at the same time, electrification of road
transport, winter heating and industrial production increase as well. The big picture: after accounting for
efficiency of electric motors and electric heat pumps, only 8-9 quads of electrical energy would be needed to
replace 27 quads of fossil fuel energy, with the latter decreasing by 34% from current levels.
The challenges: this would require a ~34% increase in US electricity generation (i.e., the same % increase in
power generation that took place from 1993 to 2022, a period of 30 years), a ~400% increase in wind and solar
power and enough backup thermal power and battery storage to handle 53% of a much larger grid coming from
intermittent renewables. The next few pages walk through each step; all increases in electrification are assumed
to be powered by new wind and solar. Electrification makes less sense from a decarbonization perspective if
powered by additional natural gas.
How much would Electravision cost? I have no idea; a thorough assessment would have to include increased
costs to both ratepayers and taxpayers. For example, it would need to incorporate the impact on US power
prices (since 2019, power prices have been rising faster than core inflation, food prices and gasoline prices) and
the incremental costs of subsidies to taxpayers (e.g., the $800 bn to $1.1 trillion cited on page 1).
Electravision roadmap for the US, Quadrillion BTUs of primary energy, new EIA convention
(primary energy for fossil fuels, delivered energy for renewables and nuclear)
0 10 20 30 40 50 60 70 80 90

Current Coal Nat Gas Oil

Decarbonization of the current grid via more wind and solar

Step 1

Decarbonization of road transport via EV cars and trucks

Step 2

Decarbonization of winter heating via residential and commercial heat pumps

Steps 3/4

Decarbonization of industrial production via industrial heat pumps

Step 5 Wind/Solar Other REN Nuclear

Source: EIA data, JPMAM assumptions, 2024. Other renewables is primarily hydropower

Current Electravision Change

Primary energy delivered (quads) 85.4 66.5 -22%
Fossil fuel primary energy use (quads) 79.4 52.2 -34%
Wind+solar generation (quads) 2.2 10.5 376%
Wind+solar generation (TWh) 645 3,069 376%
Total electricity generation (quads) 14.7 19.7 34%
Total electricity generation (TWh) 4,321 5,775 34%
Wind+solar share of electricity generation 15% 53% 38%
Renewable share of electricity generation 23% 59% 36%
Source: JPMAM, 2024. Using new EIA conventions for primary energy.

6
Electrification assumptions differ. One example: “Electrify: An optimist’s playbook for our clean energy future”
by Saul Griffith from MIT. In an October 2023 article, Griffith cites possibilities for electrification of short haul
aviation which is outside the scope of our analysis. We do not assume any electrification of shipping or aviation.
In last year’s paper we explained how short haul aviation trips (less than 200 miles) account for less than 5% of
total aviation emissions, so even if we had included it, it would not have made much of an impact.
5
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Electravision
14th Annual Energy Paper Executive Summary

# 1: Decarbonization of the US grid via more wind and solar power, before increased electrification
• Current: wind = 10% of generation, solar = 5% of generation
• Electravision: wind share increases to 19%, solar share increases to 18%; curtailment of 5%
• Coal share of generation falls from 20% to 5% with remainder coming from reduction in natural gas
The first stage of Electravision would be more wind and solar on the existing grid. Keep in mind that this may
not result in widespread reductions of natural gas capacity. “Capacity credits” estimate the MW of gas capacity
that can be disconnected for every MW of wind and solar added to the grid, and are just 10%-25% in the US (see
page 18). In other words, the total cost of Electravision includes redundancy of generation capacity in addition
to the cost of new generation capacity, transmission and storage. Note how Germany’s thermal power has not
declined despite large additions of wind and solar capacity since 2002. German CO2 emissions and thermal
capacity factors have declined, but its thermal plants still must be built and maintained.
In the US, pumped storage accounts for 70% of utility-scale energy storage while chemical batteries account for
another 28%. Within chemical batteries, almost the entire amount is 4-6 hour li-Ion capacity. If wind/solar
buildouts increase and natural gas peaker plants are decommissioned, long duration energy storage (LDES) will
be a critical part of Electravision since 4-6 hour li-ion battery storage will not be enough. New chemical LDES
battery approaches such as vanadium redox and sodium sulfur are just beginning to be deployed alongside
compressed air storage. But to be clear, global LDES capacity is still very small, just ~0.5% of all forms of energy
storage. Low LDES capacity is due to high costs, low technological readiness and the need for improved round-
trip efficiencies (i.e., compressed air 40%-60%, flow batteries 65%-80% vs li-Ion at ~90%).
Wind and solar share of electricity consumption Long duration energy storage by technology and by region
Percent 1.4 GW total
28% 100% Others
Europe Others
90%
24% Sodium UAE
80% sulfur battery
20% 70% US

60% Vanadium
16% China Japan
US redox flow
50% battery
12% Germany
40%
8% 30% Compressed
EM ex.
China 20% air energy
4% storage China
10%
0% 0%
1990 1995 2000 2005 2010 2015 2020 By technology By region
Source: EI Statistical Review of World Energy, IEA, JPMAM, 2023 Source: "2023 Global Long-Duration Energy Storage Update", BNEF, Sept 2023

Germany installed capacity of wind, solar and thermal Germany thermal capacity factors
power, Gigawatts Percent
80 80%

70
Wind 70%
Thermal Power
60
(Coal and Natural Gas) Coal
Solar 60%
50

40 50%

30
40%
20
30%
10 Natural Gas
0 20%
2002 2005 2008 2011 2014 2017 2020 2023 2002 2005 2008 2011 2014 2017 2020
Source: Fraunhofer Institute, JPMAM, 2024 Source: EI Statistical Review, Fraunhofer Institute, JPMAM, 2023

6
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Electravision
14th Annual Energy Paper Executive Summary

Step 2: Decarbonization of US passenger cars, light trucks and Class 8 trucks: growing but still in its infancy
• Current fleet: EVs represent 1.5% of passenger cars and light duty vehicles, and 0.6% of Class 8 trucks
• Electravision: 50% of passenger car energy petroleum use displaced by EVs, assuming 23 MPG for ICE cars/
light trucks and 0.33 kWh per mile for EVs; 50% of freight and commercial light truck energy petroleum use
displaced by EVs assuming 6.5 MPG for ICE trucks and 1.7 kWh per mile for EVs7
US EV sales are now ~10% when including BEVs and PHEVs8, but they’re just 1.5% of the car fleet and even less
as a share of the Class 8 truck fleet; it takes time to change the fleet when vehicles last for 12 years or more.
More states are following California’s lead to phase out internal combustion engine cars; states comprising one
third of the existing car fleet plan to do so. EV capital commitments by manufacturers also suggest that the
future of road vehicle transport is electric, but I wonder how long it will take.
BNEF estimates that the US fleet will be 50% EVs by 2037, but that requires accelerated adoption from current
levels. I wonder what BNEF would say about the following: US EV inventories on dealer lots reached an all-time
high at 114 days in December 2023, double the figure from the prior year (these figures exclude Tesla and Rivian
which sell direct to consumers). Ford and GM are cutting back on EV production, GM and Honda abandoned a
plan to build lower-priced EVs, and a consortium of 4,000 auto dealers asked Biden to “tap the brakes” on EV
mandates. In other words, Electravision EV forecasts are very optimistic and could take a LONG time.
US EV sales as a percent of total light-duty vehicle sales Share of US car fleet following California's fuel economy
Percent standards, Percent
9% Non Plug-In
8% Hybrid (HEV) CA and states
following its ICE-
7% phase out
6%
31%
5%
4%
Battery 59%
3% Electric
(BEV) 10%
2% Plug-In
States not
1% Hybrid
following CA States following CA
(PHEV)
0% but non-committal
2011 2013 2015 2017 2019 2021 2023
Source: ANL, JPMAM, January 2024 Source: BloombergNEF, December 2023

R&D and capex commitments for EVs and digital tech EV share of US vehicle fleet as per BNEF projections
Percent of total R&D and capex Percent of fleet, BEV plus PHEV
80% 70%

70% 60%

60% 50%
50%
40%
40%
Mercedes-Benz

30%
Volkswagen

30%
20%
Stellantis

20%
Nissan

Toyota

10%
Ford

10%
GM

0% 0%
2015 2020 2025 2030 2035 2040
Source: BloombergNEF, December 2023 Source: BloombergNEF, JPMAM, June 2023

7
ICE vehicles: 23 mpg for US vehicle fleet (DoT); 6.5 mpg for semi-truck (Phoenix Truck Driving Institute); EVs:
0.33 kWh per mile for Tesla Model X (DoE); 1.7 kWh per mile for Tesla Semi Class 8 all-electric semi-truck (Tesla)
8
Non plug-in HEVs like the Toyota Prius are generally not included in broad EV totals since they only achieve
~20% reduction in emissions per mile compared to ICE cars, with battery ranges of just 1-3 miles
7
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Electravision
14th Annual Energy Paper Executive Summary

Steps 3 and 4: Decarbonization of winter heating via residential and commercial air-sourced heat pumps
Current share of heating use
Resistance Future heat Coefficient of Furnace
Heat pump Fossil fuels
heating pump share performance efficiency
Residential 6% 74% 20% 50% 3.0 90%
Commercial 11% 75% 14% 65% 3.0 80%
Source: "Transitioning to Heat Pump Rooftop Units for US Commercial Buildings", NREL; JPMAM May 2023

In an Electravision scenario, 50%-65% of winter heating energy is met via electric heat pumps, up from 6%-11%
today. Heat pumps are the marginal source of winter heating in Europe and the US, but the challenge is the
same as with vehicles: the time required to replace the stock of combustion devices with electrical ones. New
home sales in the US and Europe are less than 1% of the existing housing stock. So, if heat pump sales are
mostly confined to new homes or to replace furnaces when they stop working (their operating lives are 20+
years), this transition will take a really long time. Also: the Ninth Circuit overruled Berkeley’s natural gas ban in
new buildings after concluding that it conflicts with Federal law; the impact of this decision could be material.
The fourth chart shows another hurdle: cost. Natural gas is cheaper than electricity per unit of energy, offsetting
heat pump efficiency benefits for homeowners. In other words, if a heat pump coefficient of performance9 is
3x but electricity costs 3x more than natural gas, there might be limited economic benefits to heat pump
adoption. Heat pump switching incentives are greater for heating oil and propane users (gold dot).
On Europe: heat pump adoption means that many buildings will have air conditioning for the first time, which
could reduce Europe’s net energy and emissions gains (see page 33) from heating electrification.
US heat pump sales exceed gas furnace sales Germany heating structure in new residential buildings
Annual sales, millions Share of new residential buildings
4.0 100%
Heat pump 90%
3.5 80%
70%
3.0
60% Natural gas or
heating oil
2.5 Gas furnace 50%
40%
2.0 30%
20%
1.5 Electric heat
10% pumps
1.0 0%
2009 2011 2013 2015 2017 2019 2021 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022
Source: Air Conditioning, Heating and Refrigeration Institute, 2023 Source: AG EnergieBilanzen, JPMAM, Q2 2023

The slow pace of housing turnover Winter price for electricity vs fossil fuels for heating
Percent Price of electricity per MJ / price of nat gas/heating oil/propane per MJ
1.0% 5.5x
elec vs natural gas
5.0x
0.8% elec vs heating oil/propane
4.5x
1 standard deviation +/-
4.0x
0.6%
3.5x
3.0x
0.4% US EU
2.5x

0.2% 2.0x
1.5x
0.0% 1.0x
New single family home sales as % Completed new homes as IL NY OH CA PA TX VA GA NC FL US
of single family housing stock % of existing housing stock Source: EIA, JPMAM, 2023. Top 10 states by electricity consumption;
Source: Census, OECD, JPMAM, 2023 assuming 90% gas furnace efficiency. Residential pricing.

9
For more information on heat pump coefficient of performance, see this page from our 2022 energy paper
8
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Electravision
14th Annual Energy Paper Executive Summary

Step 5: Decarbonization of industrial production via industrial air- or ground-sourced heat pumps
• Current: 14% of industrial sector energy use is sourced from electricity
• Electravision: 30% sourced from electricity, assuming industrial heat pump coefficient of performance = 2.0
The electricity share of US industrial energy use has been stable for over 30 years, stuck at ~15%. According to
NREL, at least half of all industrial process heat requires temperatures of 200⁰C or less. If that’s the case, why
doesn’t the industrial sector take more advantage of industrial heat pumps that require less energy than
furnaces? In other words, why isn’t electrification’s share of industrial energy use rising? Optimists believe
we’re on the cusp of a major transition to industrial heat pumps. Perhaps, but in the absence of any signs of
change, a doubling of industrial electrification is a very optimistic Electravision scenario.
One explanation for the slow pace of change: electricity is more expensive per unit of energy than natural gas
for industrial users, as shown in the third chart. In other words, you need fewer BTUs due to efficiency of heat
pumps but you pay more for them. Furthermore, a lot of industrial equipment is fully amortized, so companies
might be slow to replace it at high upfront cost without the certainty of much lower operating expenses.
Electricity share of US industrial energy use unchanged Cumulative industrial process heat use by temperature
for decades, Share of industrial energy use Quadrillion BTU
45% 12
Natural gas
40%
10
35%
Petroleum 8
30%
25% 6
20%
Electricity 4
15%
10% 2
Temperature accessible by
5% industrial heat pumps
0
0% 0 200 400 600 800 1000 1200 1400 1600
1989 1994 1999 2004 2009 2014 2019 Temperature ( C)
Source: EIA, JPMAM. 2023. Source: "Manufacturing Thermal Energy Use in 2014", McMillan (NREL), 2019

Electricity: 2x-6x more costly than gas for industrial heat Industrial energy consumption by sector
Electricity cost per MJ divided by natural gas cost per MJ, industrial
users, assuming 85% industrial furnace efficiency and type, trillion BTUs
7x
Fuel Feedstock Total
6x Chemicals 2,815 4,326 7,141
5x Petroleum/coke 3,342 903 4,245
Paper 2,488 3 2,491
4x
Primary metals 1,734 307 2,041
Pennsylvania

3x
Food 1,511 - 1,511
Louisiana
California

Germany

2x Non-metallic minerals 1,161 - 1,161

Indiana

France
Illinois

Japan
Texas

China
Spain
Ohio

All others 247 599 846

Italy

1x
UK

0x
Total 13,298 6,138 19,436
Source: EIA, Eurostat, CEIC, JPMAM. September 2023. States shown are Source: EIA, 2018
largest industrial users of US primary energy.

An alternative to electric heat pumps: co-located solar power with thermal energy storage, in which the energy
is used for heat and is stored without the use of electrochemical batteries. One example: long-term energy
storage in bricks inside an insulated steel container; fans flow air over the bricks to access the heat. Very 20 th
century but potentially cheaper and more scalable than electrochemical alternatives.

9
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Electravision
14th Annual Energy Paper Executive Summary

The second explanation has to do with the chemistry of industrial energy use, which Lawrence Berkeley
National Laboratory examined in a piece on high and low potential for electrification.
LBNL cited primary metals (ex-steel), secondary steel10, machinery, wood products, plastics and rubber as
sectors with high electrification potential since fossil fuels are mostly used for process heat which could be
replaced by electric heat. Electrification potential is also high for certain mining activities related to transport,
excavation, pit crushing and belt conveying systems. That’s the table on the left.
On the right: low/medium electrification potential sectors. Chemicals, pulp/paper and food take advantage of
integrated systems in which fuel combustion waste heat (CHP) powers related processes. CHP-intensive sectors
are harder to electrify since producers would need to purchase energy previously obtained at little to no cost,
and/or redesign the entire process. Other hard to electrify sectors include non-metallic minerals such as glass,
brick and cement which require temperatures in excess of 1400°C, and which are non-conductive solids (i.e.,
harder to electrify production of things that do not conduct electricity). Finally, oil/coal refining exploits “own-
use” fuel consumption, a source of energy lost when switching to electricity.
As shown in the pie chart, sectors with high electrification potential use roughly 23% of US industrial energy;
medium potential sectors use 33%, and low potential sectors use 28%; the remainder was not analyzed.

Industrial sectors with high electrification potential Industrial sectors with medium/low electrification potential
Fuel consum ption shares: Fuel consum ption shares:
Heat Process Heat Process
Sector requirement HVAC Heat CHP Sector requirement HVAC Heat CHP
Primary metals ex. steel 1200°C 6% 75% 7% Food/beverages 120°C-500°C 4% 25% 40%
Fabricated metal 430°C-680°C 20% 61% 7% Chemicals 100°C-850°C 1% 32% 43%
Machinery 730°C 46% 39% 4% Pulp and paper 650°C 2% 21% 63%
Secondary steel 1425°C-1540°C 4% 87% 0% Non-metallic minerals 870°C-1600°C 3% 90% 1%
Wood products 180°C 10% 50% 14% Oil/coal products 220°C-540°C 0% 58% 22%
Vehicle parts (drying) 150°C 31% 33% 12%
Plastics and rubber 260°C 20% 33% 24%
Source: LBNL, "Electrification of buildings and industry ", March 2018. Source: LBNL, "Electrification of buildings and industry ", March 2018.

US industrial energy use by electrification potential Steel production

Includes: iron, Primary: Secondary: Share of
secondary steel,
aluminum, metal
Blast oxygen Electric arc global steel
products, wood, furnace furnace production
Medium (Chemicals,
plastics and High food processing) Global 72% 28%
mining 23% 33%
5.9 quads 8.7 quads China > 90% < 10% 53.7%
India 55% 45% 7.5%
Japan 75% 25% 4.7%
Low
(Refining, paper, US 30% 70% 4.3%
cement, glass, Russia 66% 34% 4.0%
primary steel)
Not analyzed 28% Korea 68% 32% 3.7%
(Constuction and 7.3 quads Source: Steel climate impact benchmarking report, USGS, April 2022
agriculture)
17%
Source: EIA (2020), LBNL, "Electrification of buildings and industry", 2018.

10
Secondary steel refers to recycled steel which is melted down and regenerated in electric arc furnaces.
Around 85% of all steel is already recycled globally, which is why primary steel production (producing it from
cast iron made from iron and coke) is the predominant production method. Pilot projects using green hydrogen
as a reducing agent to strip oxygen from iron oxide (instead of using carbon) are still in their infancy.
10
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Electravision
14th Annual Energy Paper Executive Summary

Electravision summary: the transmission buildout might be the hardest part

The charts show what Electravision would entail in the US: a 34% decline in fossil fuel consumption, a similar
increase in electricity generation and a ~400% increase in wind/solar generation; after which intermittent
renewable energy would represent 53% of the power coming from a much larger grid. I did not put an end date
on the charts since I have no idea how long this would take to accomplish.
US fossil fuel consumption US electricity generation US wind & solar electricty generation
Quad BTUs TWh TWh
95 6000 3000
Electravision
Electravision
90 5500 2500
85
5000
80 2000
75 4500
1500
70 4000
65 1000
3500
60
3000 500
55
Electravision
50 2500 0
1980 2000 2020 2040 1980 2000 2020 2040 1980 2000 2020 2040
Source: Energy Institute, JPMAM, 2023 Source: Energy Institute, IEA, JPMAM, 2023 Source: Energy Institute, IEA, JPMAM, 2023

Building wind/solar capacity and convincing owners of vehicles, furnaces and other devices to electrify might
not be the hardest part. Building more transmission might be. Utilities spend almost as much on transmission
and distribution as they do on power generation. In October 2023, the Department of Energy released a report
on transmission needed by 2035 in a scenario that sounds like Electravision: “higher load and high clean energy
growth”. The DoE’s estimate of required growth in transmission and interregional transfer capacity is very large
(see box/table), particularly compared to declining growth in new transmission shown on the next page.
Notably, the DoE does not believe that more distributed storage necessarily results in lower transmission needs.
Without legislative and cultural changes allowing transmission to replicate the growth of the interstate highway
system, fiber optic cables, national rail, civil aviation, waterways and other infrastructure, Electravision will
remain just that: a vision.
Transmission growth needed by 2035
Median growth vs 2020, high load & high clean energy scenario
Region Growth Interconnection Growth
US transmission grid, from the DoE October 2023 report
Plains (PL) 408% PL-TX 3519% Current within-region transmission: 85,000 GW-miles
Delta (DE) 231% PL-SW 3238% New transmission req. by 2030: 33,000 GW-miles (+39%)
Midwest (MW) 174% MO-PL 2102% New transmission req. by 2035: 108,000 GW-miles (+128%)
Mountain (MO) 173% DE-PL 1019%
New England (NE) 126% NE-NY 835% US utility annual spending by category
Southwest (SW) 118% MW-PL 730% Cents of capex per kWh of electricity sales, US$ 2022
Texas (TX) 113% DE-SE 572% 12
Southeast (SE) 102% MA-MW 474%
10 Other
Mid-Atlantic (MA) 61% MW-SE 416%
New York (NY) 46% MA-NY 412% Transmission
8
Northwest (NW) 31% FL-SE 295%
Florida (FL) 24% MO-NW 202% Distribution
6
California (CA) 4% MA-SE 140%
CA-MO 130% 4
MO-SW 129% Electricity Generation
CA-SW 102% 2
DE-MW 30%
CA-NW 25% 0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
Source: "National transmission needs study ", DOE, October 2023
Source: EIA, February 2023

