The hard stuff: Navigating the physical realities of the energy transition

The research was led by Mekala Krishnan, an MGI partner in Boston; Chris Bradley, a McKinsey senior partner and a director of MGI in Sydney; Humayun Tai, a senior partner in the New York office; Tiago Devesa, an MGI senior fellow in Lisbon; Sven Smit, McKinsey senior partner in Amsterdam and chairman of MGI; and Daniel Pacthod, a senior partner in the New York office. We give particular thanks to Lola Woetzel (alumna), a former McKinsey senior partner and director of MGI, who helped us drive the research that led to this report.

A group of McKinsey colleagues coauthored chapters dedicated to the seven domains of the energy system: for power, Jesse Noffsinger, a McKinsey partner in Seattle, and Diego Hernandez Diaz, a McKinsey partner in Geneva; for mobility, Timo Möller, a McKinsey partner in Cologne and coleader of the McKinsey Center for Future Mobility; for industry, Michel Van Hoey, a McKinsey senior partner in Luxembourg; Christian Hoffmann, a McKinsey partner in Düsseldorf; Ken Somers, a McKinsey partner in Brussels; and Adam Youngman, a McKinsey senior asset leader in Los Angeles; for buildings, Daniel Cramer, a senior McKinsey asset leader in New York; for raw materials, Michel Foucart, a McKinsey associate partner in Brussels; Michel Van Hoey; and Patricia Bingoto, a McKinsey senior expert in Zurich; for hydrogen and other energy carriers, Rory Clune, a senior partner in Boston; and for carbon and energy reductions, Clint Wood, a McKinsey partner in Houston, and Santhosh Shankar, a US-based McKinsey expert. For their considered contributions to the research, we also thank Olivia White, McKinsey senior partner and a director of MGI in San Francisco, and Jan Mischke, MGI partner in Zurich.

The project team was led by Masud Ally, Francisco Galtieri, Kasmet Niyongabo, and Luc Oster-Pecqueur, and comprised Kemi Ajala, Sanjana Are, Maya Berlinger, Andrea Boza Zanatta, Susan Cheboror, Patrick Chen, Thibault Courqueux, Anurag Dash, John Grabda, Muriel Jacques, Myer Johnson-Potter, Pauline Leeuwenburg, Pierre Salvador, Girish Selvaraj, Anna Schneider, Casey Timmons, Tse Uwejamomere, Marnix Verhoeven, and David Wu. We are grateful to Janet Bush, MGI executive editor, who helped write and edit the report, and Juan M. Velasco, who helped with data visualization.

For kindly sharing their insights, we thank advisors Simon Dietz, professor, Grantham Research Institute on Climate Change and the Environment; Marion Dumas, professor, Grantham Research Institute; and John Ward, founder, Pengwern Associates, and visiting senior fellow, Grantham Research Institute.

We are also grateful to the following for taking the time to discuss the findings of this research and sharing their views with us: Jesse Jenkins, assistant professor of mechanical and aerospace engineering at the Andlinger Center for Energy and the Environment at Princeton University; Ted Nordhaus, founder and executive director of the Breakthrough Institute; Vijay Modi, a professor of mechanical engineering at Columbia University and faculty member of the Earth Institute; and Gregory F. Nemet, Vilas Distinguished Achievement Professor, La Follette School of Public Affairs, University of Wisconsin-Madison; and Daniel Schrag, the Sturgis Hooper Professor of Geology, Professor of Environmental Science and Engineering at Harvard University, and Co-Director of the Science, Technology, and Public Policy Program at Harvard’s Kennedy School.

