2. Minimising unjust harm

Unjust harm is already being caused by climate change and is set to escalate. There are hazard, exposure, and vulnerability components to this. The geography of several climate hazards falls disproportionately on the least advantaged (Mendelsohn et al., Reference Mendelsohn, Dinar and Williams2006), including exposure to extreme heat and humidity stress, which is biased to already hot places (Freychet et al., Reference Freychet, Hegerl, Lord, Lo, Mitchell and Collins2022). Furthermore, within a given geography and society, the least advantaged people tend to be more exposed and vulnerable to the impacts of climate change (Hallegatte & Rozenberg, Reference Hallegatte and Rozenberg2017).

This unjust harm is being caused by the activities that cause climate change, notably cumulative emissions of greenhouse gases from the fossil-fuelled energy system and the meat-oriented food system. While global warming increases roughly in proportion to cumulative emissions (expressed in terms of CO₂ equivalents, CO₂e), some key climate harms increase disproportionately with global warming, exacerbating their unjust consequences. Some also unfairly redistribute harm. Notably, global warming reduces cold-related deaths mostly among more advantaged people, while nonlinearly increasing heat-related deaths (Lüthi et al., Reference Lüthi, Fairless, Fischer, Scovronick, Ben, Coelho, Guo, Guo, Honda, Huber, Kyselý, Lavigne, Royé, Ryti, Silva, Urban, Gasparrini, Bresch and Vicedo-cabrera2023) mostly among less advantaged people (Bressler, Reference Bressler2025). The activities causing climate change also cause other inequitable harms. Notably, there are ∼5 million deaths per year from air pollution related to fossil fuel combustion (Lelieveld et al., Reference Lelieveld, Haines, Burnett, Tonne, Klingmüller, Münzel and Pozzer2023), and the poor are disproportionately exposed to air pollution (Rentschler & Leonova, Reference Rentschler and Leonova2023).

A faster transition means a faster decline in relevant activities and less cumulative emissions than a slower transition. Therefore, a faster transition minimises unjust harms. This can be quantified, with the example of heat-related mortality. The mortality cost of carbon synthesises existing studies to quantify the future heat-related deaths caused from a tonne of CO₂e emitted today (Bressler, Reference Bressler2021, Reference Bressler2025), with or without adaptation – meaning the assumption that increasing income will reduce vulnerability over time. Assuming adaptation and considering deaths until 2300, the (marginal) mortality cost of emissions in 2025 is 1.37 × 10⁻⁴ deaths/tCO₂e, or without adaptation 3.93 × 10⁻⁴ deaths/tCO₂e (Bressler, Reference Bressler2025). Thus, emitting 7300 tCO₂e (assuming adaptation) or 2540 tCO₂e (without adaptation) is expected to cause one premature death during 2025–2300. This is equivalent to the lifetime emissions of 5.7 or 2.0 average US citizens, or of 146 or 51 average internal combustion engine vehicles (ICEVs) (see Appendix). This mortality is weighted to hot countries and less advantaged people, with 75% occurring in South Asia and Sub-Saharan Africa (Bressler, Reference Bressler2025). Applying the principle of intergenerational equity to include premature deaths caused beyond the year 2300 would increase these mortality costs, especially in the case without adaptation, where they could more than double.

Thus, a faster transition can avoid massive, inequitable mortality. Accelerated transformation can also create new harms and injustices, but do they outweigh those alleviated?

3. An example of weighing harms and injustices

A well-publicised concern of increased harms is that the extractive industry needs to provide critical minerals for renewable energy and electrification (Pereira et al., Reference Pereira, Gianelli, Achieng, Amon, Archibald, Arif, Castro, Chimbadzwa, Coetzer, Field, Selomane, Sitas, Stevens, Villasante, Armani, Kimuyu, Adewumi, Lapola, Obura and Sumaila2024). Some of its practices are widely cited in the popular media as a cause of harm. A key example is artisanal cobalt (Co) mining in the Katanga region of the DRC, which provides a critical metal for lithium-ion batteries used in EVs.