11
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Electravision
Executive Summary
14th Annual Energy Paper

Last year we covered the difficult and time-consuming process of building high voltage transmission lines. While
projects less than 150 miles have been completed in 5-10 years, projects more than 400 miles (e.g., from Wichita
to St Louis) can require 15-20 years to complete. Another issue: rising costs. Of all 47 categories of core goods
in the US producer price inflation report, the highest increase since 2019: transformer equipment, at 71% (see
p. 21). Wood Mackenzie reports lead times of two years for buyers of generator transformers and power
transformers, and Electravision has barely begun.
US transmission line growth US primary energy consumption
Miles added per year Quadrillion BTUs
4,000 100
95
3,500
90
3,000
85
2,500 80
2,000 75
70
1,500
65
1,000
60
500 55
0 50
2013 2015 2017 2019 2021 '65 '70 '75 '80 '85 '90 '95 '00 '05 '10 '15 '20
Source: S&P Global, JPMAM, 2024. Note: Transmission lines > 100 kV. Source: EI Statistical Review of World Energy, JPMAM, 2023

An important caveat: our Electravision scenario assumes that total US energy needs will not change much
over the next two decades. Unchanged US energy demand is consistent with the last couple of decades; the
energy needs of a growing US population have been offset by improving energy efficiency. However, the rise
of AI might change that. One illustrative example: the PJM (mid-Atlantic) region has made sharp increases to
projections of future power demand for Dominion Resources, a utility serving 6 million customers in 15 states.
These increases are entirely due to an increase in data centers which serve advanced computing/AI needs11.
Constellation Energy estimates that the AI revolution could require more power in the US than the future electric
vehicle fleet12. If that’s the case, the productivity benefits from AI better be large enough to offset the increase
power load. Bottom line: the rise of AI could make the journey to Electravision longer, harder and most costly.
PJM progression of power demand forecasts for Dominion PJM forecasts for Dominion Resources power demand
Resources, Gigawatts Terawatt hours
42 180
Total
PJM 2023
160
38 PJM 2022
140
PJM 2020 Commercial
34 120
PJM 2019
100
30 PJM 2021
80 Data centers
(part of commercial)
26 60
40 Residential
22
20
0 Industrial
18
2019 2021 2023 2025 2027 2029 2031 2033 2035 2037 2039 2011 2014 2017 2020 2023 2026 2029 2032 2035 2038
Source: PJM 2023 Power Demand Outlook, January 28, 2024 Source: Dominion Resources Integrated Resource Plan, January 28, 2024

11
Power-hungry Microsoft plans to use AI to accelerate development of dedicated nuclear fission plants and
has invested in nuclear fusion startups. It’s unclear if either of these efforts will succeed, and I am skeptical on
both. A more likely savior: sub-quadratic scaling methods and other techniques which eventually reduce the
computational intensity of large language models and other forms of AI.
12
“AI is ravenous for energy. Can it be satisfied?”, WSJ, December 23, 2023; Constellation Energy
12
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Electravision
14th Annual Energy Paper Executive Summary

Meanwhile, in the trenches: the North American Electricity Reliability Corporation (NERC) and the Federal
Energy Regulatory Commission (FERC) are raising red flags on the grid even before Electravision happens. If
you think electricity outages are bad, wait until you read about near-misses of natural gas outages.
The electricity grid, outage risk and EPA proposals
Due to retirement of nuclear and dispatchable thermal generation and addition of intermittent solar and wind
resources, US cities face rising risks of electricity outages. The MISO region which stretches from Minnesota to
Louisiana is cited by NERC as having the greatest risk of power outages, even in normal conditions. Outage risk
in more extreme weather conditions is cited for New York, New England and the entire Western US. NERC cites
peak loads rising at “an alarming rate” due to electrification, coinciding with more intermittent generation and
80-110 GW of nuclear and fossil fuel retirements by 2033 (~7% of current installed capacity). “Reserve margins”
indicate buffers to deal with a spike in summer demand and are shown in the chart on the left13.
As shown on the right, unanticipated generation losses have been increasing during storms. In 2022, 90 GW of
sidelined generation capacity represented ~13% of anticipated generation resources in the Eastern US. For
everyone upset about wind outages during storms, look at the pie chart: it’s the natural gas system that led to
the majority of unplanned electricity outages during the December 2022 winter storm, not renewables whose
winter output is expected to be low. To be clear, NERC is not arguing in favor of more renewables here; they
argue for more investment in equipment winterization, gas storage and transmission.
Generation capacity buffer during peak summer demand Peak unavailable national electricity generation due to
Anticipated reserve margin cold weather, Gigawatts
50% 100
45% 90
ERCOT (Texas) 80
40%
70
35%
PJM (Mid-Atlantic) 60
30%
Southeast 50
25% 40
California
20% SPP (Plains) 30
15% New York 20
Northwest 10
10%
New England 0
5%
MISO (Midwest) Polar Vortex SW Event 2018 Event 2021 Event 2022 Event
0% (Jan 6-8, (Feb 1-5, (Jan 15-19, (Feb 8-20, (Dec 21-26,
2024 2026 2028 2030 2032 2014) 2011) 2018) 2021) 2022)
Source: "2023 Long-Term Reliability Assessment", NERC, December 2023 Source: "Winter Storm Elliott Report", FERC, NERC, October 2023

A cry for help from MISO regarding EPA proposals

MISO now warns of “immediate and serious challenges” to Unplanned generation outages, derates and failures to
grid reliability due to wind/solar intermittency, and disclosed start by fuel type during 2022 storm, % of unavailable MW
that it averted a capacity shortfall in 2023 only due to
Oil/Distillate Oils
postponement of planned generation retirements. Demand 4%
continues to grow due to electrification while new projects are Coal/Lignite
23% Wind
delayed due to supply chain and regulatory issues. 4%
MISO’s report warned of adverse consequences from current Other
3%
EPA proposals which would effectively require all coal plants, Hydro
some existing natural gas plants and all new gas plants to (a) Gas
1%
Solar
retire by a certain date, (b) retrofit with carbon capture and 63%
1%
storage or (c) co-fire with green hydrogen. MISO cites cost Nuclear
1%
and technological readiness as constraints to implementation
of EPA proposals.
Source: "Winter Storm Elliott Report", FERC, NERC, October 2023
“MISO’s Response to the Reliability Imperative”, Feb 2024

13
A note on NERC anticipated reserve margins. Variable resources such as wind and solar are “derated” by NERC to
their expected peak value. For summer wind, expected peak capacity is often only 10%-20% of nameplate capacity.
For summer solar, expected values are closer to 60% of nameplate. In winter, wind is derated to 5%-15% of
nameplate while solar is derated close to zero. Source: NERC Reliability Assessment Director
13
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Electravision
14th Annual Energy Paper Executive Summary

Local gas distribution systems

As bad as electricity outages are, widespread natural gas outages in large cities would be worse. This is what
happened and what almost happened in December 202214:
• Temperatures 20-30 degrees below normal (see map) resulted in the failure of gas production wellheads,
pipelines and distribution systems. Dry gas production in the lower 48 states fell by 16%, with Marcellus
and Utica production falling by 23%-54%
• On Christmas Eve morning 2022, interstate natural gas pipelines serving Con Edison experienced drops in
pressure due to production losses and operational issues. By noon the pipelines informed Con Edison that
pressure would not improve unless demand decreased
• Con Edison was close to having to cut gas service to customers, leaving them unable to heat buildings while
outside temperatures were in the single digits. Con Ed supplemented its gas supply by drawing on its backup
LNG facility until systemwide pressure was normalized, and ended up narrowly avoiding a gas system outage
• If a gas outage occurred, local gas distribution companies would need to go building-by-building and shut
off gas valves to ensure that residual gas does not seep through units whose pilot lights are out. During the
system restoration process, the main distribution system would be purged; then workers would have to
ensure at each point of service that heating and cooking gas lines are safely purged and operational before
restoring service and relighting pilot lights. Homes or buildings with safety issues would need remediation
before any gas restoration. Even losing service to 130,000 customers could have taken five to seven weeks
(!!) to restore. A large outage could also cause extensive property damage due to burst water pipes within
homes and buildings since water expands when it freezes
FERC/NERC solutions include winterization of gas equipment, more LNG backup supply and more fixed non-
interruptible gas contracts15 which in turn would require more gas production and distribution. Reinvestment
in the natural gas industry is not popular at a time when all eyes are focused on the renewable transition. But
if the transition to Electravision takes more time than expected, such investment may be inevitable.
Departure from Average Daily Minimum temperature, Con Edison average meter station inlet pressure
Dec 24, 2022, Degrees in Fahrenheit Pounds per square inch gauge
850
800
750
700
650
600
Condition Red Terminated
550
500
450
Gas System Emergency,
400 Condition Red
350
12/20/22 12/22/22 12/24/22 12/26/22 12/28/22
Source: “Winter Storm Elliott Report”, FERC, NERC, October 2023 Source: "Winter Storm Elliott Report", FERC, NERC, October 2023

14
“Winter Storm Elliott Report: Inquiry into Bulk-Power System Operations During December 2022”, FERC, NERC
and Regional Entity Staff Report, October 2023
15
When gas customers sign firm contracts, both pipelines and supply are designed to meet customer demand
at all times. In contrast, interruptible customers pay a much lower price than firm rates since they do not have
pipeline capacity set aside for them. Even so, customers with interruptible contracts sometimes draw when
not permitted, particularly during severe storms. While there are financial penalties involved, that doesn’t solve
the real-time problem of customers situated upstream consuming more than their designated amounts, causing
severe problems for customers downstream. This is what happened to Con Ed during Winter Storm Elliott.
14
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Vaclav

The latest from Vaclav Smil

For many years, Vaclav served as our technical science advisor on this paper. His insights and suggestions were
invaluable and the opportunity to learn from him is one of the highlights of my 36 years at JP Morgan. Vaclav
turned 80 last year and we still correspond on a variety of topics. He sent me an essay he just completed which
he has allowed me to share, which you can read here. Some passages from its introduction (bolding is mine):
“In mass terms, we will never run out of fossil fuels: enormous quantities of coal and hydrocarbons will
remain in the ground after we end their use because it would be too expensive to extract them.
Although the world of the early 2020s is in no imminent danger of running out of fossil fuels, in the long
run they would have to be replaced even in the absence of any connections to global warming. Their
conversions made the modern civilization possible, but their production, processing and transportation
are often environmentally disruptive, with impacts ranging from land dereliction to water pollution;
their combustion generates not only CO2 but also such pollutants as CO, nitrogen (NO, NO2) and sulfur
(SO2 and SO3) oxides and particulate matter; their highly uneven distribution contributes to worldwide
economic inequalities, and the quest for secure fossil fuel supplies has led to many detrimental policies
and contributed to recurrent conflicts”
“Clearing of forests, large-scale cropping and animal husbandry have been with us throughout recorded
history but the rising combustion of fossil fuels has been by far the greatest contributor of CO2 during
the past two centuries, followed by CH4 (from rice fields, landfills, cattle, and natural gas production),
and N2O (mostly from nitrogenous fertilizers). Realization that these trace gases could affect climate is
more than 150 years old”
“There has been an exponential rise of attention paid to global climate change. Much has been learned,
much remains uncertain but basic facts are indisputable. Ice core analyses show CO2 levels close to
270 parts per million (ppm) by volume during the preindustrial era; in 1958 when Mauna Loa monitoring
began they reached 313 ppm; by the year 2000 they were 370 ppm and by the end of 2023 they reached
420 ppm, more than 50% above the late 18th-century level…This rise, together with contributions by
CH4 and N2O, has translated to about 1⁰C of global warming compared to the 19th-century mean. All
continents have been affected, recent decadal warming gains have been steadily rising and the eight
years between 2015 and 2022 were the warmest years on record”
The remainder of Vaclav’s essay walks through accomplishments and setbacks in the transition to date, future
energy demand, green hydrogen for steel production, materials required per MW for wind vs natural gas, the
energy intensity of new transmission capacity, transition costs and a lot of other topics. Vaclav’s essay is a great
read for anyone looking to learn more about the transition from one of the world’s preeminent energy scientists.
Atmospheric CO2 concentrations Ocean heat content and sea level
Parts per million Zettajoules, vs 1957 baseline Millimeters, vs 1993 baseline
425 375 120
For the last decade, the ocean has been absorbing
10 ZJ of heat per year... which is equal to ~20x
400 annual global primary energy consumption 100
300
375 80
225
350 Heat content
60
150
325 40
Sea level
300 75
20

275 0 0
1750 1780 1810 1840 1870 1900 1930 1960 1990 2020 1955 1965 1975 1985 1995 2005 2015 2025
Source: National Oceanic and Atmospheric Administration, May 2023 Source: NASA/NOAA, JPMAM, 2023. 1 ZJ = 1021 joules.

15
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
China
14th Annual Energy Paper

All Eyes on China: the world’s largest energy user is building a lot of everything
Twenty years ago, Chinese energy consumption was a fraction of US levels and not much higher than Germany.
Today, China’s primary energy and power consumption are larger than the US and Germany combined. That’s
why there’s so much attention paid to China’s progress on decarbonization.
China primary energy consumption by type China electricity generation by source
Exajoules TWh
180 9,000
Nuclear US Other US
160 Hydro Germany 8,000 Oil Germany
Other renewables Bioenergy
140 Gas 7,000 Gas
Oil Nuclear
120 Coal 6,000 Solar
Wind
100 5,000 Hydro
Coal
80 4,000
60 3,000
40 2,000
20 1,000
0 0
2000 2004 2008 2012 2016 2020 2000 2004 2008 2012 2016 2020
Source: EI Statistical Review of World Energy, JPMAM, 2023 Source: EI Statistical Review of World Energy, JPMAM, 2023

A lot of attention is paid to China’s continued construction of new coal plants, and we have shown the chart
below for many years. China approved 106 GW of new coal plants in 2023, increasing its total coal pipeline to
200 GW. China and India coal consumption rose by 5% and 8% in 2023, offsetting a 20% decline in the US/EU
and driving global coal consumption up by 1.4% in 2023 to a new record level.
That said, China is also building out large amounts of renewables and nuclear as well:
• China installed ~230 GW of new wind and solar power in 2023 with growth rates of ~100% y/y
• As shown on p.4, China’s electrification of energy use has surpassed the US and Europe and is still rising
• China’s BEV share of vehicle sales is now ~25% with another 10% from PHEVs, much higher than the US
• As of February 2023, China had 57 GW of nuclear in operation, 30 GW under construction, 46 GW of
proposed plants and 175 GW in the distant planning stage
• The coal share of China’s primary energy consumption has fallen by ~1.3% per year since 2011, declining
from 70% to 55%
• China’s investments in the grid have reduced its solar curtailment to just 2%, down from 10% in 202016
• The chart at the bottom right shows how China is the primary global driver of energy transition investment

The impact of China on global coal capacity Largest 2023 transition investment countries
Coal capacity: additions and retirements, gigawatts China $676
100 US $303
Electrified transport
Germany $95
80 Renewable energy
UK $74
Power grids
60 France $55
Electrified heat
Brazil $35
40 Clean industry
Other Spain $32
Energy storage
20 additions Japan $32
Nuclear
China net India $31
additions CCS
0 Italy $30
US/EU net Hydrogen
retirements EU-27 $341
-20 Clean shipping
Other Rest of World $277
retirements
-40 $0 $100 $200 $300 $400 $500 $600 $700
'00 '02 '04 '06 '08 '10 '12 '14 '16 '18 '20 '22 US$, billions
Source: Global Energy Monitor, JPMAM, January 2024 Source: BloombergNEF, JPMAM, 2024

16
“How China became the global renewables leader”, Wood Mackenzie, November 20, 2023
16
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
China
14th Annual Energy Paper

That’s good news but China’s energy demand has been growing, unlike stagnant energy consumption in the
West. As a result, renewable shares of China’s energy and electricity use are not as indicative of emissions
declines as they are in developed countries. The charts below show absolute levels of emissions and fossil fuel
consumption. Is China close to an emissions plateau? It’s tempting to think it might be given China’s ongoing
investments in renewable energy and nuclear power. The next 3-4 years will tell us what we need to know;
keep in mind that the prior plateau from 2013-2017 ended up being a head fake.
Fossil fuel consumption CO2 emissions from energy
Quadrillion BTUs Billion tonnes of carbon dioxide
140 12
China
China
120 10

100
US 8
80
6 US
60
Europe 4
40
Europe
2
20

0 0
'65 '70 '75 '80 '85 '90 '95 '00 '05 '10 '15 '20 '65 '70 '75 '80 '85 '90 '95 '00 '05 '10 '15 '20
Source: EI Statistical Review of World Energy, JPMAM, 2023 Source: EI Statistical Review of World Energy, JPMAM, 2023

In last year’s paper we included a long section on renewable supply chains and China. The challenge for the G7
is illustrated below. The Y axis shows an index of environmental health and corruption by country (the lower
the value, the more environmentally unhealthy and corrupt each country is), along with descriptions of each
country’s presence in renewable supply chains. Other than Australia, most of these countries score poorly,
particularly China. As a result, while the G7 can try and develop its own natural resource and processing supply
chains for renewables, they will be competing with countries like China that can almost always do so more
cheaply and with much less regulatory oversight. Two examples from prior Eye on the Market pieces: China’s
unregulated rare earth element industry reportedly accounts for 40%+ of its total REE output, and Chinese
polysilicon plants (which dominate global solar supply chains) routinely dump toxic waste into rivers and lakes,
which is one reason why 70% of China’s rivers and lakes are unfit for human use.
Transition minerals are generally sourced and processed in countries with poor environmental protections and a lot of corruption
Index, environmental health score (75%) + corruption score (25%); 0 = least environmentally healthy and most corrupt
75
Lithium, Manganese
70

65
Copper, Lithium, Neodymium,
60 Polysilicon, Graphite, Rare
Lithium, Copper Copper, Nickel, earths, Aluminum, Battery
55 Graphite, Niobium anodes and cathodes, PV
Copper components and assembly,
50 Manganese Nickel,
Aluminum Mineral processing
Graphite Manganese,
45 Nickel Platinum,
Chromium
40
Nickel Copper, Cobalt
35

30 Rare earths

25

20
G7 Australia Chile Peru Brazil Gabon Mozambique Russia PhilippinesSouth Africa Indonesia Congo Myanmar China
Environmental health: PM 2.5, NOx, SO2, CO, heavy metals, ozone, lead, drinking water, biodiversity, etc
Source: Yale EPI Environmental Index, Transparency International, JPMAM. 2023.

17
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Essential charts

Essential energy transition charts

Every year we include charts that track the progress, costs and challenges of the energy transition. This page:
global wind/solar generation soar past nuclear; the IEA consistently underestimates solar capacity additions;
ISO capacity credits show that when 1 MW of wind/solar is added to the grid, thermal gas/coal requirements
only fall by 10%-25% of that amount due to wind/solar intermittency; fossil fuel shares of primary energy are
falling faster in China and Europe than in the US; global hydropower capacity factors have been declining; and
coal shares of primary energy have been flat or falling with the exception of Vietnam and Indonesia.
Global nuclear vs wind + solar electricity generation IEA consistently underestimated solar capacity growth
Terawatt-hours GW of global annual solar additions vs IEA 10-year forecasts
3,500 400
Actual solar additions
350 2022 iea
3,000 2021 iea
300 2020 iea
2,500 2019 iea
Nuclear 250 2018 iea
2,000 2017 iea
2016 iea
200 2015 iea
1,500 2014 iea
Wind + Solar 150 2013 iea
1,000 2012 iea
100 2011 iea
2010 iea
500 50 2009 iea

0 0
'65 '70 '75 '80 '85 '90 '95 '00 '05 '10 '15 '20 2000 2005 2010 2015 2020 2025 2030 2035 2040
Source: EI Statistical Review of World Energy, JPMAM, 2023 Source: Carbon Brief, BNEF. 2023. IEA = International Energy Agency.

How much natural gas capacity can be reduced per MW of Fossil fuel share of primary energy since 1965
new wind and solar power? % of primary energy consumption from coal, oil and nat gas
%, computed for 2021, assuming new wind and solar = 10% of demand 100%
100%
90% 95%
80%
90%
MISO (Upper Midwest)

70%
ISNE (New England)
CAISO (California)

SOCO (Southeast)
PJM (Mid Altantic)

60% 85%
China
ERCOT (Texas)

50%
SPP (Midwest)

80% Rest of world

40%
NYISO (NY)

30% US
75% Europe
20%
10% 70%
0% '65 '70 '75 '80 '85 '90 '95 '00 '05 '10 '15 '20
Source: EIA data, JPMAM computations, 2022 Source: EI Statistical Review of World Energy, JPMAM, 2023

Falling water Top Ten consumers of coal by energy content

Hydropower capacity factor, Index, 1991 = 100 %, Coal share of primary energy consumption
140 80%

130 70% South Africa

China
120 60%
India
China 50% Vietnam
110
Indonesia
40%
100 Japan
30% South Korea
World
90 Canada Germany
20%
Russia
80 Brazil
10% US
EU
70 0%
1991 1996 2001 2006 2011 2016 2021 '90 '95 '00 '05 '10 '15 '20
Source: IEA, JPMAM, 2023 Source: EI Statistical Review of World Energy, JPMAM. 2023.