Many McKinsey colleagues gave us input and guidance. We want to thank Enric Auladell Bernat, Deston Barger, Henrik Becker, Christian Begon, Michele Benoit, Krysta Biniek, Milo Boers, Brodie Boland, Janice Bolen, Michaela Brandl, Greg Callaway, Julian Conzade, Peter Cooper, Andreas Cornet, Matteo Cutrera, Thomas Czigler, Danny Van Dooren, Treina Fabre, Javier Ferrer, Lauritz Fischer, Wenting Gao, Godart van Gendt, Nicolas Goffaux, Jose Luis Gonzalez, Anna Granskog, Darya Guettler, Rajat Gupta, Marcin Hajlasz, Bernd Heid, Tom Hellstern, Russell Hensley, Anna Herlt, Ruth Heuss, Ann Hewitt, Autumn Hong, Blake Houghton, Thomas Hundertmark, Lionel Johnnes, Adam Kendall, Arjen Kersing, Per Klevnäs, Anna (Orthofer) Kortis, Kevin Laczkowski, Joh Hann Lee, Mateusz Lesniak, Christopher Martens, Eduardo Mencarini, Takashi Nakachi, Tomas Nauclér, Geoff Olynyk, Alex Panas, Jan Paulitschek, Sebastian Reiter, Gustavo Ribeiro, Daniel Riefer, Alexandre Van de Rijt, Moritz Rittstieg, Giulio Scopacasa, Suvojoy Sengupta, Bram Smeets, Hady Soliman, Brandon Stackhouse, Stephanie Stefanski, Michelle Stitz, Carlo Tanghetti, Tom Thys, Felix Tigges, Joaquin Ubogui, José Urgel, Steven Vercammen, Tom Voet, Maurits Waardenburg, Jeremy Wallach, Markus Wilthaner, Marita Winslade, and Nicola Zanardi.

In MGI’s operations team, we would like to thank Rachel Robinson and Rishabh Chaturvedi. For their help with digital production, we thank Chuck Burke and David Batcheck; and for their communications expertise, Rebeca Robboy, Nienke Beuwer, Shannon Ensor, and Ashley Grant. We are also grateful to communications colleagues in McKinsey’s Global Energy and Materials Practice and McKinsey’s Sustainability Practice, Lisa Farrugia and Kristen Jennings. Thanks also go to McKinsey’s design team, especially to Nathan R. Wilson and Janet Michaud. Finally, we appreciate the collaboration with other members of McKinsey’s digital production team, including Sean M. Conrad, Mary Gayen, Paromita Ghosh, Stephen Landau, and Regina Small.

As with all MGI research, this work is independent and has not been commissioned or sponsored in any way by any business, government, or other institution. While we gathered a variety of perspectives, our views have been independently formed and articulated in this report. Any errors are our own.

Understanding the physical realities of the energy transition—namely the physical properties of low-emissions solutions and the nature of the physical transformation—is critical to many aspects of designing a successful transition.

First, understanding the physical properties of low-emissions solutions can help design a new system that delivers performance on a par with the current system and does so reliably. This matters because the energy system is vital for driving economic growth and progress. As discussed later, this is not a trivial task and it requires a careful understanding of the performance and advantages of low-emissions technologies, innovation needs, and how such technologies can effectively be brought together in an interconnected system to deliver performance.

Second, looking at the nature and scale of the underlying physical transformation helps design a feasible transition. In an energy system made up of thousands or millions, and in some cases billions, of individual assets, the transformation that would be needed is monumental. With such a massive scale-up, bottlenecks in the build-out of supply chains could lead to shortages of critical minerals and manufactured goods. Installing or building new low-emissions assets at the scale and pace needed may be similarly difficult if not planned for well.

Third, and relatedly, applying a physical lens to different components of the energy system can highlight critical interdependencies, which similarly need to be factored into the design of a reliable and feasible energy transition.