In 2019, estimates ranging from 60–80,000 to >200,000 artisanal miners provided ∼12% of the cobalt supply from DRC (Gulley, Reference Gulley2023). Together with large-scale industrial mining, DRC provided around 70% of the global cobalt supply (Ritchie, Reference Ritchie2023). Artisanal mining brings benefits to miners but also causes harms, notably deaths and injuries from accidents in highly dangerous working conditions (Sovacool, Reference Sovacool2019), and it involves child labour (Gulley, Reference Gulley2023). Death rates are uncertain. The World Bank (2020) underestimates 65 per year based on media reports. Siddharth Kara estimates up to 2000 per year based on a questionnaire (∼1–2% of the workforce). Some mines have reported annual deaths of 0.4–0.5% of the workforce (Tsurukawa et al., Reference Tsurukawa, Prakash and Manhart2011), which equates to 240 to >1000 per year. I take ∼500 (65–2000) deaths/year as a central estimate (and range). Governance action is urgently needed to reduce this death toll (Sovacool, Reference Sovacool2019).

About a third (34%) of the world’s cobalt supply in 2021 went into EV batteries (the rest went into other batteries and electronics) (Ritchie, Reference Ritchie2023), so ∼170 (22–680) deaths per year from artisanal mining in DRC are associated with worldwide production of EVs. Not all battery technologies require cobalt (e.g. lithium iron phosphate batteries), but if we assume there is some artisanal mined cobalt from DRC in each of the 17 million EVs sold worldwide in 2024 (IEA, 2024b), the average EV was associated with 1.0 × 10⁻⁵ (1.3 × 10⁻⁶–4.0 × 10⁻⁵) deaths from artisanal mining in DRC over its lifetime. Thus, roughly every 100,000 EVs are currently associated with one artisanal mining death in DRC.

While seeking to avoid these mining deaths, the deaths avoided by substituting ICEVs with EVs should also be considered. The global fleet of around 1.5 billion cars and vans in 2023 (dominated by ICEVs) emitted 3.8 GtCO₂/year, which has a heat-related mortality cost, just from that year’s emissions, of 520,000 to 1.5 million (with or without adaptation). For the average ICEV with total lifetime emissions ∼50 tCO₂e (see Appendix), this translates to 6.9 × 10⁻³–2.0 × 10⁻² deaths/ICEV (with or without adaptation), or one future heat-related death for roughly every 100 (51–146) ICEVs over their lifetimes. Substituting an ICEV with a battery electric vehicle (BEV) reduces total lifetime emissions by 50–80%, depending on the electricity mix (Bieker, Reference Bieker2021). This avoids (3.5–5.5) × 10⁻³ deaths/BEV (assuming adaptation) or (1.0–1.6) × 10⁻² deaths/BEV (without adaptation) (see Appendix), which is about 1000 times (100–10,000 times) more heat-related deaths avoided per EV than currently caused in artisanal mining in DRC.

In addition, the air pollution generated over the lifetime of the average ICEV is responsible for 3.45 × 10⁻³ deaths/ICEV (see Appendix). BEVs have zero tailpipe emissions and, even in the worst case of running on 100% coal power, can avoid 3.35 × 10⁻³ deaths/BEV (see Appendix), comparable to the heat-related deaths avoided assuming adaptation.

This does not cover all causes of mortality. Accidents in other parts of the EV battery supply chain cause comparable lost life years to the lower estimate of artisanal mining deaths in DRC, and sulphur dioxide, zinc, and arsenic emissions from producing EV batteries cause more lost life years than the higher estimate of artisanal mining deaths in DRC (Arvidsson et al., Reference Arvidsson, Chordia and Nordelöf2022). Extracting and refining oil also causes mortality (Onyije et al., Reference Onyije, Hosseini, Togawa, Schüz and Olsson2021). Nevertheless, the mortality costs of carbon and air pollution remain orders of magnitude larger.

Thus, from a utilitarian global perspective, accelerating the transition to BEVs is (overwhelmingly) just. From a Rawlsian perspective, deaths caused in artisanal mining are among the least advantaged people worldwide, but so too are a far greater number of heat-related deaths avoided worldwide (including in DRC, Somalia, and Niger) (Bressler, Reference Bressler2025), while air pollution deaths avoided tend to be among poorer people in developing (especially middle-income) countries. Thus, from a prioritarian global perspective, accelerating the transition to BEVs is (overwhelmingly) just.