18
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Essential charts

On this page: the US has achieved US energy independence for the first time in 40 years while Europe and China
compete for global energy resources; energy intensive manufacturing has been shifting to the developing world
for reasons related to developed world outsourcing and the developing world’s own consumption; renewable
jet fuel alternatives are very expensive, resulting in cost and supply challenges if the US military attempts to
decarbonize energy use for aviation17. As per the last two charts, CCS faces a fundamental physical challenge:
while coal and gas plants account for the bulk of industrial emissions, they have among the lowest flue gas
concentrations of CO2, which increases the cost and complexity of capturing and sequestering it.
Energy dependence and independence A shift in energy intensive manufacturing to the emerging
Net imports of oil, natural gas and coal in million tonnes of oil equiv. world, % of global production
1,000 Europe 100%
Net importer Emerging economies
800 China 90%
80%
600
70%
400
60%
200 50%
0 40%
-200 US 30%
20%
-400
Russia 10%
-600 Developed economies
Net exporter 0%
-800 1995 2022 1998 2023 1998 2023 1998 2023 2005 2022
1980 1990 2000 2010 2020 Manufacturing Ammonia Steel Cement Plastic
Source: EI Statistical Review of World Energy, JPMAM, 2023 Source: UN DESA, Worldsteel, PlasticsEurope, USGS, JPMAM, 2024

Renewable jet fuel cost estimates Federal gov't energy consumption

Cost (Euros per liter) Trillion BTUs
4.5 350
Hydroprocessed esters and fatty acids
Gasification
US civilian agencies
4.0 300
Power to liquids US Department of Defense
3.5 Alcohol to jet 250
Synthesized isoparaffins
3.0
Jet fuel from fossil fuels 200
Sugar cane molasses
Renewable electricity

2.5
150
Used cooking oil

Municipal waste

2.0
Palm distillate

Agri residues

Agri residues
Energy crops

Energy crops

100
1.5
Sugar cane
Corn grain

50
Palm oil

1.0
Soy oil

0
0.5
Jet fuel Diesel Gasoline Other Electricity Nat gas Other
0.0
Vehicles and equipment Facilities
Source: Royal Society Policy Briefing, February 2023. Energy crops
include oilseed, miscanthus and poplar. Source: DOE, JPMAM, 2022

Annual US GHG emissions from industrial sector CO2 concentration in flue gas streams
Million tonnes of CO2 equivalent Percent by volume
800 100%
Non-ferrous metals, pulp/paper
Natural gas CCGT power plants

90% Ethylene oxide

Ammonia processing
Ethanol, biomass fermentation

700
Nat gas processing, post-sep.
Gasification biofuels

80%
Ammonia fuel combustion

600
Hydrogen production
Natural gas power plant

Integrated steel mills

70%
500 60%
Aluminum production
Cement production

Fuel oil power plant

Steel blast furnace

Hydrogen via SMR

Cement plant
Coal power plants

Coal fired plant

Ethylene plant

400 50%
Gas processing

Oil refineries

40%
Pulp & paper

300
Bioenergy
Refineries

30%
Ammonia

200
Ethanol

20%
100 10%

0 0%
Source: Energy Futures Initiative. February 2023. Source: IPCC, Swedish Env. Research Institute, Penn State, JPMAM. 2022.

17
And to be clear, an EV version of the 60-tonne M1 Abrams tank is absurd; it would require a battery that
weighs 2/3 of the tank itself to match its 250 mile range
19
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Essential charts

On this page: global ICE passenger car sales rebounded in 2023; lithium ion battery pack prices continue to
decline after a temporary spike in 2022; EV metals costs per battery type explain why LFP batteries are preferred
by automakers (i.e., no cobalt or nickel); China and Europe rank at the top of EVs as a share of car sales; US
gasoline demand appears to have peaked since miles traveled have reached pre-COVID levels while gasoline
demand has not; and the mileage of internal combustion engine cars by model year continues to rise. Notes:
global gasoline demand has risen above its 2019 peak, in contrast to the US data; and Germany EV sales have
declined in share terms due to the expiration of certain purchase subsidies.
Global passenger vehicle sales by drivetrain Volume-weighted average lithium-ion battery pack price
Vehicles, millions 2023 US$ per kWh Percent, y/y change
90 $1,600 10%
80 Plug-in
hybrid Price change
70 $800
60 Battery 0%
electric Battery
50 pack price
Hybrid $400
40
30 -10%
Internal
$200
20 combustion

10
0 $100 -20%
2015 2016 2017 2018 2019 2020 2021 2022 2023e '10 '11 '12 '13 '14 '15 '16 '17 '18 '19 '20 '21 '22 '23
Source: BloombergNEF, December 2023 Source: BloombergNEF, December 2023

Estimated metals cost per EV battery type EV share of passenger vehicle sales, largest car markets
US$ per 60 kWh battery Percent, BEV plus PHEV
$3,500 35%
Lithium Nickel Manganese Cobalt Oxide battery
NMC: Li, Ni, Mn, Co, Cu, Al, steel 30% China
$3,000 Lithium Nickel Cobalt Aluminum Oxide battery
NCA: Li, Ni, Co, Al, Cu, steel France
$2,500 25% Germany
Lithium Iron Phosphate battery
LFP: Li, Cu, Al, steel, iron UK
$2,000 20%

$1,500 15% Spain

S. Korea
$1,000 10% US
Italy
$500 5% Japan
India
$0 0%
2020 2021 2022 2023 2024 2017 2018 2019 2020 2021 2022 2023
Source: Uni. of Birmingham (UK), ANL, Bloomberg, JPMAM, Feb 13, 2024 Source: BloombergNEF, Bloomberg, JPMAM, Q3 2023

US gasoline consumption vs vehicles miles traveled US real-world average fuel economy by model year
Thousand barrels per day Billion miles traveled Miles per gallon
9,500 280 28
9,000 260 26
8,500 Gasoline demand 240
24
8,000 220
22
7,500 Miles traveled 200
20
7,000 180
18
6,500 160
6,000 140 16

5,500 120 14

5,000 100 12
1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020
Source: Bloomberg, JPMAM, December 2023 Source: EPA, JPMAM, 2023

20
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Essential charts

On this page: transformer and other transmission equipment have seen the highest inflation of all producer
goods since 2019; biofuels produced from corn, sugarcane and palm oil are generally small at 1% of primary
energy, with the exception of Brazil and Indonesia; hydraulically fractured oil and gas production has been a
growing part of US energy supplies, now accounting for >50% of primary energy consumption; planned CCS
capacity is 6%-7% of current emissions in the US and Europe but CCS project completion rates have historically
been low, so the projections are speculative. Last chart: the large energy deficit resulting from production of
synthetic fuels, using the example of a Sabatier reactor to produce green methane from atmospheric CO 2 and
green hydrogen. In addition to the energy deficit, there’s also the cost of carbon capture equipment and
electrolyzers to consider. We will discuss synthetic fuels in greater detail next year.
Core goods PPI component inflation Biofuel production as a percent of primary energy
% increase vs 2018 for each of the 47 core goods categories Percent
80% 7%
Transformers &
70% Power Regulators 6%
60%
5%
50%
40% 4%

United Kingdom
30% 3%

South Korea
Netherlands
20%

Indonesia
Argentina

Colombia
2%

Germany
Thailand

Portugal

Sweden

Canada
Finland

Austria
10%

Poland

France

China
Brazil

Spain

India
1%
0%

US
-10% 0%
Source: Bloomberg, JPMAM, February 2024 Source: EI Statistical Review of World Energy, JPMAM, 2023

Percentage of US oil and gas production derived from Hydraulic fracturing accounted for 53% of all US primary
hydraulic fracturing through year-end 2022 energy consumption in 2022 Solar/Geoth
90% 100%
Wind
80% 90% Hydro
80% Nuclear Biomass
70%
Coal
60% 70%
Oil, gas and NGLs from
60% conventional sources
50%
50%
40%
Natural gas 40%
30% Natural gas Oil, gas and NGLs
liquids (ethane, 30% from unconventional
20% sources requiring
Crude oil propane, butane) 20%
10% hydraulic fracturing
10%
0% 0%
1995 1998 2001 2004 2007 2010 2013 2016 2019 2022
Source: EIA, US Department of Energy, JPMAM, 2022 Source: EIA, BP, Society of Petroleum Engineers, S&P Platts, JPMAM. 2022.

Current vs planned carbon sequestration by 2030 Production of 1 kg of e-methane via Sabatier reaction
Million tonnes per annum of CO2 Percent of current emissions Megajoules
900 7% 150 DACC for
atmospheric CO2 DACC: 366 kWh of electricity and 5.25
800 GJ of thermal heat per ton of CO2
In construction or development (lhs) 6% 125
700 Current capacity (lhs) Sabatier reaction Sabatier reaction: 75% efficiency
Total as % of current emissions (rhs) 5% 100 Electrolysis: 55 kWh of electricity per
600 kg of H2
500 4%
75
400 3%
300 50 Electrolysis for
2% green hydrogen
200
25 Synfuel: e-methane
1%
100
0 0% 0
US Europe China Global Energy in Energy out
Source: Global CCS Institute, OWID, JPMAM. 2024. Source: Spitfire Research, Keith et al (DACC), JPMAM, 2024

21
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Investment returns

Investment returns in oil & gas, renewable energy and nuclear-exposed shares
The profitability of the traditional energy sector improved substantially after the profitless period from 2013-
2020. However, energy sector valuation multiples remain low due to concerns about stranded asset risk, and
the degree to which some institutional investors either cannot or will not invest in the sector. From 2021 to
November 2023, the best energy trade was to be long traditional (legacy) energy sectors and be short renewable
energy. The Fed pivot in November 2023 revived renewable valuations; I don’t think there’s a clear cut winner
for 2024 now that most of the damage has been done to unprofitable energy business models. Note at the
bottom how US oil and gas production is soaring.
Oil & gas unprofitability decade ended in 2021 S&P 500 energy share of market cap and net income
Free cash flow, $ billions Percent
$24 30%
S&P 1500 Oil and Net income
$20 Gas E&P Index share
25%
$16
$12 20%

$8 15%
$4
10%
$0
5% Market cap
-$4 Shale revolution share
portfolio
-$8 0%
(22 companies)
-$12 -5%
2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 1995 2000 2005 2010 2015 2020
Source: Bloomberg, JPMAM, Q3 2023 Source: Factset, JPMAM, February 12, 2024

Percentile rank of energy sector valuations relative to S&P 1500, Investment returns: renewables vs traditional energy
start of available data to Jan 2024 Index (100 = Jan 2019)
Enterprise value to 500
cash flow (EBITDA) Free cash flow yield Renewables composite
450 (equal weighted)
Last 12 M Next 12 M Last 12 M Next 12 M
400
Oil & Gas Drilling 0 3 73 95
Oil & Gas Equipment 350
NASDAQ Clean Edge
1 0 96 98 300 Wilderhill Clean Energy
& Services
FTSE Renew/Alt Energy
Integrated Oil & Gas 4 2 87 84 250 S&P Global Clean Energy
Oil & Gas Exploration 200 MAC Global Solar
4 8 85 83
& Production
150
Oil & Gas Refining &
6 17 91 90 100
Marketing
Oil & Gas Storage & 50 MSCI World Energy Index
8 5 81 82 (oil, gas and pipelines)
Transport 0
Energy 2 4 87 85 2019 2020 2021 2022 2023 2024
Source: FactSet, JPMAM, January 2024 Source: Bloomberg, JPMAM, February 9, 2024

US oil and gas production soaring

mm barrels per day Trillion cubic feet per month
21 3.8
Crude and petroleum products (lhs)
19
Natural gas production (rhs) 3.3
17

15 2.8
13

11 2.3

9
1.8
7

5 1.3
1994 1998 2002 2006 2010 2014 2018 2022
Source: Bloomberg, JPMAM, November 2023

22
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Investment returns

A primary reason why renewable energy indexes have not performed as well as many investors expected:
despite being seen as growth stocks, their operating margins are generally below a variety of growth sectors
and traditional oil & gas sectors as well.
Operating margins by sector and index
40%

Integrated oil & gas companies

NASDAQ Cybersecurity Index

35%

Communications equipment

Electronics manufacturing
S&P Global Clean Energy

MSCI Cybersecurity Index

Bloomberg Clean Energy

30%
Wilderhill Clean Energy

Internet of Things Index

NASDAQ Clean Energy

Electronic components
Healthcare equipment

Internet infrastructure

Electronic equipment
Oil & gas exploration

MSCI Robotics Index

25%

Application software
Oil & gas equipment
FTSE Clean Energy

Healthcare supplies

MSCI Fintech Index

Life sciences tools
Large cap pharma

Systems software
Oil & gas refining
MAC Solar Index

20%

Data processing

Tech hardware
IT consulting
15%
Pipelines

Wireless
Telecom
10%

Biotech
5%
0%
-5%
Clean OIl & gas Healthcare Technology/Software Telecom Cyber,
-10% energy Robotics and
-15% Fintech
Source: Bloomberg, JPMAM, March 1, 2024

The other notable trend: a 2023 rally in companies linked to nuclear energy. As we discuss in the next section,
there’s a lot of talk of a nuclear renaissance in the West. So far there’s little evidence of this, but China, India
and Russia have a lot of MW either under construction or in the planning phase (see table). The MVIS Global
Uranium & Nuclear Energy Index tracks performance of companies in the uranium and nuclear energy industries
with at least 50% of revenues from nuclear. The spot price of uranium has reached a post-Fukushima high, a
level which may spur more mining projects. The top 3 performers in the table are uranium producers.
Global energy ETF prices Largest 10 positions in Global Uranium & Nuclear Energy ETF
Index (100 = January 2022) Return in USD
150 MVIS Global
Uranium &
Name Country since 1/1/2023
140 Nuclear Global Uranium & Nuclear
- 43%
130 Energy Energy ETF
120 Cameco Corp Canada 93%
110 Nexgen Energy Canada 73%
NAC Kazatomprom Kazakhstan 60%
100
Constellation Energy US 51%
90 ISE Global
China General Nuclear Power Co China 37%
Wind
80 CEZ AS Czech Republic 20%
Energy
70 MAC Global Endesa Sa Spain 10%
60 Solar PG&E Corp US 1%
Energy Public SVC Enterprise US 1%
50
Jan-22 Jul-22 Jan-23 Jul-23 Jan-24 Fortum Oyj Finland -22%
Source: Bloomberg, JPMAM, February 13, 2024 Source: Bloomberg, JPMAM, February 12, 2024

Nuclear power on the drawing board, gross GW

Under
Proposed Planned
construction
Emerging countries 61 95 304
of which China 30 46 175
Developed countries 9 13 48
Source: World Nuclear Association, 2024

23
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Nuclear power

Nuclear power: Elusive measures of true cost, German decommissioning/deindustrialization and NY State
The most elusive questions in energy, other than those related to commercialized fusion18, revolve around real-
world costs of building nuclear fission plants. In December 2023, leaders from 22 countries announced plans to
triple nuclear capacity by 2050 and extend the life of existing plants. How much will it cost to build new plants?
It depends on which countries you look to as examples. Regional differences explain most of the variations
below and these differences are unlikely to go away. The size of a fission plant and type (boiling or light water,
pressurized water, fast breeder, etc) do not explain cost differences in any meaningful way. Recent large cost
overruns in the US, UK, France and Finland are a logical starting point for any plant undertaken in the West19.
Nuclear capital costs by country and year of completion
Millions of 2023 US$ per MW
$14
Developed country
$12
Emerging country
$10

$8

South Korea, 2024

South Korea, 2019

South Korea, 2015

$6
Argentina, 2014

South Korea, ?
Pakistan, 2021

Pakistan, 2017
Slovakia, 2024

Bangladesh, ?
Belarus, 2023

Finland, 2023

France, 1999
Turkey, 2026

Russia, 2020

Russia, 2019
Japan, 2009
China, 2021

China, 2018

China, 2019
China, 2018

China, 2021

China, 2004

China, 2016

China, 2022

China, 2011

China, 2019

China, 2016
China, 2015
China, 2022
$4 India, 2021

India, 2013
UAE, 2023
France, ?

UK, 1989

UK, 1985
UK, 2027

UK, 1965

UK, 1995
UK, 1988

UK, 2030

UK, 1966

UK, 1977
US, 2016
US, 2024

Egypt, ?

Brazil, ?

India, ?

India, ?
$2
UK, ?

$0
Source: Britain Remade, World Nuclear News, Reuters, Bloomberg, AP News, IEEFA, 2023

What about the “once we build more of them, the price will go down” argument? In the US, nuclear construction
would have to really reignite for this effect to materialize. I’m not even sure that 5 plants a year would do it; a
pace of 5-10 plants per year might be needed for such synergies to appear.
# of US nuclear plants built by year of construction start Global nuclear power plants by year of completion
30 Gross GW
40
25 35

30
20
25
Projected,
15 90%+ EM
20

10 15

10
5
5
0 0
1940 1950 1960 1970 1980 1990 2000 2010 2020 2030 1950 1960 1970 1980 1990 2000 2010 2020 2030
Source: Power Reactor System Database, JPMAM, December 2023 Source: Power Reactor System Database, JPMAM, December 2023

18
Global investments in private nuclear fusion companies have now reached $6.2 billion. Four companies
expect to deliver power in 2030, and another 19 expect to deliver power in 2035. Hope springs eternal.
19
Flamanville (France) construction began in 2007; by 2020 it was 5x over budget. Project managers had to
address structural anomalies, faulty cooling welds and fire/explosions onsite. Operation is scheduled for 2024
after more delays and cost overruns. Hinkley Point (UK) is $13 billion over budget and several years late after
the 5th budget increase in 8 years. The total cost is now estimated at $46 billion, double original projections.
Vogtle 3 (Georgia, US) came in 7 years late and $16 billion over its original $14 billion budget. Olkiluoto
(Finland) was scheduled to be completed in 2009; it was completed in 2023 and cost $12 billion, three times its
original estimate. As shown on the next page, the levelized cost of these plants is much higher than our estimate
of a wind-solar-gas system with high renewable penetration and adequate thermal backup capacity.
24
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Nuclear power

If a country is willing to build nuclear fission plants, how long would it take to complete them? The chart on
the left shows how difficult a question is this to answer for the developed world. Average completion times 20
in developed countries ranged from 4 to 10 years for plants whose construction began from 1960 to 1990, but
there are so few observations for developed countries since 1990, it’s impossible to know. Plants have been
completed in 6-8 years in developing countries such as China, India and Russia but this has nothing to do with
what can be accomplished in the West.
Average nuclear plant completion time by year and region During the 1980's, global nuclear construction shifted from
Years developed to developing countries
20 Developed market share of global new nuclear MW, 5yr rolling avg.
18
Developed country 100%
16 Emerging country 90%
14 80%
12 70%
10 60%
50%
8
40%
6
30%
4
20%
2
10%
0 0%
1950 1960 1970 1980 1990 2000 2010 2020 1950 1960 1970 1980 1990 2000 2010 2020
Year of construction start Year of construction start
Source: Power Reactor System Database, JPMAM, December 2023 Source: Power Reactor System Database, JPMAM, December 2023

What happened to NuScale and its small modular reactor project?

NuScale had a lot going for it: a $1.4 billion cost-sharing deal with the DoE, and design approval from the NRC that
bypassed the typical lengthy site permitting process since its six small modular reactors were designed to be built on the
DoE’s Idaho National Lab site. NuScale claimed that construction costs would be below $3 mm per MW, that it would be
built in ~5 years and that it would run at a 95% capacity factor.
Eventually, NuScale scrapped plans to build SMRs when cost estimates soared to $20 mm per MW due in part to rising
reinforced concrete prices; its designs require more reinforced concrete per MW than conventional reactors. Even before
cost overruns, electricity purchase commitments covered less than 25% of NuScale’s output. Then as cost projections
rose, electricity buyers gave NuScale an ultimatum: boost commitments to 80% or terminate the project. In November
2023, NuScale conceded the target was unachievable and all parties determined the project should end. SMR hyperbole
was everywhere in 2021 but is quieter now. As I wrote last year, I never understood the logic behind SMRs. Nuclear
power is one of the most capital intensive of all industrial projects, and a significant part of that capital cost is fixed per
plant irrespective of its capacity. So why would shrinking the size of a nuclear fission plant be of any economic benefit?