Fourth, understanding the physical properties and maturity of different technologies, and the nature of the physical transformation, also helps to shed light on their costs and therefore on the affordability of the transition.¹It is also important to take a holistic view of the socioeconomic impacts of different transition pathways, and to use this to help inform decision making. See Climate Transition Impact Framework: Essential elements for an equitable and inclusive transition, McKinsey Sustainability, December 2023; and “Solving the net-zero equation: Nine requirements for a more orderly transition, McKinsey Sustainability, October 27, 2021. Prior McKinsey research has highlighted the large scale-up needed in low-emissions capital spending and various challenges associated with the affordability of the transition.²See, for example, The net-zero transition: What it would cost, what it could bring, McKinsey Global Institute, January 2022; An affordable, reliable, competitive path to net zero, McKinsey Sustainability, November 30, 2023; and From poverty to empowerment: Raising the bar for sustainable and inclusive growth, McKinsey Global Institute, September 2023. While costs associated with the transition are not the core focus of this research, appreciating the physical realities of the transition is crucial to better understand cost challenges. For example, in the case of carbon capture technologies, expanding their use to new use cases would require deploying them in processes where CO₂ makes up a small portion of the gases that are emitted (that is, is present in lower concentration in flue gases) and is therefore harder to capture. This could be about three times more expensive than the cost of capture of higher-concentration use cases deployed today.³See chapter 7, Challenge 24. The massive physical scale-up of the assets needed for a new system could also lead to shortages of raw materials and, as a result, contribute to price increases and create volatility. In 2022, prices of cobalt, lithium, and nickel surged, leading to an increase in the price of batteries of nearly 10 percent globally.⁴Energy technology perspectives 2023, IEA, January 2023; IEA clean energy equipment price index, 2014–2023, IEA, September 7, 2023; and Trends in electric vehicle batteries, IEA, April 2023. A sharp drop in prices quickly followed. This volatility generated uncertainty that contributed to the postponement of new mining projects.⁵Thomas Biesheuvel, “Battery metal price plunge is closing mines and killing deals,” Bloomberg Law, January 9, 2024; and Aya Dufour, “Some minerals are ‘critical’ to the digital economy, but current prices don’t reflect that,” CBC News, March 4, 2024.

Thus, a physical lens brings focus on not just how to achieve emissions reduction feasibly but also to do so while ensuring affordability, maintaining the reliability of the energy system, and thus also securing the competitiveness of companies and economies—three other objectives that McKinsey research has identified as vital for a successful transition.⁶An affordable, reliable, competitive path to net zero, McKinsey Sustainability, November 2023.

This research focuses on understanding the physical challenges of the energy transition. Important methodological choices were made to do this.

The focus is the energy system, encompassing both production and use (including the current use of fossil fuels as feedstock for industrial processes). The system accounts for more than 85 percent of current CO₂ emissions.¹Global CO₂ emissions from energy combustion and industrial processes total about 37 gigatonnes, with about five gigatonnes in agriculture, forestry, and other land use. In the case of methane, more than 35 percent of global emissions arise from the energy system, from combustion and industrial processes, with the remainder split between agriculture at about 40 percent and waste and other sectors at about 25 percent; McKinsey EMIT database, 2023. Sources of emissions outside the energy system, including in agriculture, forestry, and other land use, are not included. Other important sustainability topics, including the preservation of natural capital and the impact of pollution beyond greenhouse gas emissions, are also not within scope. In each domain of the energy system, the analysis explores what physical asset and process transformations would be required when switching from high-emissions assets to low-emissions alternatives. Examples include switches from fossil-fuel-based power generation, such as coal power plants, to low-emissions sources like variable renewable energy in the form of solar and wind, and clean firm power like nuclear or hydropower in the power domain; from ICE vehicles to EVs in the mobility domain; and from gas boilers to low-emissions heat sources in industry or buildings. The associated infrastructure and supply chains that would need to be built to support these switches are also analyzed.

Based on these transformations, the research then identifies 25 physical challenges that must be addressed for CO₂ emissions of the energy system to be reduced while replicating the performance of the existing energy system. These challenges were identified in consultation with more than 50 industry experts and academics within and outside McKinsey alongside an extensive literature review of analysis of the energy system.²This includes reviews of the level of progress in clean technologies and associated challenges by McKinsey and others. Among other research, see, for instance, Hauke Engel, Mekala Krishnan, Hamid Samandari, Humayun Tai, Daniel Pacthod, Simran Khural, and Mackenzie Murphy, A sector progress tracker for the net-zero transition, McKinsey Sustainability, November 2023; Energy technology perspectives 2023, IEA, January 2023; Tracking clean energy progress 2023, IEA, July 2023; Net zero roadmap: A global pathway to keep the 1.5ºC goal in reach, IEA, September 2023; World energy transitions outlook 2023, International Renewable Energy Agency, 2023; Systems Change Lab data dashboard, accessed May 2024; Climate change 2022: Impacts, adaptation, and vulnerability, Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, 2022; ETP Clean energy technology guide, IEA, updated September 14, 2023; The state of clean technology manufacturing, IEA, May 2023; New energy outlook 2023, BloombergNEF, 2023; Global critical minerals outlook 2024, IEA, May 2024; The state of clean technology manufacturing, IEA, May 2023; Material and resource requirements for the energy transition, Energy Transitions Commission, July 2023; and Better, faster, cleaner: Securing clean energy technology supply chains, Energy Transitions Commission, June 2023.