If one took a narrower view of ‘society’ as just DRC, each BEV avoids (1.4–2.2) × 10⁻⁴ heat-related deaths (with adaptation) to (5.0–8.0) × 10⁻⁴ heat-related deaths (without adaptation) in DRC (see Appendix), which is 3.5–615 times the deaths caused per BEV in artisanal cobalt mining in DRC. Clearly, governance should seek to avoid deaths in artisanal cobalt mining in DRC and its other negative impacts (Banza Lubaba Nkulu et al., Reference Banza Lubaba Nkulu, Casas, Haufroid, De Putter, Saenen, Kayembe-kitenge, Musa Obadia, Kyanika Wa Mukoma, Lunda Ilunga, Nawrot, Luboya Numbi, Smolders and Nemery2018; Sovacool, Reference Sovacool2019), but from a utilitarian perspective, accelerating the transition to EVs is just for the citizens of DRC. From a prioritarian perspective, it is hard to say whether the greater number of people in DRC avoiding heat-related death (in the future) will be less advantaged than those dying in artisanal cobalt mining today, but they could be. Today, over 85% of people in DRC live in extreme poverty (<$3/day) and artisanal cobalt miners are generally better off than this (Sovacool, Reference Sovacool2019). The fraction of DRC citizens living in extreme poverty increased from 2012 to 2020, meaning DRC’s people are at risk of not being able to reduce their vulnerability to heat-related deaths in future. However, the cobalt mining region of Katanga is notable for its economic development, and the DRC’s mineral wealth offers a clear opportunity to lift its people out of poverty and reduce their vulnerability. Thus, accelerating the transition to EVs while seeking to avoid mining deaths could be just for the citizens of DRC from a prioritarian perspective. Such judgements are complicated by the large class, ethnic, and other cleavages at play in determining who benefits from, and who pays the costs of, the transition. Ideally, costs should fall disproportionately on the richest and those who have historically contributed most to causing climate change.

4. Does speed compromise justice in other ways?

Several other justice concerns with accelerated transformation have been raised (Newell et al., Reference Newell, Geels and Sovacool2022; Pereira et al., Reference Pereira, Gianelli, Achieng, Amon, Archibald, Arif, Castro, Chimbadzwa, Coetzer, Field, Selomane, Sitas, Stevens, Villasante, Armani, Kimuyu, Adewumi, Lapola, Obura and Sumaila2024, Reference Pereira, Smith, Gifford, Newell, Villasante, Achieng, Castro, Constantino, Powell, Ghadiali, Smith, Vogel and Zimm2025).

Accelerated transformation could perpetuate or entrench existing injustices, notably because incumbents are in a better position to mobilise quickly and protect their interests (Newell et al., Reference Newell, Geels and Sovacool2022). A key concern is that unjust extractivism will be perpetuated by a new wave of critical mineral extraction from poorer to richer countries, producing ‘green sacrifice zones’ (including the example above) (Pereira et al., Reference Pereira, Gianelli, Achieng, Amon, Archibald, Arif, Castro, Chimbadzwa, Coetzer, Field, Selomane, Sitas, Stevens, Villasante, Armani, Kimuyu, Adewumi, Lapola, Obura and Sumaila2024, Reference Pereira, Smith, Gifford, Newell, Villasante, Achieng, Castro, Constantino, Powell, Ghadiali, Smith, Vogel and Zimm2025). However, globally, the extractive industry for clean energy will be smaller than the fossil fuel one it replaces (Nijnens et al., Reference Nijnens, Behrens, Kraan, Sprecher and Kleijn2023). Also, the high-value minerals extracted can form the basis of a more circular economy with less extraction (Ma et al., Reference Ma, Meng, Bellonia, Spangenberger, Harper, Gratz, Olivetti, Arsenault and Wang2025), whereas fossil fuel carbon cannot. This mineral recycling potential increases the incentive to accelerate transformation now (Fabre et al., Reference Fabre, Fodha and Ricci2020). Thus, accelerating transformation could hasten the demise of extractivism.

Accelerating transformation may preclude pathways and outcomes that are more just (Pereira et al., Reference Pereira, Gianelli, Achieng, Amon, Archibald, Arif, Castro, Chimbadzwa, Coetzer, Field, Selomane, Sitas, Stevens, Villasante, Armani, Kimuyu, Adewumi, Lapola, Obura and Sumaila2024), narrowing the range of transition pathways and creating path dependency that locks us into suboptimal solutions. For example, substituting ICEVs with EVs is not the best of all possible solutions. A deeper transformation to a different mobility regime, e.g. reducing mobility and switching to electrified public transport, could (ultimately) achieve a greater reduction in emissions (Pereira et al., Reference Pereira, Gianelli, Achieng, Amon, Archibald, Arif, Castro, Chimbadzwa, Coetzer, Field, Selomane, Sitas, Stevens, Villasante, Armani, Kimuyu, Adewumi, Lapola, Obura and Sumaila2024) and associated harms (Rose, Reference Rose2023). However, if (as implied) it takes longer to switch to a different mobility regime, then more cumulative emissions can cause more harm in the intervening time. Conceivably, an accelerated transition to EVs could subsequently undergo a (delayed) switch in mobility regime, starting from a point of lower transient emissions, and with lower ultimate cumulative emissions and associated harms.