NuScale Power share price Levelized cost of 4 specific nuclear plants vs cost of
Price, SPAC merger date May 2022 high renewable wind-solar-gas system, US$ / MWh
$16 $250
$14
Wind-Solar-Gas @6.9/mmbtu
$200 Wind-Solar-Gas @3.45/mmbtu
$12

$10
$150
France Flamanville

$8
Finland Olkiluoto
UK Hinkley Point

Georgia Vogtle

$6
EIA pro-forma

$100
$4

$2
$50
$0
Dec '20 Jun '21 Dec '21 Jun '22 Dec '22 Jun '23 Dec '23
Source: Bloomberg, JPMAM, February 16, 2024 $0
Source: EIA, JPMAM, 2024

20
Average completion times can obscure massive delays on individual projects. Example: Watts Bar-2
Tennessee (1973, 43 years); Diablo Canyon California (1968, 17 years); and Hartlepool UK (1968, 20 years)
25
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Nuclear power

To sum up, there’s reason to be cautious about the cost and timing required for the West to increase its baseload
nuclear power. That said, decommissioning existing nuclear fission plants is an entirely different matter. That
brings us to the case of Germany and its exposure to Russian energy: had it not decommissioned nuclear, how
exposed would Germany have been to Russia in 2022 when Russia invaded Ukraine?
First let’s look at what did happen. After Russia’s invasion in 2022, European electricity and natural gas costs
spiked as Russian pipeline imports were replaced by pipeline imports from Norway and expensive LNG imports.
While power prices have now fallen from peak levels, they’re still ~2x higher than pre-war levels.
A lot has been written on European de-industrialization due to higher energy costs. Impacts include declining
German production of energy-intensive goods and German industrial companies negatively affected by high gas
prices. Roughly one third of German industrial companies plan to or already did relocate21, and its petrochemical
plants run at just 75% of capacity. The IEA cites declines of ~6% per year in European industrial power demand
in 2022/2023 (demand destruction in chemicals, steel and aluminum), which has been the main reason for
European power demand falling to levels last seen two decades ago22. There is some good news: as power
prices declined, some curtailed production has been restarted.
EU gas vs German electricity prices European imports of natural gas and LNG
US$ per MMBTU US$ per MWh, 30 day average Million cubic meters per day
$30 $140 1,000
900
$25 $120
800
$100 700
$20
600 Africa
German electricity $80
$15 $72 500 Global LNG
$60 400
$10
$8 $40 300

$5 European natural 200

gas (TTF) $20 Russia North Sea
100
$0 $0 0
2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 Jan-21 Jul-21 Jan-22 Jul-22 Jan-23 Jul-23 Jan-24
Source: Bloomberg, JPMAM, February 12, 2024 Source: J.P. Morgan Commodities Research, January 2024

German industrial production

Index (100 = January 2021)
110
Motor vehicles
105
100
Other German industrial plants impacted by high gas costs
95 Capacities impacted in '000 tonnes per year
90 Plant Impact
85 BASF Ludwigshafen: Caprolactam (95),
Cyclohexanone (279), Adipic Acid (270), Shutdown on high
80 Energy intensive Ammonia (880), Melarine (65), Toluene production costs
75 products
Diisocyanate (300), Soda Ash, Cyclohexanol
70 Yara Brunsbuettel: Ammonia (750) 65% utilization
65 Dow Leuna, Schkopau: Polyethylene (385) 15% rate reduction
2021 2022 2023 Oxxynova: Dimethylterephthalate (240) Shutdown
Source: Greg Fuzesi, JP Morgan Economics, December 2023 Source: Mitsubishi UFJ Financial Group, December 2023

21
MUFG Energy 2024 Outlook, December 2023
22
“Electricity Market Report Update: Outlook for 2023 and 2024” and “Electricity 2024”, IEA
26
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Nuclear power

Now let’s look at what might have happened if Germany had not decommissioned nuclear. In 2002, Germany
had 24 GW and 164 TWh of nuclear generation which provided 28% of its power needs. After Japan’s Fukushima
meltdown in 2011, Germany gradually decommissioned its nuclear fleet. By the end of 2021, Germany had only
4.1 GW of nuclear capacity still operating (it has since fallen to zero). Assuming that its nuclear plants were
never decommissioned, we can estimate the capacity and generation Germany would have needed from fossil
fuels in 2022. Since Germany produces its own coal, we can also estimate imported natural gas Germany would
have needed for electricity generation.
Had Germany not decommissioned its nuclear, we estimate that Germany would have needed 50% less
electricity generation from fossil fuels and 84% less generation from natural gas in 2022. Large declines are
also seen in the counterfactual need for natural gas capacity on the right. Would more German nuclear power
have made a difference in 2022? Germany was the largest European importer of gas in 2022; on the margin, its
gas demand would have been much lower in this scenario. Would it have been enough of a decline in marginal
demand to reduce regional gas and electricity prices? Unclear, but it does seem like Germany’s political
decisions, domestic subsidy expenses and economic adjustments could have been less painful23.
Charts: how exposed would Germany have been to Russia in 2022 had it not decommissioned nuclear power?
German fossil fuel GENERATION required in 2022 with and German fossil fuel CAPACITY required in 2022 with and
without nuclear decommissioning, TWh without nuclear decommissioning, GW
275 60
Actual 2022 Actual 2022
250
225 50 Counterfactual assuming no
Counterfactual assuming no -27%
200 decommissioning of nuclear since 2002 decommissioning of nuclear since 2002
175 40
150 -49% 30
125
100 -42%
20
75
50 10
25 -84%
0 0
Electricity generation Electricity generation Fossil fuel Natural gas
from fossil fuels from natural gas generation capacity generation capacity
Source: Fraunhofer Institute, Energy Institute, JPMAM, 2023 Source: Fraunhofer Institute, Energy Institute, JPMAM, 2023

Germany installed nuclear capacity

GW
25

20

15

10

5
Fukushima nuclear
meltdown
0
2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2024
Source: Fraunhofer Institute, JPMAM, 2023

23
Some analysts link Germany’s decision to decommission nuclear power and increase natural gas reliance on
Russia with Ostpolitik (a deliberate plan to increase economic links with Russia with the goal of integrating it
into Europe). The “Putinverstehers” (Putin-admirers) were led by former German Chancellor Gerhard Schroder,
who earned millions in fees from Russian energy companies for promoting its interests. In an April 2022 NYT
interview after the apparent failure of Ostpolitik, Schroeder said “I don’t do mea culpa. It’s not my thing.”
27
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Nuclear power

One final look at nuclear decommissioning, this time from New York. In 2020 and 2021, New York State shut
down Indian Point’s nuclear plants with the intention of replacing its generation with renewable energy. That’s
not what has happened so far: three new natural gas plants (Bayonne Energy Center, CPV Valley Energy Center
and Cricket Valley Energy Center) have filled the gap along with mostly gas-fired electricity imports from states
like Pennsylvania. The chart below on GHG emissions illustrates how the closure of Indian Point has temporarily
pushed New York City regional emissions per MWh above the US average and above ERCOT. So, while Texas is
more “red” than New York City, it’s now more “green” as well.
The Champlain Hudson Power Express (Canadian hydropower) and the Clean Path project (upstate New York
wind/solar) should replace Indian Point’s lost generation. While Champlain Express is expected to start bringing
power to New York City in 2025, Clean Path is not set to deliver power until 2027. As shown below, both new
megaprojects are expected to exceed Indian Point’s lost generation as long as Canadian hydropower capacity
factors remain stable, and when assuming that CPNY wind/solar capacity factors are equal to those on existing
NY projects. [Note: a prior version of this paper incorrectly estimated these figures and has been revised
accordingly]. New York has an advantage over Massachusetts and other Eastern states that need new clean
energy capacity since it shares a border with Canada and can decide unilaterally to approve projects like CHPE.
As we have written in prior years, Massachusetts has seen two separate Canadian hydropower import projects
fail due to objections from neighboring states regarding construction of high voltage direct current lines. These
states are attempting to build offshore wind but are running into challenges there as well (see page 34).
New York: electricity generation by source Wind and solar capacity factors by state
Terawatt hours Wind Solar
60 Natural
Gas 7 highest 7 lowest 7 highest 7 lowest
50 State CF State CF State CF State CF
1 NE 46% WV 27% UT 29% MA 18%
40
2 ND 43% NH 26% NV 28% NY 17%
30 3 IA 43% MA 25% AZ 28% PA 17%
Nuclear
Hydro 4 SD 43% NV 24% NM 27% WI 17%
Imports 5 KS 42% NY 24% CA 27% VT 17%
20
6 MT 39% OR 23% TX 26% RI 17%
10 Wind &
Solar 7 IL 38% UT 21% AR 25% NJ 17%
0 Coal Source: EIA forms 860/923, JPMAM, 2023. All states w ith more than
1990 1995 2000 2005 2010 2015 2020 50 MW of solar and more than 100 MW of w ind
Source: EIA, JPMAM, 2023

The GHG impact of shutting down Indian Point Planned replacement of Indian Point generation
CO2 emissions, pounds per MWh TWh
1,000 18
16
900 NYC / Long Island
14
Champlain Express hydro
800
US 12 (1.25 GW)
10 Lost power from
ERCOT (TX)
8 Indian Point nuclear
700
(2 GW, 90% capacity
6 factor)
600
4 Clean Path wind/solar (3.8
GW, 20% capacity factor)
2
500
2019 2020 2021 2022 0
Source: EPA eGRID data, 2022 Source: JPMAM, 2024

28
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Lazard epiphany

Lazard’s epiphany does not go far enough: the inadequacies of levelized cost, Part 2
Last year, I wrote a long section on the problems with levelized costs of energy when applied to renewables. I
described LCOE as the cocktail napkin of energy math24 and cited Paul Joskow at MIT who began to write about
this over a decade ago (see text box). Not much has changed, other than a gradual recognition by more analysts
that LCOE is a mostly useless measure when comparing renewables to baseload power.
The most committed disciples of LCOE over the last two decades have been Lazard’s energy team, propagators
of their annual LCOE report. After 16 years, the Lazard team has had an epiphany: while it’s not incorporated
in their base case figures which are still of little practical use, Lazard now includes a new supplemental exhibit
on the cost of “firming the intermittency” of wind and solar power. In other words, the costs required to provide
power when there’s not enough wind or sun to meet the load demand. For MISO, SPP and PJM regions, Lazard
incorporates the cost of peaker combustion turbines into wind and solar costs; and for CAISO, they incorporate
the cost of utility scale energy storage, presumably fueled by overbuilding solar power. In their new exhibit,
some revised LCOE estimates for wind and solar power are at or above median costs for combined cycle natural
gas plants (their firming costs are shown by the gold bars). To be clear, should storage and solar costs continue
to decline, so would Lazard’s respective firming costs in regions like CAISO.
There are still odd things in Lazard’s report. Lazard assumes an operating life of only 20 years for new natural
gas combined cycle and combustion turbine plants25. This is strange, particularly since the EIA reports the
average age of existing natural gas plants at 22 years, and the EIA assumes 30-year cost recovery for new ones.
Sargent & Lundy consults for the EIA on LCOE estimates and confirmed that they assume 30-40 years for natural
gas plant operating lives. A longer assumed operating life would reduce natural gas LCOE estimates. This
assumption is another sign of how Lazard sometimes tilts the playing field in their analysis, and explains why it
took 16 years for firming costs to show up in their report.
Levelized costs of energy: Incorporating the cost of backup thermal power and energy storage
US$ per MWh
$225

$200 Gas Peaking

Firming cost: combustion turbine for SPP, MISO and PJM; storage for CAISO ($115 - $221/M Wh)
$175
Unsubsidized LCOE
$150

$125

$100
Gas Combined Cycle
$75 ($39 - $101/M Wh)

$50

$25

$0
Solar Wind Solar PV + Wind Solar Wind Solar PV + Wind
Storage Storage

MISO CAISO SPP PJM

Source: "Lazard's Levelized Cost of Energy Analysis - Version 16.0", Lazard, April 2023.

Paul Joskow (MIT): LCOE is “inappropriate for comparing intermittent generating technologies like wind
and solar with dispatchable generation…and also overvalues intermittent generating technologies
compared to dispatchable baseload generation”…“LCOE comparisons of baseload and intermittent, non-
dispatchable generation make little sense; what’s needed instead is a system-wide model rather than
simplistic LCOE calculations”

24
“Growing Pains”, 2023 Eye on the Market Energy Paper, pages 14-18
25
“LCOE Plus”, Lazard, April 2023, page 39
29
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Advanced
electrification topics
14th Annual Energy Paper

Advanced Electravision topics: timing, temperature, transmission and turbines

Electravision entails more than just building new renewable generation capacity and transmission lines.
Consumers of electricity may have to adjust to new pricing models for power since the time of day when they
consume it will be more important than it is today. Furthermore, estimating emissions benefits from
Electravision EV and heat pump adoption will require more precise analysis than current models. This section
gets into advanced topics: real-world EV emissions; capacity and transmission requirements as a function of
driver and homeowner electricity use; and the challenges of offshore wind development in the US.
[a] Estimated emissions reductions from EV adoption may be too optimistic unless they reflect “marginal”
emissions rates rather than “average” emissions rates
Many studies compare lifecycle emissions of EVs to internal combustion engine cars. The general conclusion:
after around two years of operation, the average EV matches the carbon intensity of the average ICE car or light
truck. After that, the EV accrues net carbon benefits through to the end of its life. The initial EV carbon deficit
reflects the carbon intensity of battery manufacturing. One example: a 2022 piece from the Union of Concerned
Scientists26. The first chart shows UCS estimated lifetime net emissions benefits for EV cars and light trucks
assuming the emissions intensity of the current population-weighted electricity grid, and a future grid that is
95% renewable. Based on UCS estimates, after 15k-20k miles (around two years of driving), EV trucks and cars
accrue net emissions benefits vs ICE vehicles. This is a consistent finding across many studies27.
Cumulative emissions benefits from EVs vs ICE vehicles Lifecycle emissions, EVs vs ICE cars and light trucks
Tonnes CO2e net emissions benefit Emissions, grams of CO2e per mile
100 700
95% ren
Vehicle manufacturing
600
75 Battery manufacturing
Current 500
Vehicle operation
50 95% ren 400

Current 300
25
200

0 100
Initial EV deficit due to battery manufacturing 0
-25 ICE car Electric car ICE truck Electric truck
0k 25k 50k 75k 100k 125k 150k 175k 200k (32 MPG) (300-mi range, (19 MPG) (300-mi range,
Miles driven 0.31 kWh/mi) 0.48 kWh/mi)
Source: "Driving Cleaner", Union of Concerned Scientists, July 2022 Source: "Driving Cleaner", Union of Concerned Scientists, July 2022

However: there’s an issue that needs to be further explored. UCS EV CO2 estimates are based on the average
emissions rate of the current grid: i.e., the sum of all CO2 emissions from electricity generation divided by total
MWh. But for the next EV added to the grid, what matters is marginal emissions rates (i.e., what does the grid
look like when EVs are actually charged) and not average emissions rates. UCS is aware of this distinction but
uses average emissions rates due to an inability to “make assumptions about charging behavior or the response
of electricity generation units to an increase in demand”. We agree that the former is hard to do, but some
analysts have made progress on the latter. This is particularly important for the future when more EVs are
anticipated as a share of the fleet.

26
“Driving Cleaner: How Electric Cars and Pick-Ups Beat Gasoline on Lifetime Global Emissions”, Union of
Concerned Scientists, Reichmuth et al, July 2022
27
Similar findings to UCS results: “Comparative lifecycle GHG emissions of a mid-sized BEV and ICE vehicle” from
the IEA; BNEF’s Electric Vehicle Outlook; and a study from the European Federation for Transport & Environment
30
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Advanced
electrification topics
14th Annual Energy Paper

Researchers at Convergent Science estimated upper and lower bounds of marginal emissions rates by US region,
calculated over every hour in a given year. Each estimate reflects the energy sources that are most correlated
with changes in marginal demand. As shown, estimated marginal emissions rates for the US are 1.5x – 1.8x
higher than average emissions rates28.
Average vs marginal emissions rate, kg of CO 2 per MWh
Lower bound Upper bound
MARGINAL MARGINAL Ratio of marginal
Geographical AVERAGE emissions rate (FF emissions rate rates to average
region emissions rate and renewables) (FF only) rates
Pacific Northwest 186 92 584 0.5 - 3.1
California 196 396 469 2.0 - 2.4
Texas 401 508 529 1.3 - 1.3
New England 216 408 433 1.9 - 2.0
Midwest/South 560 749 766 1.3 - 1.4
New York 174 478 520 2.7 - 3.0
Mid-Atlantic 395 690 724 1.7 - 1.8
South 454 687 704 1.5 - 1.5
Midwest 492 523 762 1.1 - 1.5
Source: Tristan Burton and Kelly Senecal (Convergent Science), April 13, 2022

How would marginal emission rates impact the UCS analysis? Let’s assume that on a national population-
weighted basis that marginal emissions rates are 1.5x higher than average rates (and that’s before incorporating
higher emissions rates of battery manufacturing in China). The next chart shows the revised net emissions
benefits for EV vs ICE sedans assuming marginal emissions rates. The result: the lifetime net emissions benefit
for the sedan falls roughly in half from 32 tonnes of CO2 to 17 tonnes instead.
The red line is based on average marginal emissions rates over the entire year; as a result, it assumes no ability
to influence the timing of EV charging through pricing incentives. Furthermore, it uses today’s grid and does
not account for the pace of wind and solar additions which are growing faster than EV adoption. The bottom
line on EV emissions: more wind and solar, time-of-use electricity pricing incentives, centralized charging
programs (see box) and more public/private charging infrastructure whose use coincides with solar power
peaks would increase EV net emissions benefits closer to UCS assumed levels.
Cumulative emissions benefits from EVs vs ICE sedan
Tonnes CO2e net emissions benefit Centralized EV charging optimization
50 95% ren Seven leading US utilities are conducting pilot programs
40
to remotely control time-of-day charging of EVs to
Current:
average
reduce peak load and transmission needs. The remote
30 emissions access takes place through the charging device or
20 Current:
wirelessly with the vehicle. Many utilities are also
marginal piloting distributed energy resource management
10 emissions
systems which remotely control behind-the-meter
0 thermostats, water heaters and batteries
Initial EV deficit due to battery manufacturing Source: Darcy Insights “Benchmarking Study, Managed
-10
0k 25k 50k 75k 100k 125k 150k 175k 200k EV Charging Programs”, February 2024
Miles driven
Source: UCS, Convergent Science, JPMAM, July 2022

28
“Data-Driven Greenhouse Gas Emission Rate Analysis for Vehicle Comparisons”, Burton and Senecal, 2022
31
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Advanced
electrification topics
14th Annual Energy Paper

[b] Clustering of charging behavior can substantially increase peak loads and the need for transmission
Researchers at the Technical University of Denmark analyzed what would happen to the grid if EVs, currently
10% of the fleet in Denmark (BEV+PHEV), reached 90%+ of the fleet29. For example, if all EV charging were
evenly distributed throughout the day, incremental capacity needs would only be 1 GW and peak loads would
only rise by 17%. At the other end of the extreme, if every EV owner charged at the highest possible power30 at
the same time of day, peak loads would increase by 550% with an additional 33 GW of capacity required. That
is an absurdly pessimistic scenario, but so is the first one which is too optimistic.
Denmark: capacity and peak load increases at 100% EV penetration
Additional Increase in peak
Confluence of EV charging Key assumptions
power required consumption
Evenly distributed charging 1.0 GW 17% 15,000 km driven per year per vehicle
3.7 kW @ 30% coincidence factor 3.3 GW 55% EV mileage = 5 km per kWh
3.7 kW @ 100% coincidence factor 11.1 GW 185% Denm. electricity consumption = 36 TWh
11 kW @ 100% coincidence factor 33.0 GW 550% Denm. peak power consumption = 6 GW
Source: Mattia Marinelli, Technical University of Denmark, November 2023 Denm. current generation capacity = 18 GW
The reality lies in between, highlighting the need for pricing policies for EV charging that do not just reflect the
availability of surplus renewable energy, but also reflect the impact of EV charging on peak loads, capacity
needs and transmission infrastructure. This is precisely what Denmark now does; residential customers pay
four separate components on their electricity bills: a spot price which depends on the day-ahead market; a grid
impact fee which depends on the month/hour of the day; a fixed energy fee; and a VAT tax. The second
component is the key one and contrasts with the US where the grid/infrastructure component of electricity bills
is typically fixed irrespective of when electricity is consumed. On the right: note how electricity prices can spike
substantially in Denmark based on the time of day, which ends up shifting consumption. Synergi, a Finnish
electricity management company, reports similar results: EV owners using smart charging can reduce their
annual energy bills by 70% as their charging is typically deferred to 11 pm – 3 am.
Denmark grid infrastructure fees are time-of-day dependent Danish customers respond to higher electricity prices
Retail price, Danish Krone per kWh, October-to-March average day Y/y change in electricity consumption (H1 '23) Change in prices
2.5 15.0% 60%
Value-added tax 12.5% Actual electricity 50%
consumption
2.0 Flat energy fee 10.0% 40%
Grid infrastructure fee 7.5% 30%
1.5 5.0% 20%
2.5% 10%
1.0 0.0% 0%
-2.5% -10%
-5.0% Electricity prices -20%
0.5
-7.5% -30%
-10.0% -40%
0.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 1 3 5 7 9 11 13 15 17 19 21 23
Hour of day Hour of day
Source: Marinelli (Technical University of Denmark), August 25, 2023 Source: Marinelli (Technical University of Denmark), August 25, 2023