The 25 challenges are prioritized based on the potential of new, low-emissions technologies to abate emissions. Some exclusions help bound the scope of the work.

First, challenges that are expected to affect only a small portion of total emissions are not included. Second, incremental improvements to existing assets that do not involve major switches in technologies are not directly discussed as individual challenges; two examples are improved ICE fuel efficiency and insulation in buildings. Nevertheless, their collective impact is recognized in Challenge 23. Third, as noted, the challenges focus only on the energy system; those related to the transition of agriculture and other land use are not discussed directly, although the role of land as a physical challenge is discussed as part of the power domain challenges. Fourth, the focus is on challenges of a physical nature; any challenges that are purely related to market adoption or policies are excluded. Fifth, this research does not explicitly cover challenges related to labor. Finally, since this work focuses on analyzing the physical realities of the transition, costs are not the main focus, although, as noted, physical realities can help shed light on cost challenges.

The choice and precise boundaries of the 25 challenges is subjective to a degree, and some challenges are broader in scope than others. Different taxonomies, granularity, or segmentation of some challenges would certainly be possible. For example, circularity and recycling are important cross-cutting challenges that are discussed in the context of individual materials, such as plastics and critical minerals, but they could be deemed challenges in themselves. The list of 25 is neither collectively exhaustive (as noted, a prioritization lens has been used) nor mutually exclusive (many challenges share interdependencies).

The challenges are categorized into three levels, reflecting both the progress made to date in addressing them and the nature of the hurdles to overcome. Three features of difficulty, discussed further in chapter 3, are considered to do this: technological performance; gnarly interdependencies with other challenges; and degree of, and constraints on, scaling.

In examining the 25 challenges, this research builds on existing analyses of the transition in three ways. First, the examination of the performance of individual technologies is done in the context of specific use cases rather than their technological maturity in general. Second, this analysis goes beyond assessing technological maturity to consider other physical challenges, such as the required scale-up of supply of critical minerals. Finally, it considers how the system as a whole interacts—including how a particular individual technology relies on others—and the implications of that interaction.

Of course, the precise boundaries between the levels of challenges can be debated, and the classification into levels can vary by region. Parts of a Level 3 challenge could be categorized as Level 1 or 2. For instance, overall, decarbonizing cement is a Level 3 challenge that requires substantial technological innovation, but some decarbonization approaches, such as using biomass for heating or deploying clinker substitutes, are already widespread in some markets.

A global view of challenges is taken, but deployment of different technologies varies among regions. Some challenges may be more or less important—and difficult—depending on the region.

Across challenges, the research looks at the required deployment of low-emissions assets in 2050, comparing it with today’s levels using McKinsey’s 2023 Achieved Commitments scenario.³This research uses the 2023 McKinsey Achieved Commitments scenario because it provides detail across different economies and types of assets about the deployment levels that would be required for those economies to meet the climate commitments they have made. The scenario assumes that countries that have committed to net zero (some by 2050, some later) meet those commitments, and that warming reaches 1.6ºC relative to preindustrial levels by 2100. See Global energy perspective 2023, McKinsey, October 2023. Other net-zero scenarios may have slightly different combinations of technologies and rates of deployment, but the broad trends and themes described in this research would still apply. In some instances, this research also uses insights from other external scenarios for reasons of data availability.

Among the external sources of data used in this report are publicly available data from the International Energy Agency (IEA) in Paris, namely Energy technology perspectives 2023, IEA, January 2023; and Net zero roadmap: A global pathway to keep the 1.5ºC goal in reach 2023 update, IEA, September 2023. All are license CC BY 4.0. We note that some analysis in this research was derived from IEA material, and MGI is solely liable and responsible for it; it is not endorsed by the IEA in any manner. This holds true for all providers of the data that went into our analysis. We gratefully acknowledge their input, but the conclusions and any errors are our own.