A counter case can be made that accelerated transformation could create (rather than preclude) opportunities for deeper transformation, because it will cause greater disruption for incumbent actors. Accelerated transformation to EVs is a pertinent example. Already, ever cheaper batteries coming from the electrification of personal mobility are opening up other pathways of accelerating change to lower emissions (Lenton, Reference Lenton2025; Sharpe & Lenton, Reference Sharpe and Lenton2021). Also, cars are the dominant driver of demand for oil. In an accelerated transformation, oil companies and those invested in them will experience greater stranded assets sooner (Mercure et al., Reference Mercure, Pollitt, Viñuales, Edwards, Holden, Chewpreecha, Salas, Sognnaes, Lam and Knobloch2018; Semieniuk et al., Reference Semieniuk, Holden, Mercure, Salas, Pollitt, Jobson, Vercoulen, Chewpreecha, Edwards and Viñuales2022), while clean energy assets will grow faster. This should cause oil companies to diversify sooner (or go out of business) and those invested in them (e.g. pension funds) to move their investment elsewhere sooner (or suffer losses). If investors do not respond fast enough, then financial harm will fall upon pension holders in richer countries (Semieniuk et al., Reference Semieniuk, Holden, Mercure, Salas, Pollitt, Jobson, Vercoulen, Chewpreecha, Edwards and Viñuales2022). This may open a space for deeper cultural questioning of the incumbent paradigm.

Speed may trade off with procedural justice (Newell et al., Reference Newell, Geels and Sovacool2022), but this is uncertain and should not preclude prioritising justice concerns (Steel et al., Reference Steel, Vásquez-fernández, Crookall, Cripps, DesRoches and Mintz-Woo2026). Notably, participatory democratic processes can help forefront justice considerations but tend to be slow to conduct. However, they tend to lead to policy proposals that (if enacted) would greatly accelerate transformation. The deeper problem is that their policy proposals are generally not acted upon (P. J. Newell et al., Reference Newell, Geels and Sovacool2022; Lenton, Reference Lenton2025). Furthermore, incumbent actors in the fossil fuel sector have co-opted some ‘just transition’ processes to deliberately slow action that is against their vested interests (Newell et al., Reference Newell, Geels and Sovacool2022) – making a mockery of the concept.

More broadly, speed does not conflict with distributive justice. In the law, ‘justice delayed is justice denied’ – a speedy resolution of cases is widely seen as just for those suffering injustice. Vulnerable communities suffering escalating heat-related mortality may feel the same.

5. Maximising just benefits

Transforming the energy system has justice benefits, in addition to the harms avoided.

Renewable energy is already providing the cheapest source of new electricity almost everywhere (IRENA, 2025b). This can disproportionately benefit the poor, for example, off-grid solar power plus battery storage is starting to bring access to affordable electricity to the ∼700 million people that do not have it (IRENA, 2025a; Lenton, Reference Lenton2025). Several low- to middle-income countries have increased their reliance on low-carbon energy and reduced emissions while increasing per capita GDP and reducing income inequality (Reich et al., Reference Reich, Grace, Agrawal and Nagendra2025). Across 23 higher-income countries, a greater share of renewable energy substantially increases within-country energy justice (Sen et al., Reference Sen, Hosan, Karmaker, Chapman and Saha2024).

Renewable energy sources are more equitably distributed than fossil fuels (Overland et al., Reference Overland, Juraev and Vakulchuk2022). Just 12 countries supply 80% of net fossil fuel exports, whereas around 74% of the world’s population live in net fossil fuel-importing countries, including many of the poorest people. Switching to in-country generation of renewable energy benefits their balance of trade and economic conditions, and energy security. Meanwhile, the minority of fossil fuel exporters may be harmed if they do not transition their economies in a timely fashion, but they tend to be more advantaged people.

Clean energy gets cheaper the more that is deployed, thanks to several reinforcing feedbacks (Lenton, Reference Lenton2025). Hence, despite upfront costs, transforming the energy system saves a large amount of money, and accelerating transformation saves more money (Way et al., Reference Way, Ives, Mealy and Farmer2022). From a utilitarian perspective, transforming faster maximises benefits to total or average welfare. From a prioritarian perspective, transforming faster can bring greater benefits to the least advantaged people.