29
Mattia Marinelli (Technical University of Denmark), DTU Wind and Energy Systems, November 2023
30
As a reminder, watts = volts times amps. The 3.7 kW scenario is based on 230 volts (typical voltage in Europe)
times 16 amps (typical max for domestic appliances) times 1 phase. The 11 kW scenario is based on 230 volts
times 16 amps times 3 phases. Three-phase and single-phase power refer to different wiring configurations; in
the US, single-phase is used in most residences while three-phase is normally reserved for commercial and
industrial users with heavier loads. But in Europe, many Nordic and German residences use three-phase power
as well for cooking and heating.
32
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Advanced
electrification topics
14th Annual Energy Paper

[c] European benefits from heat pump adoption might be partially reduced by increased air conditioning use
Europeans live in colder climates than the US. While only 8% of the US population lives above the 44th parallel,
82% of Europe’s population lives above that level. That’s one reason why air conditioning use in Europe has
historically been so much lower than in the US. Only 1 in 10 households has air conditioning in Europe vs 90%
in the US, and air conditioning represents only 1% of building energy use in Europe vs 16% in the US.
But: two things are changing. First, Europe is getting hotter, as shown in the first chart on rising cooling degree
days. And second, Europe is installing more heat pumps for winter heating, most of which can be used for air
conditioning as well. It will be important to monitor the degree to which increased air conditioning use offsets
some of the benefits of more efficient, less emissions-intensive winter heating in Europe.
Increased need for cooling in the EU Household electricity use for air conditioning
Cumulative cooling degrees per year, Celsius kWh per dwelling
160 2,000
Cooling degrees: difference
140 1,800
between avg daily temperature
and 24 C for all days where avg 1,600
120 temperature exceeds 24 C
1,400
100
1,200
80
1,000

European Union
60 800

Luxembourg

Netherlands
Switzerland
40 600

Germany
Slovenia

Hungary
Portugal
Bulgaria

Belgium
Czechia
Croatia
Cyprus

Austria
Greece

France
20 400

Spain
Malta

Italy
200 US
0
'79 '82 '85 '88 '91 '94 '97 '00 '03 '06 '09 '12 '15 '18 '21 0
Source: Eurostat, JPMAM, 2022 Source: EIA, OdysseMure, JPMAM, 2019

[d] Smart thermostats help conserve energy by encouraging people to use less heat and air conditioning when
they’re not home or asleep. But if too many smart thermostat setpoints coincide, there could be a
concentrated burst of power demand that grids must be designed to accommodate31
Researchers looked at time-of-day setpoints used by homeowners with smart thermostats. The charts show
average winter setpoints and indoor temperatures. Setpoints to increase heat are concentrated around 7 am,
which may reflect default settings rather than user choices. Setpoint timing is important from a load perspective
since more power is required to increase temperatures to a given level than to sustain them there. And: peak
smart thermostat demand tends to be concentrated around times of low renewable resource availability (7 am,
8 pm). Smart thermostats will need to be accompanied by policies that avoid time-of-day concentrations. Some
real-world trials involving price signals have shown high responsiveness; but re-entry loads would have to be
properly managed or else you end up with the same peaking problem pushed out to another time of day.
Winter weekday average setpoints and indoor temperature Winter weekend average setpoints and indoor temperature
of smart thermostat users, Degrees Celsius of smart thermostat users, Degrees Celsius
20.0 20.0
Indoor temperature
19.8 19.8
19.6 Indoor temperature 19.6
Setpoint
19.4 19.4
19.2 19.2
19.0 19.0
Setpoint
18.8 18.8
18.6 18.6
18.4 18.4
0:00 4:00 8:00 12:00 16:00 20:00 0:00 0:00 4:00 8:00 12:00 16:00 20:00 0:00
Time of day Time of day
Source: Cornell University, Lee & Zhang, September 15, 2022 Source: Cornell University, Lee & Zhang, September 15, 2022

31
“Unintended consequences of smart thermostats in the transition to electrified heating”, Cornell, Lee & Zhang, 2022
33
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Advanced
electrification topics
14th Annual Energy Paper

[e] Offshore wind turbulence in the US contrasts with greater capacity growth elsewhere
While the US has 146 GW of onshore wind, it has almost no offshore wind. Some Northeastern states aim to
develop their own offshore wind given failure of projects to import Canadian hydropower. However, the last
12 months have seen US offshore wind developers walk away from 8.5 gigawatts of wind projects due to
withdrawal of original bid submissions after as much as 50% inflation in component costs. These projects are
likely to proceed, but it will take more time and higher power costs. The US is an underachiever on offshore
wind compared to countries below. NREL highlights the supply chain buildout required to meet the Biden
administration’s 30 GW offshore wind target by 2030: a lot of new factories and ~$22 bn in capital investment32.
The global wind turbine market is split into two groups: ~15 Chinese manufacturers which mostly supply its
domestic market, and Western firms that supply everyone else (GE, Vestas, Nordex and Siemens Gamesa). As
Western turbine makers experience cost and margin challenges, Chinese firms are increasingly competitive in
Western markets. As shown below, S&P Global estimates that Chinese turbines are 70% cheaper.
One more thing on offshore wind. Research from Europe shows that offshore wind farms can generate “wakes”
which reduce wind speeds at adjacent wind farms, even as far away as 55 km. The decline in offshore wind
capacity factors can be as much as 20%, resulting in production and revenue losses33. The study shows that
most offshore wind farms in Northern Europe are less than 50 km away from the nearest wind farm.
Offshore wind share of total wind capacity 8.6 GW of US offshore wind capacity: already cancelled or
50% expected to cancel, Gigawatts
9
45%
8 Empire Wind 1
40%
7
Sunrise Wind
35% Park City Wind
6
30%
Commonwealth Wind
25% 5

4 Beacon Wind 1
United Kingdom

20%
South Korea
Netherlands

15% 3 Empire Wind 2

Germany
Denmark

Portugal
Vietnam

Sweden
Belgium

Norway

10% 2
Taiwan

Ireland
Japan
China

Spain

Dots: Expected to
5% SouthCoast Wind
US

1 cancel
0% Solid: cancelled
0
Source: IRENA Energy, JPMAM, 2022 Source: Intelatus Global Partners, Individual project reports, JPMAM, 2024

US offshore wind supply chain requires at least $22bn in Average selling price of onshore wind turbines
investment to meet 2030 target, Cumulative investment, US$ bn US$ per watt
$24 $1.2 Western
COVID starts Russia-Ukraine manufacturers
conflict
$20 Installation $1.0
vessels

$16 $0.8
Ports
$12 $0.6
Wind turbines
$8 $0.4
Substructures
Electrical
$4 $0.2 Chinese
Steel plates
manufacturers
Other
$0 $0.0
2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2019 2020 2021 2022 2023
Source: NREL, January 2023 Source: S&P Global, Q3 2023

32
“A Supply Chain Road Map for Offshore Wind Energy in the United States”, NREL, Shields et al, January 2023
33
“Wind farm-induced wakes and regulatory gaps”, University of Bergen (Norway), Finseras et al, October 2023
34
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Why-drogen Epilog

Why-drogen Epilog: A tough year for the green hydrogen industry

In 2021 I wrote a skeptical long-form section entitled “Why-drogen” given all the unanswered questions on
green hydrogen production, transmission and consumption. Since then, the hydrogen economy has seen some
advances: Saudi Arabia’s 2.2 GW NEOM ammonia/hydrogen project based on wind/solar and Shell’s 200 MW
Holland Hydrogen project based on offshore wind are now under construction. And in the US, there’s 10 million
tonnes per annum of planned green/blue hydrogen production, which would be enough to replace almost all
US hydrogen consumption that is currently produced by natural gas and coal…if it gets built.
But there have also been a lot of cancelled projects due to the realization that per unit of energy, hydrogen for
road transport or winter heating is much less efficient than directly electrified alternatives. While developers
lined up at the public trough for $350 billion in global hydrogen subsidies in 202334 (tax credits, grants and R&D
funding), the energy math usually argues against it other than for green hydrogen to replace fossil fuel-derived
hydrogen currently used in ammonia/fertilizer, in oil refining and in direct reduction of iron ore to iron metal.
This section walks through the latest developments, most of which imply that “No-drogen” is the common
answer to “Why-drogen”.
How much electricity is available to power an electric How much heating value is available when starting with
motor when starting with 100 kWh of wind power? kWh 100 kWh of wind power? Trillion BTU
80 1.1
70 1.0
0.9
60 0.8
50 0.7
0.6
40
0.5
30 0.4
20 0.3
0.2
10 Hydrogen fuel Electric Hydrogen Electric heat
0.1 boiler pump
cell vehicle vehicle
0 0.0
After electrolysis, compression, fuel After transmission, charging, electric After H2 compression, transport, After transmission losses and
cell losses, elec. motor losses motor losses electrolysis, boiler efficiency coefficient of performance
Source: EIA, DoE, DIW Berlin, JPMAM, 2023 Source: EIA, DoE, DIW Berlin, JPMAM, 2023

Hydrogen Index returns

Total return index, November 2020 = 100
180
Wilderhill Hydrogen Index
160 Solactive World Hydrogen Index
140

120

100

80

60

40
Nov-20 May-21 Nov-21 May-22 Nov-22 May-23 Nov-23
Source: Bloomberg, JPMAM, March 1, 2024

34
BNEF Hydrogen Subsidy Tracker, January 25, 2024
35
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Why-drogen Epilog

Low demand for electrolyzers and green hydrogen

• Some electrolyzer manufacturers are downgrading production capacity plans due to policy delays and
insufficient green hydrogen demand35. Only 5% of green hydrogen projects have reached operation; and
only 10% of announced green hydrogen capacity has confirmed buyers36,37. As a result, share prices of some
electrolyzer manufacturers have fallen by 95% from 2021 levels (Plug Power, ITM Power, McPhy Energy)
• The EU established a 1.2% blending requirement for synthetic aviation fuel sourced from green hydrogen
by 2030 and 35% (!!) by 2050. Synthetic fuel producers note that airlines are reluctant to pay a green
premium, and are struggling to secure offtake agreements with airlines38
Rising electrolyzer costs and technological problems
• A March 2024 survey of alkaline and PEM electrolyzers found 46%-65% price increases since 2022, part of
which is explained by slower-than expected market scaling39
• The world’s largest green hydrogen project built in Xinjiang, China is operating at less than a third of installed
capacity due to technical problems. In the second half of 2023, this 260 MW electrolyzer was expected to
produce ~9,500 metric tons of green hydrogen; it produced only ~2,000 metric tons instead. In addition to
these technical issues, the plant is missing some safety and flexibility features promised in the contract
• While Chinese electrolyzers are much cheaper than Western counterparts, savings to Western project
developers are unlikely to offset concerns about quality, financing issues and policy risks
Rising costs of green hydrogen, matching requirements for tax credits and the need for subsidies
• BCG raised its expected minimum green hydrogen price target for 2030 from 3 euros/kg to 5 euros/kg40
• In 2020, ArcelorMittal announced new green steel plants with great fanfare; they would eventually use
green hydrogen as a reducing agent for iron ore instead of carbon. Arcelor now concedes that green
hydrogen is too costly to use and would render its steel uncompetitive, even after taking European subsidies
into account. The company will either use fossil gas in a direct reduced iron (DRI) reactor or import green
DRI from elsewhere, after which both approaches would rely on an electric arc furnace to produce steel
• The US Treasury ruled that hydrogen projects must adopt hourly matching of hydrogen production and
green energy sourcing by 2028 to qualify for tax credits. This immediately set off complaints by the hydrogen
industry which apparently wants tax credits for creating brown hydrogen while claiming that it’s green
• BNEF estimates that for the Hydrogen Energy Ministerial’s 2030 targets to be met, the hydrogen industry
would need over $2 trillion in subsidies41
Overestimated GHG benefits
• The US Department of Energy concluded that the emissions footprint of most blue hydrogen projects (brown
hydrogen + CCS) is too high to merit tax credits42
• A 30% green hydrogen blend in natural gas pipelines may only result in a 6% GHG decline due to increased
need for compression stations and due to hydrogen leakage43

35
“Green hydrogen electrolyzer makers putting brakes on capacity expansion”, Hydrogen Insight, Sept 2023
36
“Five Strategies for Optimizing Power-to-X Projects”, BCG and Oxford Global Projects, Sept 2023
37
“Hydrogen Demand: Tiny but Rising”, BNEF, Gao, November 2023
38
“Airlines are not willing to pay a premium for green hydrogen-based e-kerosene”, Hydrogen Insight, January 2024
39
“Electrolyzer price survey 2024: Rising Costs, Glitchy Tech”, BNEF, March 1, 2024
40
“Turning the European Green Hydrogen Dream into Reality: A Call to Action”, BCG, October 2023
41
“Clean Hydrogen’s Missing Trillions”, Michael Liebrich, BNEF, December 2023
42
“Blue hydrogen unlikely to qualify for US tax credits due to high upstream emissions”, Hydrogen Ins., Dec 2023
43
“Hydrogen blending in gas pipelines faces limits due to leakage: US DOE lab”, S&P Global, October 27, 2023
36
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Why-drogen Epilog

For road transport, hydrogen is uncompetitive with directly electrified alternatives

• Maersk’s APM Terminals division found that hydrogen costs for container handling equipment (tractors,
carriers and stackers) are 25%-60% higher than battery powered equipment due to greater efficiency losses,
higher fuel costs and greater equipment complexity; and concluded that these cost gaps will persist44
• Hydrogen truck makers Hyzon and Nikola both face delisting by Nasdaq for consistently trading below $1
Challenges for hydrogen shipping
• Green hydrogen’s arguably best new use case, maritime shipping, faces huge obstacles. When compressed
and chilled to 20⁰C above absolute zero (which would consume 33% of the energy in hydrogen), hydrogen’s
energy density by volume is still 4x less than maritime fuel; if converted into ammonia fuel, the energy
conversions are expensive; there are safety concerns if used in confined ship engine rooms45; and any
hydrogen leakage could have negative consequences for the ozone since global warming impacts of H2 are
12x higher than CO246. The cost challenge for shipping: even after adjusting for hydrogen’s higher energy
density by mass compared to diesel, green hydrogen could still cost 5x-6x more than maritime fuel47
Cancelled and suspended hydrogen projects48
• Electrolyzers: ATCO (Canada), Schleswig-Holstein (Germany). Cited reasons: distance between hydrogen
production facility and end use undermines commercial viability49, high construction costs
• Buses/trucks: Liverpool (UK), Glasgow (UK), Montpellier (Fr). Cited reasons: unreliable green hydrogen
supply, hydrogen vehicles as much as 6x more expensive than battery electric models
• Trains: Germany, Netherlands. Cited reasons: cheaper battery electric models, no bids by developers
• Blue hydrogen/ammonia: Nutrien (Louisiana), Shell-UK National Grid. Cited reasons: higher than expected
capital costs, uncertain future demand for clean ammonia, no support for heating/blending
• Residential heating: UK. Cited reasons: insufficient green hydrogen supply
• Green hydrogen-to-methanol: Belgium. Cited reasons: escalating costs, no long-term buyers
Next year we will look at the question of naturally occurring “white” or “gold” hydrogen deposits. By then,
the 50+ companies which have begun to look for them will have presumably either confirmed its abundance or
not. The USGS believes that there might be some in the Atlantic coastal plain, the central US, parts of the Great
Plains and the Upper Midwest. The USGS has committed to publishing its global resource potential models and
a preliminary map of the areas that are most likely to contain geologic hydrogen resources later this year50.

On shipping and fossil fuels. In 2021, 36% of all maritime shipping tonnage by weight was the movement of
oil (15%), coal (11%), natural gas (5%) and refined petroleum products (5%). For this reason, some analysts
note that if energy use for road transport, heating and industrial production were substantially decarbonized,
energy use for shipping would simultaneously fall as well. True enough, but as shown in the introduction there
are few signs that global consumption of fossil fuels is in decline. It could take decades for demand for maritime
fuels to materially decline due to decarbonization.

44
“Reaching a tipping point in Battery-Electric Container Handling Equipment”, APM Terminals, October 2023
45
“Lies leaks and lethality: is ammonia a safe fuel for ships?”, The Lodestar, August 6, 2023
46
“New study estimates global warming potential of hydrogen”, CICERO, June 7, 2023
47
“Hydrogen Half Truths Keep Shipping Fuel Hopes Afloat”, Michael Barnard, Forbes, December 29, 2023
48
Hydrogen Insight, Renew Economy, Glasgow Live, Recharge
49
Since hydrogen is very difficult to transport, ~85% of hydrogen is consumed at the point of manufacture
50
“The potential for geologic hydrogen for next-generation energy”, US Geological Society, April 13, 2023
37
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
Ignore the PUNIs
14th Annual Energy Paper

“Ignore the PUNIs”: The distraction of small countries with unique natural energy resources
While China merits the attention it gets on its energy transition, some very small countries do not. One example:
press articles on the high share of renewable power that some countries generate in one hour or one day of the
year without mentioning how the rest of the year’s energy needs are met. Or: articles highlighting renewable
transitions in countries with unique natural resources without acknowledging that they’re often irrelevant to
the rest of the world. The acronym “PUNI” refers to Panama, Uganda, Norway and Iceland, prime examples of
this phenomenon. Let’s see why.
The first chart shows each country’s renewable share of electricity generation in green alongside a cumulative
tally of global electricity consumption. Key point from the chart: there are 62 countries with renewable shares
of electricity over 60%, but they only account for 9% of global electricity consumption (Brazil and Canada
account for around half of the electricity consumption in this cohort). Now let’s see how this cohort is
accomplishing such high levels of renewable electricity generation. Hint: it’s not wind or solar.
62 countries with renewable electricity shares over 60%
account for just 9% of global electricity consumption
100%
Country renewable
90% share of electricity Cumulative share of
80%
global electricity
70% consumption
60%
50%
40%
30%
20%
10%
0%
0 20 40 60 80 100 120 140 160 180 200 220
Countries ranked by descending renewable share of electricity generation
Source: IEA, IRENA, World Bank, Harvard ECI, ESMAP, JPMAM, 2022

It's all about hydropower. The table shows how this 62-country cohort heavily relies on hydropower for its
electricity generation; wind plus solar account for just 11%. Is there potential for even more hydropower? The
International Hydropower Association estimates that global capacity could grow by 3x51. But as we reviewed in
2016, countries like the US have already exploited most of its readily available hydropower resources. Most
hydropower potential studies are done at the topographical reconnaissance level and do not incorporate real-
world constraints related to cost, local politics or environmental impact. That may explain why global
hydropower generation is only growing at ~1% per year when excluding China, and why the OECD has seen no
hydropower growth at all since the year 2000.
Electricity generation of aggregate cohort Ex-China, global hydropower growing at 1% per year since
Countries with > 60% of electricity from renewables 2000; no growth in OECD at all, Terawatt hours
Source TWh Share of TWh 4,500
Global
Hydropower 1,673,644 63% 4,000
Fossil fuels 363,855 14% 3,500 Global ex
Wind 230,857 9% China
3,000
Nuclear 198,100 7%
Biomass/biogas 126,528 5% 2,500

Solar 60,363 2% 2,000

OECD
Geothermal 9,691 0% 1,500
Source: IRENA, JPMAM, 2022
1,000

500
1965 1972 1979 1986 1993 2000 2007 2014 2021
Source: EI Statistical Review of World Energy, JPMAM, 2023

51
“Hydropower 2050: Identifying the next 850+ GW towards Net Zero”, International Hydropower Assoc., 2021
38
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
Ignore the PUNIs
14th Annual Energy Paper

Are there countries with high renewable shares that do not just rely on hydropower, geothermal and biomass
(HGB)? Just a few: Namibia, Uruguay, Denmark, Lithuania and Portugal where wind and solar provide at least
30% of electricity generation52. Note that they also benefit from substantial HGB contributions as well.
Renewable and wind/solar shares of electricity generation
Cohort: all countries with at least 60% of electricity generation from renewable energy
Ren W/S Ren W/S Ren W/S Ren W/S Ren W/S
share share share share share share share share share share
1 Eswatini 100% 0% 13 Namibia 98% 35% 25 El Salva 86% 25% 37 Cameroon 80% 0% 50 Croatia 67% 18%
2 Lesotho 100% 0% 14 Greenlan 97% 0% 26 Kyrgyzst 86% 0% 38 Malawi 79% 0% 51 Fiji 67% 1%
3 Bhutan 100% 0% 15 Central 97% 0% 27 Luxembou 85% 43% 39 Austria 79% 16% 52 Burundi 66% 0%
4 Nepal 100% 0% 16 Belize 93% 2% 28 Mozambiq 85% 1% 40 Colombia 77% 1% 53 Lithuani 66% 45%
5 Iceland 100% 0% 17 Andorra 93% 1% 29 Uruguay 85% 34% 41 Sierra L 75% 2% 54 Pakistan 65% 9%
6 Paraguay 100% 0% 18 Zambia 93% 1% 30 Venezuel 84% 0% 42 Guatemal 73% 4% 55 Honduras 64% 16%
7 Ethiopia 100% 5% 19 Tajikist 93% 0% 31 Panama 83% 12% 43 Mali 73% 3% 56 Latvia 64% 3%
8 Albania 100% 1% 20 Ecuador 92% 0% 32 New Zeal 83% 7% 44 Zimbabwe 71% 1% 57 North Ko 63% 0%
9 Costa Ri 100% 13% 21 Tokelau 92% 92% 33 Denmark 82% 60% 45 Laos 71% 0% 58 Monteneg 62% 9%
10 Norway 99% 10% 22 Kenya 89% 5% 34 Angola 82% 0% 46 Nicaragu 70% 16% 59 Peru 61% 5%
11 Democrat 99% 0% 23 Guinea 88% 0% 35 Georgia 82% 1% 47 French G 70% 6% 60 Sudan 61% 0%
12 Uganda 98% 2% 24 Brazil 88% 17% 36 Afghanis 81% 3% 48 Sweden 69% 21% 61 Portugal 60% 37%
Source: IRENA, JPMAM, 2022 49 Canada 69% 7% 62 Switzerl 60% 7%

Are these five countries relevant paradigms for the world’s largest energy consumers? Not really. They’re all
tiny in terms of energy consumption and population. When compared to larger countries, Namibia has much
higher solar irradiance; Denmark and Uruguay have much higher wind potential; Uruguay and Namibia have
much lower population density as well as lower economic complexity. This latter figure measures each country’s
ability to produce complex products across industries, which in turn drives the need for more developed energy
systems. Some also benefit from proximity to large countries for grid stabilization (Uruguay/Brazil,
Denmark/Germany). If you’re keeping score, that makes six separate factors which reduce Uruguay’s relevance
to larger countries. Portugal is the only one of the five without unique natural resource advantages.
PUNI countries deserve credit for developing wind, solar and HGB resources and reducing reliance on fossil fuels
for electricity generation. But they’re mostly inapt paradigms regarding the transition for larger developed and
developing countries. If you come across an article on any of the PUNIs, you can probably just skip it.
Wind Solar HGB Prim Energy Popul Econ Solar Wind
Country Popul
share shale share Consumption density complexity irradiance potential
% of elec % of elec % of elec Petajoules Millions Per sq mile 0-100 scale Percentile kW / capita
Renew able share of electricity > 60% and w ind/solar share of electricity > 30%
Namibia 2% 34% 63% 80 3 8 29 100 285
Denmark 55% 5% 22% 678 6 352 83 3 56
Lithuania 39% 5% 21% 324 3 111 78 4 10
Portugal 30% 7% 23% 842 10 289 74 33 13
Uruguay 31% 3% 50% 234 3 50 62 40 80
Large prim ary energy consum ers
China 9% 5% 17% 157,034 1,412 385 87 24 2
US 10% 5% 7% 89,555 332 91 90 30 16
Germany 23% 11% 12% 12,055 83 605 98 6 2
France 8% 4% 12% 9,857 68 303 88 12 9
India 4% 5% 12% 39,529 1,408 1,109 69 50 0
Japan 1% 11% 12% 16,731 126 854 100 14 15
South Korea 1% 4% 3% 12,216 52 1,340 98 20 12
Indonesia 0% 0% 19% 9,858 274 371 52 36 1
Turkey 11% 5% 27% 6,675 84 280 70 35 1
Source: IEA, IRENA, World Bank, Harvard ECI, ESMAP, JPMAM, 2022

52
We do not analyze Tokelau, a Pacific island dependency of New Zealand with a population of 1,500; or the
Grand Duchy of Luxembourg (pop. 600k) where most stores close at 6 pm every day.
39
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
Net Zero Oil
14th Annual Energy Paper

On “Net Zero” Oil: worth trying if it can be done close to very ambitious pro-forma projections
Net Zero Oil is based on the following concept: produce a barrel of oil, and simultaneously sequester enough
CO2 from the atmosphere to substantially offset the ~500 kg of CO2 emissions from combusting the oil plus
whatever additional emissions result from the sequestration approach chosen. Sounds simple; it’s not.
At JP Morgan’s inaugural Sustainability Summit last May, one of the energy CEOs in attendance noted the lack
of any discussion or support for their efforts to develop “Net Zero Oil”. I was called out for being behind the
curve since I did not mention anything about it in my presentation, and since I repeated my long-standing view
that the highest ratio in the history of science is the number of scientific papers written on carbon sequestration
divided by the actual amount of carbon sequestration. I was seen as a Net Zero Oil opponent even though I had
never heard much about it, which was my fault.
After the Summit, I agreed to spend time with the company’s engineers and write a briefing memo on Net Zero
Oil for our CEO Jamie Dimon and other members of the JP Morgan Operating Committee. The next three pages
are the verbatim contents of that memo. Bottom line: there are a lot of technical hurdles to overcome and the
cost of decarbonized oil might be extremely high; but I will keep an open mind regarding the potential for direct
air carbon capture and other related costs to come down over time for hard to abate sectors.
CO2 emissions from a barrel of oil
kg CO2e
520 8 kg
480 49 kg Transport
440 Production 59 kg
400 Refining
360
320
280
240
200 385 kg
Combustion
160
120
80
40
0
Source: IHS Markit, 2019

40
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
Net Zero Oil
14th Annual Energy Paper

Energy brief: on Net Zero Oil and Direct Air Carbon Capture
Net Zero Oil is a concept that entails capture and geologic sequestration of CO2 for enhanced oil recovery, with
sufficient sequestration to offset the lifecycle emissions of the oil produced in the process and emissions
resulting from sequestration. One example relies on Direct Air Carbon Capture (DACC) as a source for the CO2,
using zero emission energy to power the DACC process. This brief discusses the carbon math involved.
The Occidental approach: enhanced oil recovery using mostly zero emission sources of energy for DACC
Oxy is building its first DACC plant so projections and estimates should be judged accordingly53. Oxy intends to
use mostly zero carbon sources of energy for DACC, sourced from a combination of (a) dedicated solar power,
(b) zero carbon power from the grid via derivative contracts and (c) power/heat from natural gas whose
emissions are captured/sequestered. An example of the latter is projected to be sourced from NET Power
(NPWR) which aims to combust natural gas with pure oxygen to generate power and captured CO2 at a lower
cost than traditional gas plants with co-located carbon capture. NET Power is also a new venture, aiming to
deliver its first commercialized plants in 2026-2027. As a result, our energy math estimates are pro-forma and
subject to proof of concept/cost discovery as Oxy DACC and NET Power plants are constructed.
The carbon math of net zero oil/DACC using grid electricity
Most CO2-EOR projects in the US currently rely on naturally occurring underground deposits of CO2 and some
point source capture from anthropogenic sources. As a result, clearly assigning a GHG reduction benefit to oil
produced in these CO2-EOR systems is a complex exercise that involves assuming counterfactual GHG emissions
baselines and conventional oil production displacement rates. However, if exclusively atmospheric CO2 were
utilized via DACC for CO2-EOR, ascribing the GHG emissions benefits to the oil produced becomes more
straightforward. The emissions impact per barrel of CO2-EOR oil using atmospheric CO2 requires estimates of:
Assumption Value Assumption type Source
CO2 injected to produce a Limited observation
0.46 metric tons Azzolina54
barrel of oil set
Lifetime CO2 emissions from a Robust observation
0.5 metric tons of CO2 Multiple sources55
barrel of oil set
Energy intensity of DACC, 366 kWh of electricity and 5.25 GJ Theoretical; first
Keith56
including compression of heat per metric ton of CO2 plants being built
Carbon intensity of natural Robust observation
0.44 metric tons of CO2 per MWh EIA57
gas-powered electricity set
Carbon intensity of natural gas Robust observation
50.3 kg of CO2 per GJ EIA58
combustion set
Split of zero emissions power zero carbon sources are 40% in the Robust observation
EIA59
and natural gas median US state set
Upstream methane emissions
Robust observation
from natural gas production 13 g of CO2 per MJ of natural gas Littlefield60
set
and distribution

53
According to conversations with Oxy engineers in 2023, its DACC materials inputs are typically cheaper and less
volatile (calcium carbonate, caustic potash and PVC) when compared to greater reliance on steel, concrete and other
materials by facilities deploying solid sorbent DACC technology
54 “CO storage associated with CO enhanced oil recovery”, Azzolina et al, Int’l Journal of GHG Control, June 2015
2 2
55 Carnegie Endowment, Energy Futures Initiative and Clean Air Task Force

56 “A Process for Capturing CO from the Atmosphere”, David Keith (Harvard) et al., Joule, June 2018
2
57 “U.S. electricity net generation and resulting CO emissions by fuel in 2021”, EIA
2
58 “Natural gas and the environment”, EIA

59
“Net generation by state and source”, EIA
60
“Life Cycle GHG Perspective on US Natural Gas Delivery Pathways”, Env. Science & Technology, Littlefield et al, Nov 2022
41
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
Net Zero Oil
14th Annual Energy Paper

Using the assumptions above, we estimate that Oxy’s net CO2 emissions per DACC-EOR barrel would be 0.22
metric tons. Compared to lifecycle emissions of crude oil of 0.5 metric tons, this represents “55%
decarbonization”. As part of this calculation, only 79% of each metric ton of CO2 captured counts towards
emissions reductions since some CO2 must be captured to offset upstream and thermal grid emissions. The
assumptions in the table are for the Nth plant and not the first one, so this is aspirational at this point.
How might DACC-EOR oil achieve higher rates of decarbonization than the 55% baseline case?
1. Using only zero carbon power would increase EOR oil decarbonization to 67%. The impact is not large since
electricity only accounts for 20% of the energy in the typical DACC process (the rest is gas combustion)
2. If remote sensors reduced upstream methane emissions by half, EOR oil decarbonization would rise to 70%
3. If 20% more CO2 were injected per barrel of oil, EOR oil decarbonization would rise to 84%
4. Finally, if the energy load of DACC fell by 1/3 (possibly resulting from the use of industrial heat pumps for
thermal heat), EOR oil decarbonization would rise to 92%. In other words, “92% decarbonized oil”.
DACC-EOR oil decarbonization scenarios
Net emissions,
EOR oil
Scenario metric tons of
decarbonization %
CO2 per barrel
Traditional oil production, refining and combustion 0.50 NA
Oxy process with grid power 0.22 55%
Oxy process using 100% zero-carbon power 0.17 67%
Oxy process, 100% zero-carbon power, 50% reduction in upstream methane
0.15 70%
emissions
Oxy process, 100% zero carbon power, 50% reduction in upstream methane
0.08 84%
emissions, 20% more CO2 injected per barrel
Oxy process, 100% zero carbon power, 50% reduction in upstream methane
emissions, 20% more CO2 injected per barrel, 1/3 decline in energy intensity 0.04 92%
per ton of CO2 for DACC
Source: EIA, Harvard Univ., Carnegie Endow ment, Energy Futures Initiative, U. North Dakota, USGS, Occidental, JPMAM, June 2023

Challenges. The thermodynamics of DACC are very challenging, particularly when compared to point source
CCS. As a reminder, CO2 is only 0.04% of the atmosphere compared to 10%-15% concentrations in power plant
flue gas and 80%+ in some industrial flue gases. The DACC energy assumptions used in the analysis above are
based on an academic paper from 2018; Oxy is now trying to substantiate or improve upon those estimates.
If we take Oxy assumptions at face value, DACC would Energy required for DACC vs CCS
use 2.4x more energy than CCS. But is this realistic? A GJ of energy per metric ton of CO2
7
Shell Quest steam methane reformation project with CCS Thermal energy use
in Edmonton captures and buries ~1 mm metric tons of 6
Electricity use
CO2 per year. The source of CO2: concentrated syngas 5
from hydrogen production under 10,000x more partial 4
pressure than atmospheric CO2; ~78% of the syngas CO2 3
is captured61. According to Shell62, its CCS project
2
requires 2.8 GJ of energy per ton of CO2, including energy
for compression. Could Oxy’s DACC process really be 1
accomplished at only a 2.4x energy premium to Shell 0
Oxy DACC Shell Quest CCS
Quest? That needs to be proven given thermodynamic PROPOSED ACTUAL
challenges involved. Source: Quest CCS Project, Keith (Harvard) et al., 2022.

61
To be clear, Shell Quest is only capturing CO2 from syngas and not from the smokestack. After taking energy
demands of CCS into account, Shell Quest is only capturing 35% of total project CO2 emissions
62
Quest GHG and Energy Report, 2021
42
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
Net Zero Oil
14th Annual Energy Paper

One last geologic issue: CO2 resurfacing. Any CO2 that unexpectedly resurfaces would need to be recaptured,
recompressed and reinjected, all of which requires additional energy; or it would need to be deducted from the
net carbon benefits of DACC-EOR oil. Large scale sequestration projects are typically tracked regarding actual
geologic CO2 retention rates, particularly when the project qualifies for 45Q tax credits.
Implications for DACC as an emissions mitigation approach for hard-to-abate sectors
Learning curves have been steep across a wide range of energy technologies, generating lower unit costs and
increased efficiency. Wind, solar, EV battery and utility scale storage costs are notable examples. Other
examples include improved efficiency of natural gas combined cycle plants, residential and industrial heat
pumps and internal combustion engines; and the increase in wind capacity factors as a function of turbine/rotor
dimensions and offshore locations. But other improvements have been harder to come by, such as the low
round trip efficiency of electrolysis/hydrogen fuel cells used in transport (still just ~26%63) and the negligible
improvement in crystalline silicon PV cell efficiency (from 24% in the year 2000 to 26% in 2023).
Will there be a steep DACC learning curve64? If there were, DACC could be used to offset emissions from hard-
to-abate sectors such as aviation, primary steel production, cement, chemicals and agriculture. But if there
isn’t, it will be a very expensive form of abatement. As shown below, the latest estimates for first-of-a-kind
DACC plants are over $800 per ton of CO2. Even if these costs fell by two thirds, they would probably be the
most expensive solution you would ever see on a carbon abatement cost curve.
DACC-EOR conclusions:
• Even if DACC and Net Power plants are completed as planned, “Net Zero Oil” would probably refer to oil
that is significantly but not entirely decarbonized
• Net Zero Oil is projected to only be 100% decarbonized if CO2 injection rates increase, DACC costs decline
sharply, 100% zero carbon energy is used and upstream methane emissions fall
• DACC is currently very energy intensive; if/when 5-8 DACC plants are built, we will have a better sense for
the learning curve decline in capital costs and in energy consumption per ton of CO2

DACC first of a kind plant cost estimates

US$ per short ton of CO2
$1,800
Other
$1,600
Energy
$1,400 Fixed unit opex
$1,200 Unit capex
$1,000
$800
$600
$400
$200
$0
High temperature Low temperature Electrochemical
Regeneration type
Source: BCG, World Economic Forum, June 2023

63
Center for Sustainable Road Freight (UK); this figure includes AC/DC conversion, electrolysis, compression,
transfer and fuel cell losses
64
The Oxy/Carbon Engineering DACC assumption of ~1,800 kWh per tonne of CO2 (including process heat) is
lower than other estimates we have seen from the IEA, the World Resources Institute and Climeworks. An even
greater reduction in energy intensity of ~1/3 from Oxy levels was estimated by Feron in a 2022 paper presented
at the International Conference on Greenhouse Gas Technologies. The approach involves the use of amino
acids, cooling towers and industrial heat pumps, and has no direct thermal energy requirement.
43
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Bio-oil

A proper burial: can sequestration of bio-oil gain more traction than CCS?
In 2023, JP Morgan announced a $200 mm carbon removal plan designed to offset its Scope 1 emissions which
are expected to be ~800,000 metric tons over the next decade. Most of JP Morgan’s carbon removal plan relies
on existing point source CCS technology, but there’s another sequestration approach the firm is using as well.
JP Morgan entered into a carbon removal contract with Charm Industrial, a company which converts corn stover
(stalks, leaves and cobs) and wastewood from wildfire prevention treatments into something called “bio-oil”.
Bio-oil is produced via pyrolysis, which refers to the decomposition of biomass into solid char, liquid bio-oil and
gaseous syngas. The process occurs at very high temperatures in the absence of any oxygen in order to avoid
combustion. The carbon-rich liquid bio-oil is then sequestered underground, counting as carbon removal since
the source of the carbon is “biogenic”; in other words, it was recently in the atmosphere before it was absorbed
as CO2 via photosynthesis by corn plants. Burying bio-oil appears to be the best option since its energy density
is too low for use as a transportation or heating fuel.
As shown below using common switchgrass as an example, product yields from pyrolysis vary with the
temperature at which it occurs. According to Charm Industrial, its pyrolysis process yields roughly 50% bio-oil
and 25% each in bio-char and syngas.
Pyrolysis yields from switchgrass
Product yield
50%
Bio char
Charm Industrial process
40%
Bio-oil: sequestered and buried. For every cubic meter
of bio-oil produced, the associated mitigation is roughly
30% Bio oil 1.8 tonnes of CO2. Emissions associated with transport,
pyrolysis and injection reduce that figure by ~20%
20%
Syngas: used to generate energy for pyrolysis process
Syngas
10% Bio-char: buried with bio-oil or applied to soil to
improve nutrient retention, enhance soil structure
0%
375⁰ 400⁰ 425⁰ 450⁰ 475⁰ 500⁰ 525⁰ 550⁰ 575⁰ 600⁰ 625⁰
Pyrolysis temperature (C⁰)
Source: Journal of Analytical and Applied Pyrolysis, 2011

As a devout skeptic of point source CCS and direct air carbon capture, could bio-oil be any different?
• There’s no need for bio-oil pipelines since it is transported by truck and contains 2.3x more carbon per unit
of volume than compressed CO265
• Bio-oil can be buried in common wells that had previously been used for industrial waste or for oil
extraction, or old salt caverns left behind by oil and gas exploration. In other words, bio-oil does not require
more specialized geological formations typically required by CCS/DACC
That said, pyrolysis, pumping, blending and trucking all have carbon footprints and require proper accounting.
In a recent batch of bio-oil produced and sequestered for JP Morgan, Charm cited a net carbon removal rate of
~80%66. In addition, the cost of sequestration via bio-oil is also extremely high now, at $500-$600 per tonne of
CO2. And as of December 2023, Charm owned just three pyrolyzers with plans to build more in 2024. Last point:
there are also risks that after coming into contact with hot porous rocks, bio-oil becomes more viscous and
polymerizes, constraining the ability of a cavern to absorb more injected fluids.

65
Assuming bio-oil density of 1.2 grams per mL and 42% carbon share of bio-oil by weight; density of compressed
CO2 = 0.8 metric tons per cubic meter; and carbon’s weight of 27% in CO2
66
This net carbon removal rate was inspected and confirmed by Isometric, which has launched a bio-oil
sequestration protocol for measurement, reporting and verification
44
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Bio-oil

The biggest issue of course is scale. Charm has reportedly sequestered 7,000 tonnes of CO2 to date and aims
to get to ~20,000 tonnes per year in the next few years. For context: one of Charm’s clients is Meta, a relatively
carbon-light company with a footprint of just 66,000 tonnes of scope 1 CO2 emissions in 2021. Within the
Communication Services and Information Technology sectors, Meta is well below the cluster of higher emissions
companies. Note how around half of this cohort have practically no scope 1 emissions at all. Some tech
companies pay very high prices to mitigate their scope 1 footprints, which is easy to do when emissions are low.
Robert Hoglund at Carbon Gap has done some interesting research showing that GHG emissions are almost
perfectly inversely correlated with profits per tonne of GHG. In other words, the big emitters are probably way
less price inelastic regarding prices paid per ton of mitigation than most tech companies.
Meta’s scope 1 emissions only include fuel consumption in its buildings, planes and vehicles67. Its total carbon
footprint is 8.5 million tonnes of CO2, and Meta is a drop in the bucket compared to emissions of the S&P 500
and the entire US economy. We will keep watching this space, but I am not sure how bio-oil is any more scalable
than CCS, and it might even be less scalable. Charm’s business model may be profitable for its backers; but that
is a separate issue from whether bio-oil can meaningfully contribute to big picture decarbonization efforts.
Multiple of current
Metric tonnes Charm Industrial
of CO2e sequestration
Charm sequestration to date 7,083
Charm scale up target (annual) 18,667 3x
Meta scope 1 emissions 66,934 9x
Meta scope 1, 2, and 3 emissions 8,533,471 1,205x
Scope 1 emissions of S&P 500 1,476,977,428 208,524x
US CO2 emissions from energy 5,586,000,000 788,649x
Source: Bloomberg, EPA, individual company filings, JPMAM, 2023

Scope 1 Emissions of S&P 500 companies Scope 1 Emissions of S&P 500 Info Tech & Comm Services
Million metric tonnes of CO2e Million metric tonnes of CO2e
100 4.0
90 3.5
80 Top 12: Micron, Intel, Texas
3.0 Instruments, AT&T, Disney, ON Semi,
70 Corning, NXP Semi, Comcast, Lam
60 2.5 Research, Microchip Tech, Charter
50 2.0
40 1.5
30
1.0
20 META META
10 0.5

0 0.0
1 51 101 151 201 251 301 351 401 451 1 10 19 28 37 46 55 64 73 82
Source: Bloomberg, JPMAM, January 3, 2024 Source: Bloomberg, JPMAM, January 3, 2024

67
Scope 1 emissions occur from sources controlled or owned by an organization, such as emissions associated
with fuel combustion in their boilers, furnaces and vehicles
Scope 2 emissions mostly refer to electricity consumption for HVAC and data centers and reflect the CO 2
intensity of grids that electricity is sourced from. JP Morgan uses direct renewable power and renewable energy
credits to offset part of its Scope 2 emissions
Scope 3 emissions refer to activities such as employee travel/commuting. JP Morgan uses “nature-based”
credits to offset part of its Scope 3 emissions. Most nature-based credits are derived from forestation, grassland
and afforestation projects. The latter refers to efforts to develop forests where none have existed in recent
times; this can be challenging since if a given biome were conducive to forest growth, it would typically already
exist there. Leakage, compliance and verification issues have been challenges for users of nature-based credits
45
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Gaza and rooftop
solar power
14th Annual Energy Paper

What if? A thought experiment on the reconstruction of Gaza and the role of distributed solar power
During WWII, Germany drained conquered European territories of resources and labor to feed its war machine,
with industry and agriculture forcibly reoriented at the expense of native populations. The tide turned with the
Battle of Stalingrad in February 1943 in which one million Russian troops died. By the end of the war, Germany
was in shambles: 40% of its dwelling units were destroyed or damaged, food production and caloric intake fell
in half, industrial production fell by 30% and one third of all males born from 1916-1924 died during the war68.
In Italy: bridges, industrial enterprises and hundreds of thousands of homes were destroyed first by Allied
bombers and then by the retreating Germany Army. In Japan: 2.5 million people died during the war, major
parts of Tokyo and other cities were in ashes and one third of the nation’s wealth was destroyed.
When the US State Department designed the Marshall Plan to revive Germany and the rest of Europe, it was
highly unpopular with more than 50% of Americans opposing it. Eventually the Plan passed and boosted
Europe’s recovery. It is generally regarded as a 20th century policy success: when the war ended, policymakers
focused on what would come next and tried to influence the course of future events. This was in stark contrast
to the aftershocks in Germany following the reparations required by the Treaty of Versailles after WWI.
Post WWII Axis power recoveries Post WWI Treaty of Versailles and its aftermath in Germany
GDP per capita, 1990 US$ Index (100=1950) Percent
$8,000 800 14%
Germany Treaty of
Marshall & 700 Versailles 12%
$7,000
Italy GARIOA
Plans 600
$6,000 Japan German 10%
500 unemployment
$5,000 8%
400
$4,000 6%
300
$3,000 4%
200
$2,000 German inflation 2%
100
$1,000 0 0%
1930 1935 1940 1945 1950 1955 1960 1905 1910 1915 1920
Source: Angus Maddison, World Economy Historical Statistics, 2024 Source: Angus Maddison, World Economy Historical Statistics, 2024

If Americans were ambivalent about helping Germany after the war, you can imagine how they felt about Japan
given its attack on US soil in December 1941. Even so, while there was no explicit Marshall Plan for Japan, the
US still provided Japan roughly half the economic aid that it had provided to Germany. The US also helped
engineer Japan’s recovery via an advantageous exchange rate set at 360 Yen/$. With the benefit of a cheap
exchange rate, the era of Japanese trade surpluses began and lasted well into the 21st century.
US assistance to Germany and Japan, 1945-1952 The cost of select government programs
2005 US$, billions Percent of GDP
$30 1.2%
$29.3 Total assistance
$25 Economic reconstruction 1.0%

$20 0.8%

$15 0.6%
$15.2
0.4%
$10
$9.3
0.2%
$5
$5.2
0.0%
$0 Marshall Plan Manhattan Project Project Apollo
Germany Japan (atomic bomb) (moon landing)
Source: Congressional Research Service, March 2006 Source: Vaclav Smil (U Manitoba), Eichengreen (Berkeley), JPMAM, 2024

68
“Unbalanced sex ratios in Germany caused by World War II and their effect on fertility”, European Economic
Review, Kesternich et al, September 2020
46
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Gaza and rooftop
solar power
14th Annual Energy Paper

I am reminded about this history given the situation in Gaza. According to satellite analysis, 55%-65% of all
buildings in Gaza had likely been damaged or destroyed by February 2, 2024 by Israel in response to the October
7 attacks and hostage-taking by Hamas. The accompanying map illustrates spatial findings from researchers
Jamon Van Den Hoek and Corey Scher with white boxes indicating damaged and destroyed areas. Separately,
the Hebrew University of Jerusalem found that Israel has destroyed 40% of buildings within the one-kilometer
buffer zone it plans to maintain along the Israel-Gaza border69.
Building damage in Gaza
Percent of buildings likely damaged or destroyed Other damage assessments
90% • 90% of schools
80%
• All 7 universities
70%
• 2/3 of hospitals
60%
• 22% of agricultural land, including
50%
greenhouses and olive groves
40%
• More than half of all bakeries
30%
20%
• Half of all water/sanitation facilities
10%
High estimate: 50%+ building damage • Remaining wastewater treatment plants
Low estimate: 99%+ building damage
0% and sewage pumping stations are non-
Gaza North Khan Deir Al-
Rafah Total operational due to lack of electricity
City Gaza Younis Balah
Source: Damage analysis of Sentinel-1 satellite data, Van Den BBC, Foreign Policy, CSIS, France24, UNOCHA
Hoek (Oregon State) and Scher (CUNY), February 2, 2024

What if the West financed reconstruction of the electricity grid in the Gaza strip once buildings are rebuilt?
And what if this effort were centered around rooftop solar to increase Gaza’s energy independence? This is not
a futuristic idea: in 2022 China added 61 GW of distributed solar capacity, even more than the 45 GW of utility-
scale solar it added70. There are a million caveats involved with this scenario but before you get too caught up
in them, remember what Germany, Italy and Japan looked like in 1946. The lesson from the prior page:
coordinated international investment can go a long way after wartime destruction.
Furthermore, Gaza had been successful in adding over 8,000 rooftop solar sites to its electricity mix by the
end of the prior decade. Of the solar PV shown below, over 90% was rooftop solar while the remainder was
ground-based. This rooftop solar growth took place despite unfavorable financing conditions, ongoing conflict
with Israel, intraparty conflicts within Gaza, Israeli maritime blockades of Gaza, sub 1% growth rates, 45%
unemployment and Israeli-imposed restrictions on agriculture and industry. Before going further I will point out
that the source for this information is a paper from three professors at the Hebrew University of Jerusalem71.
Gaza average daily electricity mix, 2007-2019
Megawatts
225
200 Total electricity As of 2021, the Gaza Strip depended primarily on Israel
175
for energy. Israel supplied almost 100% of the fossil fuels
powering Gaza’s sole electricity power plant, which
150
Interconnected electricity grid with Israel provided between 60 and 80 MW of electricity. Israel also
125
supplied 120 MW via its interconnected electricity grid.
100
Since Gaza energy demand is estimated at 450 MW, it
75 suffered from a shortage of 250–270 MW. This energy
Gaza Power Plant (fuel from Israel)
50 shortage severely impacted essential services (health,
25 Egypt water, sanitation) in Gaza and undermined its economy,
Solar PV mainly manufacturing and agriculture (Fischhendler et al,
0
'07 '08 '09 '10 '11 '12 '13 '14 '15 '16 '17 '18 '19 Hebrew University of Jerusalem, 2022)
Source: "Light at the End of the Panel", New Political Economy, 2022

69
France24 (February 2, 2024) citing Adi Ben Nun of the Hebrew University of Jerusalem
70
“China takes its climate fight to the rooftops”, Bloomberg News, March 23, 2023
71
“Light at the End of the Panel: The Gaza Strip and the Interplay Between Geopolitical Conflict and Renewable Energy”,
New Political Economy, 2022, Fischhendler, Herman & David, Hebrew University of Jerusalem
47
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Gaza and rooftop
14th Annual Energy Paper solar power

While Gaza’s rooftop solar began in refugee camps and dense urban areas, eventually more efficient larger sites
were developed. Support for solar development in Gaza came from the World Bank, UNDP, OPEC and the EU.
CSIS estimates that Gaza had developed the densest rooftop solar system in the world72; at least a third of Gaza’s
population and more than 50% of its businesses used solar panels in March 2023. Gaza’s success in building
rooftop solar was part of what is a global trend: for the last several years, global rooftop solar PV additions have
roughly matched new utility-scale solar installations.
The potential for rooftop solar in a rebuilt Gaza
The chart below (right) shows electricity consumption in Gaza alongside an estimate of what a reconstructed
grid could look like. Before the war, 18 of Gaza’s 365 sq km was made up of high-density structures73. After a
lengthy period of reconstruction74, if we assume that 25% of the associated roofspace is converted to rooftop
solar at ~90 watts per square meter75 with a capacity factor of 20%76 at a cost of $1 per watt, solar power could
provide ~350 kWh per capita per year at an upfront capital cost of ~$500 million. In other words, a modest
amount of international aid in a rebuilt Gaza could substantially improve energy supplies and energy security.
Global solar PV capacity additions Electricity consumption per capita in Gaza
GW kWh per capita per year
350 400 1.1 MW/sq km (25% of high density Gaza structures
Commercial, industrial and residential converted to rooftop solar @20%CF)
300 350
Utility scale
250 300

200 250

150 200
Average annual electricity
100 150 consumption (kWh) per capita

50 100

0 50
2016 2017 2018 2019 2020 2021 2022 2023 2024 2003 2005 2007 2009 2011 2013 2015 2017 2019 2021
Source: IEA, 2023 Source: CSIS, Environment America, JTC, Our World in Data, JPMAM, 2023

Any efforts to galvanize international aid for Gaza might be made more difficult by the history of Hamas
diverting international aid77, and its founding charter. Since 2007 when Hamas took power in the Gaza strip,
it has financed its activities via levies/taxes on trade with Egypt, new import taxes, business taxes and taxes on
income earned by Palestinians working in Israel. While aid from Qatar resulted in reconstruction of homes and
infrastructure, black markets for construction materials also ended up reportedly being used by Hamas for
military purposes. From 1994 to 2020, Europe and the US accounted for 75% of the $40 billion in international
aid provided to Gaza and the West Bank78, but this figure doesn’t include Iran. US Congressional estimates of
Iranian financial support for Hamas range from $100 mm to $350 mm per year. As for the Hamas founding
charter and its core principles, you can read about it here and judge for yourself79.

72
“Gaza’s solar power in wartime”, CSIS, Will Todman et al, November 21, 2023
73
“Land/cover use in Gaza in 2023”, Dr. He Yin, Kent State University Department of Geography
74
UNCTAD projects that it could take many decades for Gaza’s economy to recover to prewar levels
75
“Buildings & Parking Lots: Ready for a Recharge”, Morgan Stanley Research, May 6, 2022
76
“National Survey Report of Photovoltaic Applications in Israel”, International Energy Agency, 2017
77
“How the West Inadvertently Funded Hamas”, Wall Street Journal, October 19, 2023
78
“International Aid to the Palestinians”, August 4, 2022, Omar Shaban, Arab Center DC
79
“Understanding Hamas’s Genocidal Ideology: A close read of Hamas’s founding documents clearly shows its
intentions”, Bruce Hoffman (Georgetown, Council on Foreign Relations), The Atlantic, October 2023. Quick
summary: destruction of Israel, creation of an Islamic State based on Sharia Law and rejection of all political
settlements and negotiations
48
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4 Gaza and rooftop
14th Annual Energy Paper solar power

Whatever international aid levels were, they did little to prevent the crippling economic situation in Gaza that
existed before the latest war and which is illustrated below. Note how the income gap between Israel and Gaza
is the third largest in the world of all contiguous countries, behind only Yemen/Saudi Arabia and North/South
Korea. The myriad reasons for this are well beyond the scope of this analysis.
At the end of last year’s paper, I wrote about the impossible odds facing nuclear fusion, space-based solar power
and electric planes. The odds of a Western-funded Gaza solar rebuild also seem impossible right now. One
thing’s for sure: like post-WWI Germany in 1919, the story is almost certainly not going to end where things
stand now. It could either get much better, or much worse.
GDP per capita Unemployment rate
2015 US$ Percent, smoothed 4 quarters
$1,400 50%
West Bank
$1,200
40% Gaza
$1,000
30%
$800

$600 20%
$400
10% West Bank
$200 Gaza
$0 0%
2000 2003 2006 2009 2012 2015 2018 2021 2024 2002 2006 2010 2014 2018 2022
Source: Palestinian Central Bureau of Statistics, JPMAM, Q3 2023 Source: Palestinian Central Bureau of Statistics, JPMAM, Q3 2023

Lowest per capita income and income growth, 1994-2022 Largest per capita GDP gaps across contiguous countries
2022 per capita income, PPP US$ Tajikistan-China
$10,000 Mozambique-South Africa
$9,000 Namibia Jordan-Israel
$8,000 Egypt-Israel
Guatemala Angola Afghanistan-Iran
$7,000
Venezuela Lebanon-Israel
$6,000 Côte d'Ivoire
Syria-Israel
$5,000 Syria Kenya
North Korea-Russia
$4,000 Sudan Cameroon Afghanistan-Turkmenistan
$3,000 Zimbabwe Afghanistan-China
Yemen Senegal
$2,000 Yemen-Oman Country A per capita GDP
Madagascar NigerChad
Gaza Gaza-Israel divided by
$1,000 DR Congo
Yemen-Saudi Arabia Country B per capita GDP
$0
-75% -50% -25% 0% 25% 50% North Korea-South Korea
Change in real per capita income, 1994-2022 0% 2% 4% 6% 8% 10%
Source: Conference Board, IMF, JPMAM, 2024 Source: IMF, World Bank, JPMAM, 2023

Religious demography of Israel-Palestine region since the Year 1 AD

Percent of population
100%
90%
Muslims
80% Christians
70%
60%
50%
40%
Pagan/Druze/
30% Other/Unknown Jews
20%
10%
0%
0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000
Source: Lyman Stone, Demographic Intelligence/McGill University, 2023

49
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Appendix I

Appendix I: on US and European energy supply/demand and global LNG markets

Despite an almost total decline in Suez Canal containership crossings, WTI oil prices in February 2024 were
actually below where they were when the Israel-Hamas conflict began. The US is less exposed than it was in the
1970’s to changing oil prices. This can partially be explained by the following:
• The US is a net energy exporter vs its net import position in the 1970’s. US net crude imports are down 75%
from the 2005 peak, and net refined product imports of 4 mm bpd in 2005 flipped to net exports of 4 mm
bpd of refined products by 2019. As shown on page 21, this is mostly a by-product of the shale revolution
• The oil intensity of US GDP growth is 65% lower than it was in the 1970’s
• Annual global oil consumption growth has declined from 8%-10% in the early 1970’s to 0%-2% today
(although as you can see in the 4th chart, there’s no peak in oil consumption yet)
• Geopolitical benefits to OPEC of an oil embargo would be less clear now: 75% of Saudi oil exports go to Asia,
China gets half its oil from the Middle East and the US gets most of its imported oil from Canada, Mexico
and other non-OPEC sources
• Saudi Arabia also has spare capacity to bring online if needed. Spare capacity is measured in different ways;
as shown below it is at a high level for a non-recessionary period
US energy dependence and independence Oil intensity and consumption
Net imports of oil, natural gas and coal as a share of total primary MJ per US$ of GDP (chained 2017$) Percent
energy consumption 6 10%
35% US oil intensity
Net importer 5 8%
30%
25%
4 6%
20%
15% 3 4%
10%
5% 2 2%
0% Global 5-yr rolling oil
1 0%
-5% demand growth
Net exporter
-10% 0 -2%
1970 1980 1990 2000 2010 2020 1965 1975 1985 1995 2005 2015
Source: EI Statistical Review of World Energy, JPMAM, 2023 Source: Bloomberg, Energy Institute, JPMAM, 2023

Estimated OPEC spare capacity Global oil demand

Million barrels per day Million barrels per day
12 105
Bloomberg 100
10
J.P. Morgan 95
8 90
85
6
80
4 75
70
2
65
0 60
2000 2003 2006 2009 2012 2015 2018 2021 2024 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030
Source: Bloomberg, J.P. Morgan, January 2024 Source: IEA, JP Morgan Global Commodities Research, January 2024

50
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Appendix I

As for Europe, the region has survived based on a combination of fortuitously warm winter weather, greater
energy efficiencies, reduced energy consumption and increasing amounts of imported LNG:
• 15% decline in gas consumption in 2023 vs 2022
• By December 2023, imported LNG volumes tripled vs January 2021
• 46 mtpa of LNG regasification capacity added in 2023, a 25% increase vs January 2022 levels
The global LNG market may remain tight until a large increase in supply in 2025, depending in part on the
outcome of recently announced restrictions in the US on the export of LNG.
Update: in January 2024, the Biden administration froze approval for new liquefaction plants exporting US LNG
to countries without a free trade agreement with the US (note that there is no such agreement between the US
and EU). To be clear, current LNG exports and projects under construction will not be affected, and there is
an exemption for US allies in case of emergency. In other words, no impact over the short or medium term on
US LNG exports to Europe or Asia. The ban seems mostly political in nature in an election year given minimal
climate or national security implications. There are only four projects in the DoE approval queue that would be
affected (Sempra, Commonwealth LNG and Energy Transfer). A Venture Global project in Louisiana could also
be impacted; its FERC approval is pending and is required before consideration by the DoE.
European natural gas demand European imports of natural gas and LNG
Million cubic meters per day Percent, y/y change Million cubic meters per day
50 5% 1,000
900
-50 -5% 800
700
-150 -15%
600 Africa
500 Global LNG
-250 -25%
400
Electricity generation
-350 Industrial -35% 300
Residential and commercial 200
-450
YoY % change (rhs) -45% 100 Russia North Sea
Jul-22

Jul-23
Jun-22

Nov-23
Dec-22
Jan-23

Jun-23
Apr-22

Aug-22
Sep-22
Oct-22
Nov-22

Mar-23

Aug-23
Sep-23
Oct-23
May-22

Apr-23
Feb-23

May-23

0
Jan-21 Jul-21 Jan-22 Jul-22 Jan-23 Jul-23 Jan-24
Source: Mitsubishi UFJ Financial Group, December 2023 Source: J.P. Morgan Commodities Research, January 2024

Projected new European LNG regasification capacity Global LNG supply

Million tons per annum Million tonnes per annum
120 60
110 Pre-war Russian gas exports to the EU
100 50
90 Italy
Spain
Other
80
40
70 France
60 Finland/Estonia
30
50 Germany Middle East
40 Netherlands 20
30
20 US
10 10
Australia
0
Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 0
22 22 23 23 23 23 24 24 24 24 25 25 25 25 26 '13 '14 '15 '16 '17 '18 '19 '20 '21 '22 '23 '24 '25 '26 '27 '28
Source: Mitsubishi UFJ Financial Group, December 2023 Source: Mitsubishi UFJ Financial Group, December 2023

51
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper Appendix II

Appendix II: EV misadventures with Electrify America, Waymo and Jeep

Electrify America: yes, but only part of the time
MotorTrend published a travelogue by a new owner of a Ford F-150 Lightning EV. Called away on an emergency
trip, the family piled into their F-150 on a full charge and headed north, following the vehicle’s recommended
route along Interstate 5. The route was selected due to its Electrify America charging network.
• Unfortunately, 4 of the 6 Electrify America towers at the designated stop were offline with a large wait for
the two working towers
• Electrify America was unable to remotely fix the broken towers and Ford’s software is apparently unable to
adjust trips accordingly. As a result, the family went to another local charging station despite its lower
charge rate, but that charger wasn’t working either
• The family returned to the original Electrify America station and waited two hours to plug in their F-150,
charging at 33 kW which was 10% of the tower’s advertised capacity. After one hour on the charger, the
battery was 64% full so they scheduled another stop 100 miles north only to find another subpar tower
which only charged at 40 kW. After 25 minutes they gave up, booked a hotel and charged overnight
“Our last Ford F-150 Lightning EV pickup road trip was a nightmare”, MotorTrend, Christian Seabaugh,
December 22, 2023. This travelogue is consistent with data from UC Berkeley and JD Power: almost 30% of
chargers are not working in the SF Bay Area, and 20% of visits to charging stations nationwide end up with
drivers not charging80. Best in class: Tesla, with a 99%+ global charging station reliability rate.
Waymo: electric vehicles can gaslight as well
In Phoenix last December, two Waymo self-driving vehicles crashed into the same pickup truck being towed by
a tow truck. According to Waymo, this was due to an “exceptionally rare” combination of events: the tow truck
was pulling the pickup truck in a backwards position and partially occupying another lane. These conditions were
apparently too complex and unusual for Waymo vehicles to decipher. A Waymo EV approached from behind
and crashed into the pickup truck. The tow truck kept on driving and several minutes later, a second Waymo
EV crashed into the same pickup truck, also after approaching from behind. The director of the Robotics and
Autonomy Center at George Mason University took exception to Waymo’s characterization of these events as
being that unusual, and is now “exhausted by the company’s gaslighting”81. NHTSA Recall: #24E-013
Jeep: internal (spontaneous) combustion engine
My affinity for Jeep Wranglers was sorely tested after two recalls of my 2021 Jeep Wrangler plug-in hybrid 4xe.
The first recall was for “loss of motor power”, while the second was for “thermostat gasket failure leading to
coolant leakage”. In addition, the heater for the electric battery failed and was on back-order for several weeks.
Now, another recall. According to the December 2023 recall notice, “The high voltage battery on your vehicle
may fail internally. The defect has not been identified and the root cause is still being investigated. An internally
failed battery could lead to a vehicle fire with the ignition on or off”. The recall continues: “Customers are
advised to refrain from recharging these vehicles, and not to park inside buildings or structures or near other
vehicles until the vehicle has the final repair completed.” And the kicker: “The remedy for this condition is not
yet available”. Jeep recall notice B9A/NHTSA 237-787

80
“Reliability of Open Public Electric Vehicle Direct Current Fast Chargers”, UC Berkeley, Rempel et al, April 2022,
and “JD Power 2023 U.S. Electric Vehicle Experience (EVX) Public Charging Study”, JD Power, August 2023
81
Professor M. Cummings on LinkedIn, George Mason University, February 20, 2024. Gaslighting: manipulating
someone into questioning their own perception of reality, from the 1944 film Gaslight
52
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper

IMPORTANT INFORMATION
KEY RISKS
This material is for information purposes only, and may inform you of certain products and services offered by private banking businesses, part of JPMorgan
Chase & Co. (“JPM”). Products and services described, as well as associated fees, charges and interest rates, are subject to change in accordance with the
applicable account agreements and may differ among geographic locations. Not all products and services are offered at all locations. If you are a person with a
disability and need additional support accessing this material, please contact your J.P. Morgan team or email us at accessibility.support@jpmorgan.com for
assistance. Please read all Important Information.

GENERAL RISKS & CONSIDERATIONS

Any views, strategies or products discussed in this material may not be appropriate for all individuals and are subject to risks. Investors may get back less than
they invested, and past performance is not a reliable indicator of future results. Asset allocation/diversification does not guarantee a profit or protect against
loss. Nothing in this material should be relied upon in isolation for the purpose of making an investment decision. You are urged to consider carefully whether
the services, products, asset classes (e.g. equities, fixed income, alternative investments, commodities, etc.) or strategies discussed are suitable to your needs.
You must also consider the objectives, risks, charges, and expenses associated with an investment service, product or strategy prior to making an investment
decision. For this and more complete information, including discussion of your goals/situation, contact your J.P. Morgan team.

NON-RELIANCE
Certain information contained in this material is believed to be reliable; however, JPM does not represent or warrant its accuracy, reliability or completeness, or
accept any liability for any loss or damage (whether direct or indirect) arising out of the use of all or any part of this material. No representation or warranty
should be made with regard to any computations, graphs, tables, diagrams or commentary in this material, which are provided for illustration/ reference
purposes only. The views, opinions, estimates and strategies expressed in this material constitute our judgment based on current market conditions and are
subject to change without notice. JPM assumes no duty to update any information in this material in the event that such information changes. Views, opinions,
estimates and strategies expressed herein may differ from those expressed by other areas of JPM, views expressed for other purposes or in other contexts, and
this material should not be regarded as a research report. Any projected results and risks are based solely on hypothetical examples cited, and actual results
and risks will vary depending on specific circumstances. Forward-looking statements should not be considered as guarantees or predictions of future events.

Nothing in this document shall be construed as giving rise to any duty of care owed to, or advisory relationship with, you or any third party. Nothing in this
document shall be regarded as an offer, solicitation, recommendation or advice (whether financial, accounting, legal, tax or other) given by J.P. Morgan and/or its
officers or employees, irrespective of whether or not such communication was given at your request. J.P. Morgan and its affiliates and employees do not provide
tax, legal or accounting advice. You should consult your own tax, legal and accounting advisors before engaging in any financial transactions.

For J.P. Morgan Asset Management Clients:

J.P. Morgan Asset Management is the brand for the asset management business of JPMorgan Chase & Co. and its affiliates worldwide.
To the extent permitted by applicable law, we may record telephone calls and monitor electronic communications to comply with our legal and regulatory
obligations and internal policies. Personal data will be collected, stored and processed by J.P. Morgan Asset Management in accordance with our privacy policies
at https://am.jpmorgan.com/global/privacy.

ACCESSIBILITY
For U.S. only: If you are a person with a disability and need additional support in viewing the material, please call us at 1-800-343-1113 for assistance.
This communication is issued by the following entities:
In the United States, by J.P. Morgan Investment Management Inc. or J.P. Morgan Alternative Asset Management, Inc., both regulated by the Securities and
Exchange Commission; in Latin America, for intended recipients’ use only, by local J.P. Morgan entities, as the case may be.; in Canada, for institutional clients’
use only, by JPMorgan Asset Management (Canada) Inc., which is a registered Portfolio Manager and Exempt Market Dealer in all Canadian provinces and
territories except the Yukon and is also registered as an Investment Fund Manager in British Columbia, Ontario, Quebec and Newfoundland and Labrador. In the
United Kingdom, by JPMorgan Asset Management (UK) Limited, which is authorized and regulated by the Financial Conduct Authority; in other European
jurisdictions, by JPMorgan Asset Management (Europe) S.à r.l. In Asia Pacific (“APAC”), by the following issuing entities and in the respective jurisdictions in which
they are primarily regulated: JPMorgan Asset Management (Asia Pacific) Limited, or JPMorgan Funds (Asia) Limited, or JPMorgan Asset Management Real Assets
(Asia) Limited, each of which is regulated by the Securities and Futures Commission of Hong Kong; JPMorgan Asset Management (Singapore) Limited (Co. Reg. No.
197601586K), which this advertisement or publication has not been reviewed by the Monetary Authority of Singapore; JPMorgan Asset Management (Taiwan)
Limited; JPMorgan Asset Management (Japan) Limited, which is a member of the Investment Trusts Association, Japan, the Japan Investment Advisers Association,
Type II Financial Instruments Firms Association and the Japan Securities Dealers Association and is regulated by the Financial Services Agency (registration number
“Kanto Local Finance Bureau (Financial Instruments Firm) No. 330”); in Australia, to wholesale clients only as defined in section 761A and 761G of the Corporations
Act 2001 (Commonwealth), by JPMorgan Asset Management (Australia) Limited (ABN 55143832080) (AFSL 376919). For all other markets in APAC, to intended
recipients only.

For J.P. Morgan Private Bank Clients:

YOUR INVESTMENTS AND POTENTIAL CONFLICTS OF INTEREST
Conflicts of interest will arise whenever JPMorgan Chase Bank, N.A. or any of its affiliates (together, “J.P. Morgan”) have an actual or perceived economic or other
incentive in its management of our clients’ portfolios to act in a way that benefits J.P. Morgan. Conflicts will result, for example (to the extent the following
activities are permitted in your account): (1) when J.P. Morgan invests in an investment product, such as a mutual fund, structured product, separately managed
account or hedge fund issued or managed by JPMorgan Chase Bank, N.A. or an affiliate, such as J.P. Morgan Investment Management Inc.; (2) when a J.P. Morgan
entity obtains services, including trade execution and trade clearing, from an affiliate; (3) when J.P. Morgan receives payment as a result of purchasing an
investment product for a client’s account; or (4) when J.P. Morgan receives payment for providing services (including shareholder servicing, recordkeeping or
custody) with respect to investment products purchased for a client’s portfolio. Other conflicts will result because of relationships that J.P. Morgan has with
other clients or when J.P. Morgan acts for its own account.

53
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper

Investment strategies are selected from both J.P. Morgan and third-party asset managers and are subject to a review process by our manager research teams.
From this pool of strategies, our portfolio construction teams select those strategies we believe fit our asset allocation goals and forward-looking views in order
to meet the portfolio's investment objective.
As a general matter, we prefer J.P. Morgan managed strategies. We expect the proportion of J.P. Morgan managed strategies will be high (in fact, up to 100
percent) in strategies such as, for example, cash and high-quality fixed income, subject to applicable law and any account-specific considerations.

While our internally managed strategies generally align well with our forward-looking views, and we are familiar with the investment processes as well as the risk
and compliance philosophy of the firm, it is important to note that J.P. Morgan receives more overall fees when internally managed strategies are included. We
offer the option of choosing to exclude J.P. Morgan managed strategies (other than cash and liquidity products) in certain portfolios.

The Six Circles Funds are U.S.-registered mutual funds managed by J.P. Morgan and sub-advised by third parties. Although considered internally managed
strategies, JPMC does not retain a fee for fund management or other fund services.

LEGAL ENTITY, BRAND & REGULATORY INFORMATION

In the United States, bank deposit accounts and related services, such as checking, savings and bank lending, are offered by JPMorgan Chase Bank, N.A.
Member FDIC.

JPMorgan Chase Bank, N.A. and its affiliates (collectively “JPMCB”) offer investment products, which may include bank managed investment accounts and
custody, as part of its trust and fiduciary services. Other investment products and services, such as brokerage and advisory accounts, are offered through J.P.
Morgan Securities LLC (“JPMS”), a member of FINRA and SIPC. Insurance products are made available through Chase Insurance Agency, Inc. (CIA), a licensed
insurance agency, doing business as Chase Insurance Agency Services, Inc. in Florida. JPMCB, JPMS and CIA are affiliated companies under the common control
of JPM. Products not available in all states.

In Germany, this material is issued by J.P. Morgan SE, with its registered office at Taunustor 1 (TaunusTurm), 60310 Frankfurt am Main, Germany, authorized by
the Bundesanstalt für Finanzdienstleistungsaufsicht (BaFin) and jointly supervised by the BaFin, the German Central Bank (Deutsche Bundesbank) and the
European Central Bank (ECB). In Luxembourg, this material is issued by J.P. Morgan SE – Luxembourg Branch, with registered office at European Bank and
Business Centre, 6 route de Treves, L-2633, Senningerberg, Luxembourg, authorized by the Bundesanstalt für Finanzdienstleistungsaufsicht (BaFin) and jointly
supervised by the BaFin, the German Central Bank (Deutsche Bundesbank) and the European Central Bank (ECB); J.P. Morgan SE – Luxembourg Branch is also
supervised by the Commission de Surveillance du Secteur Financier (CSSF); registered under R.C.S Luxembourg B255938. In the United Kingdom, this material
is issued by J.P. Morgan SE – London Branch, registered office at 25 Bank Street, Canary Wharf, London E14 5JP, authorized by the Bundesanstalt für
Finanzdienstleistungsaufsicht (BaFin) and jointly supervised by the BaFin, the German Central Bank (Deutsche Bundesbank) and the European Central Bank
(ECB); J.P. Morgan SE – London Branch is also supervised by the Financial Conduct Authority and Prudential Regulation Authority. In Spain, this material is
distributed by J.P. Morgan SE, Sucursal en España, with registered office at Paseo de la Castellana, 31, 28046 Madrid, Spain, authorized by the Bundesanstalt für
Finanzdienstleistungsaufsicht (BaFin) and jointly supervised by the BaFin, the German Central Bank (Deutsche Bundesbank) and the European Central Bank
(ECB); J.P. Morgan SE, Sucursal en España is also supervised by the Spanish Securities Market Commission (CNMV); registered with Bank of Spain as a branch of
J.P. Morgan SE under code 1567. In Italy, this material is distributed by J.P. Morgan SE – Milan Branch, with its registered office at Via Cordusio, n.3, Milan 20123,
Italy, authorized by the Bundesanstalt für Finanzdienstleistungsaufsicht (BaFin) and jointly supervised by the BaFin, the German Central Bank (Deutsche
Bundesbank) and the European Central Bank (ECB); J.P. Morgan SE – Milan Branch is also supervised by Bank of Italy and the Commissione Nazionale per le
Società e la Borsa (CONSOB); registered with Bank of Italy as a branch of J.P. Morgan SE under code 8076; Milan Chamber of Commerce Registered Number: REA
MI 2536325. In the Netherlands, this material is distributed by J.P. Morgan SE – Amsterdam Branch, with registered office at World Trade Centre, Tower B,
Strawinskylaan 1135, 1077 XX, Amsterdam, The Netherlands, authorized by the Bundesanstalt für Finanzdienstleistungsaufsicht (BaFin) and jointly supervised by
the BaFin, the German Central Bank (Deutsche Bundesbank) and the European Central Bank (ECB); J.P. Morgan SE – Amsterdam Branch is also supervised by De
Nederlandsche Bank (DNB) and the Autoriteit Financiële Markten (AFM) in the Netherlands. Registered with the Kamer van Koophandel as a branch of J.P.
Morgan SE under registration number 72610220. In Denmark, this material is distributed by J.P. Morgan SE – Copenhagen Branch, filial af J.P. Morgan SE,
Tyskland, with registered office at Kalvebod Brygge 39-41, 1560 København V, Denmark, authorized by the Bundesanstalt für Finanzdienstleistungsaufsicht
(BaFin) and jointly supervised by the BaFin, the German Central Bank (Deutsche Bundesbank) and the European Central Bank (ECB); J.P. Morgan SE –
Copenhagen Branch, filial af J.P. Morgan SE, Tyskland is also supervised by Finanstilsynet (Danish FSA) and is registered with Finanstilsynet as a branch of J.P.
Morgan SE under code 29010. In Sweden, this material is distributed by J.P. Morgan SE – Stockholm Bankfilial, with registered office at Hamngatan 15,
Stockholm, 11147, Sweden, authorized by the Bundesanstalt für Finanzdienstleistungsaufsicht (BaFin) and jointly supervised by the BaFin, the German Central
Bank (Deutsche Bundesbank) and the European Central Bank (ECB); J.P. Morgan SE – Stockholm Bankfilial is also supervised by Finansinspektionen (Swedish
FSA); registered with Finansinspektionen as a branch of J.P. Morgan SE. In Belgium, this material is distributed by J.P. Morgan SE – Brussels Branch with
registered office at 35 Boulevard du Régent, 1000, Brussels, Belgium, authorized by the Bundesanstalt für Finanzdienstleistungsaufsicht (BaFin) and jointly
supervised by the BaFin, the German Central Bank (Deutsche Bundesbank) and the European Central Bank (ECB); J.P. Morgan SE Brussels Branch is also
supervised by the National Bank of Belgium (NBB) and the Financial Services and Markets Authority (FSMA) in Belgium; registered with the NBB under
registration number 0715.622.844. In Greece, this material is distributed by J.P. Morgan SE – Athens Branch, with its registered office at 3 Haritos Street, Athens,
10675, Greece, authorized by the Bundesanstalt für Finanzdienstleistungsaufsicht (BaFin) and jointly supervised by the BaFin, the German Central Bank
(Deutsche Bundesbank) and the European Central Bank (ECB); J.P. Morgan SE – Athens Branch is also supervised by Bank of Greece; registered with Bank of
Greece as a branch of J.P. Morgan SE under code 124; Athens Chamber of Commerce Registered Number 158683760001; VAT Number 99676577. In France, this
material is distributed by J.P. Morgan SE – Paris Branch, with its registered office at 14, Place Vendôme 75001 Paris, France, authorized by the Bundesanstaltfür
Finanzdienstleistungsaufsicht(BaFin) and jointly supervised by the BaFin, the German Central Bank (Deutsche Bundesbank) and the European Central Bank (ECB)
under code 842 422 972; J.P. Morgan SE – Paris Branch is also supervised by the French banking authorities the Autorité de Contrôle Prudentiel et de Résolution
(ACPR) and the Autorité des Marchés Financiers (AMF). In Switzerland, this material is distributed by J.P. Morgan (Suisse) SA, with registered address at rue du
Rhône, 35, 1204, Geneva, Switzerland, which is authorised and supervised by the Swiss Financial Market Supervisory Authority (FINMA) as a bank and a
securities dealer in Switzerland.
This communication is an advertisement for the purposes of the Markets in Financial Instruments Directive (MIFID II) and the Swiss Financial Services Act
(FINSA). Investors should not subscribe for or purchase any financial instruments referred to in this advertisement except on the basis of information contained
in any applicable legal documentation, which is or shall be made available in the relevant jurisdictions (as required).
In Hong Kong, this material is distributed by JPMCB, Hong Kong branch. JPMCB, Hong Kong branch is regulated by the Hong Kong Monetary Authority and the
Securities and Futures Commission of Hong Kong. In Hong Kong, we will cease to use your personal data for our marketing purposes without charge if you so
request. In Singapore, this material is distributed by JPMCB, Singapore branch. JPMCB, Singapore branch is regulated by the Monetary Authority of Singapore.
Dealing and advisory services and discretionary investment management services are provided to you by JPMCB, Hong Kong/Singapore branch (as notified to

54
E Y E O N T H E M A R K E T • M I C H A E L C E M B A L E S T • J . P . M O R G A N • M AR C H 2 0 2 4
14th Annual Energy Paper

you). Banking and custody services are provided to you by JPMCB Singapore Branch. The contents of this document have not been reviewed by any regulatory
authority in Hong Kong, Singapore or any other jurisdictions. You are advised to exercise caution in relation to this document. If you are in any doubt about any
of the contents of this document, you should obtain independent professional advice. For materials which constitute product advertisement under the Securities
and Futures Act and the Financial Advisers Act, this advertisement has not been reviewed by the Monetary Authority of Singapore. JPMorgan Chase Bank, N.A., a
national banking association chartered under the laws of the United States, and as a body corporate, its shareholder’s liability is limited.

With respect to countries in Latin America, the distribution of this material may be restricted in certain jurisdictions. We may offer and/or sell to you securities
or other financial instruments which may not be registered under, and are not the subject of a public offering under, the securities or other financial regulatory
laws of your home country. Such securities or instruments are offered and/or sold to you on a private basis only. Any communication by us to you regarding such
securities or instruments, including without limitation the delivery of a prospectus, term sheet or other offering document, is not intended by us as an offer to
sell or a solicitation of an offer to buy any securities or instruments in any jurisdiction in which such an offer or a solicitation is unlawful. Furthermore, such
securities or instruments may be subject to certain regulatory and/or contractual restrictions on subsequent transfer by you, and you are solely responsible for
ascertaining and complying with such restrictions. To the extent this content makes reference to a fund, the Fund may not be publicly offered in any Latin
American country, without previous registration of such fund´s securities in compliance with the laws of the corresponding jurisdiction.

JPMorgan Chase Bank, N.A. (JPMCBNA) (ABN 43 074 112 011/AFS Licence No: 238367) is regulated by the Australian Securities and Investment Commission and
the Australian Prudential Regulation Authority. Material provided by JPMCBNA in Australia is to “wholesale clients” only. For the purposes of this paragraph the
term “wholesale client” has the meaning given in section 761G of the Corporations Act 2001 (Cth). Please inform us if you are not a Wholesale Client now or if
you cease to be a Wholesale Client at any time in the future.
JPMS is a registered foreign company (overseas) (ARBN 109293610) incorporated in Delaware, U.S.A. Under Australian financial services licensing requirements,
carrying on a financial services business in Australia requires a financial service provider, such as J.P. Morgan Securities LLC (JPMS), to hold an Australian Financial
Services Licence (AFSL), unless an exemption applies. JPMS is exempt from the requirement to hold an AFSL under the Corporations Act 2001 (Cth) (Act) in
respect of financial services it provides to you, and is regulated by the SEC, FINRA and CFTC under US laws, which differ from Australian laws. Material
provided by JPMS in Australia is to “wholesale clients” only. The information provided in this material is not intended to be, and must not be, distributed or
passed on, directly or indirectly, to any other class of persons in Australia. For the purposes of this paragraph the term “wholesale client” has the meaning given
in section 761G of the Act. Please inform us immediately if you are not a Wholesale Client now or if you cease to be a Wholesale Client at any time in the future.

This material has not been prepared specifically for Australian investors. It: may contain references to dollar amounts which are not Australian dollars; may
contain financial information which is not prepared in accordance with Australian law or practices; may not address risks associated with investment in foreign
currency denominated investments; and does not address Australian tax issues.

References to “J.P. Morgan” are to JPM, its subsidiaries and affiliates worldwide. “J.P. Morgan Private Bank” is the brand name for the private banking business
conducted by JPM. This material is intended for your personal use and should not be circulated to or used by any other person, or duplicated for non-personal
use, without our permission. If you have any questions or no longer wish to receive these communications, please contact your J.P. Morgan team.

© 2024 JPMorgan Chase & Co. All rights reserved.

55