1. Introduction and context

In the three decades since the founding of the UNFCCC, progress on climate change mitigation has been inadequate, with a large gap between ambition and progress (Canadell, Reference Canadell2021; Jackson et al., Reference Jackson, Friedlingstein, Le Quéré, Abernethy, Andrew, Canadell, Ciais, Davis, Deng, Liu, Korsbakken and Peters2022; Liu et al., Reference Liu, Deng, Davis and Ciais2024). Paris Agreement goals (‘well below 2 °C above pre-industrial levels and pursuing efforts to limit the temperature increase to 1.5 °C above pre-industrial levels’ [UNFCCC, 2015]) are threatened by inadequate policy commitment as well as credibility and implementation gaps (Fransen et al., Reference Fransen, Meckling, Stünzi, Schmidt, Egli, Schmid and Beaton2023; Rogelj et al., Reference Rogelj, Fransen, den Elzen, Lamboll, Schumer, Kuramochi, Hans, Mooldijk and Portugal-Pereira2023). Yet, ambition has endured with COP28 in 2023, for the first time calling for a transition away from fossil fuels to keep the 1.5 °C target within reach (UNFCCC, 2023a): a compromise which nevertheless fell short of fossil-fuel phaseout as called for by small-island states, scientists, and civil society (Morton, Reference Morton2023) and on the heels of a global stocktake (UNFCCC, 2023b). Despite many countries targeting global net-zero by mid-century (Net Zero Coalition, 2023), pathways remain uncertain. Moreover, the problem of temporality in global climate policy, set by the external and finite calendars of urgent action, contrasts with the typically open and indefinite calendars of negotiations between states (Chakrabarty, Reference Chakrabarty2021).

At the same time, complexity, interconnectedness, and concentration of energy use will continue to drive systemic risk (Goldin & Mariathasan, Reference Goldin and Mariathasan2014). Energy systems, as with all systems operating at regional and global scales, entail tightly coupled, interdependent flows (of embodied expertise and knowledge, materials and finance, exchange and trade, coordination and control, and relationships and power), which have come to be central to lives across the developed as well as developing worlds (Bair, Reference Bair2005; Castells, Reference Castells1996; Ibrahim et al., Reference Ibrahim, Centeno, Patterson and Callahan2021). Indeed, the growth in systemic risk, and its manifestations into polycrisis, follows the growing interdependence of societies (Centeno et al., Reference Centeno, Nag, Patterson, Shaver and Windawi2015). Systemic risk emergence is not only associated with the great acceleration (Steffen et al., Reference Steffen, Broadgate, Deutsch, Gaffney and Ludwig2015) but has much earlier roots in centuries-long trends of growing ability to concentrate useful work (Smil, Reference Smil2015) and concomitant societal and economic complexity (Smil, Reference Smil2018; Taylor & Tainter, Reference Taylor and Tainter2016). The widespread decarbonization of energy services across the net-zero transition needs to occur in the context of widespread increases in the density of energy utilization and primary energy consumption enabled by past energy revolutions (Mattick et al., Reference Mattick, Williams and Allenby2010; Smil, Reference Smil2018), which brought about specialization, concentration, and subsequent diminishing returns experienced across energy end-uses (Kemp, Reference Kemp, Centeno, Callahan, Larcey and Patterson2023; Tainter, Reference Tainter2011).

While previous large-scale energy transitions have mostly unfolded over several decades to over a century and even then have not become truly global (Smil, Reference Smil2020; Sovacool, Reference Sovacool2016), planetary stabilization now requires establishing alternatives to fossil fuels across sectors and on unprecedented scales and urgency, within a few decades. All this needs to be achieved amidst background complexity of our world, with the potential for heightened organization to bridge the innate low and variable energy density and diminishing returns of low-carbon energy sources with widespread and growing concentrations of end-use. These challenges can further kindle design choices that induce polycrisis dynamics, giving rise to highly interconnected electricity, materials, and fuel systems, with limited redundancy amidst difficult choices.

Here we explore through a structured approach many of the different system interactions at play in the low-carbon energy transition, so as to highlight key risks and concerns that must be managed if the transition is to be both successful and serve to reduce polycrisis risks more broadly. Section 2 discusses the imperative to align the transition within the context of the polycrisis we are now facing. Section 3 uses a systems map to identify critical risk transmission channels from the net-zero transition to other systems, and vice versa. Section 4 concludes by discussing research and analysis, long-term strategic thinking, and policy and decision-making needs to limit destabilizing feedback between the energy transition and the polycrisis.

2. Aligning net-zero transition with polycrisis reduction

Climate change is having a growing physical and socioeconomic impact across the world. The scale of impacts will continue to grow, most likely nonlinearly, until the world converts to a net-zero economy while adapting to climate change. Such a worldwide net-zero transition has been delayed by the implementation gap in reducing greenhouse gas emissions, as well as inadequate and unscaled counterbalance measures to limit emissions growth, and the global economy's lack of preparation for a full net-zero transition (Bossman et al., Reference Bossman, Gubareva and Teplova2023).

Scientific literature increasingly recognizes climate change as a critical component of the global polycrisis (Lawrence et al., Reference Lawrence, Homer-Dixon, Janzwood, Rockstöm, Renn and Donges2024). Global polycrisis is defined as ‘any combination of three or more interacting systemic risks with the potential to cause a cascading, runaway failure of Earth's natural and social systems that irreversibly and catastrophically degrades humanity's prospects’ (Janzwood & Homer-Dixon, Reference Janzwood and Homer-Dixon2022). It inherits four core properties of systemic risks that also interact to produce causal cascading effects: extreme complexity, high nonlinearity, transboundary causality, and deep uncertainty (Janzwood & Homer-Dixon, Reference Janzwood and Homer-Dixon2022).

Systemic risks are threats emerging primarily within an individual natural, social, or technological system that have impacts beyond its boundaries and endanger the functionality of one or more other systems (Jacobs, Reference Jacobs2024; Janzwood & Homer-Dixon, Reference Janzwood and Homer-Dixon2022). In the context of a net-zero transition, there are several interacting crises influencing the risk channels between the climate crisis, financial (in)stability, geopolitical energy crises, and the energy transition (Hoffart et al., Reference Hoffart, D'Orazio, Holz and Kemfert2024). The energy sector plays a critical role in achieving a zero-emissions future, and substantial investments are required to achieve climate goals. However, geopolitical turmoil, such as from the ongoing Russia–Ukraine war, creates uncertainties that can impede or even reverse progress in the energy transition, thereby diminishing investments in the global push for a clean energy transition (Bossman et al., Reference Bossman, Gubareva and Teplova2023).

Moreover, disregarding climate risks associated with the energy transition might result in serious threats to the financial and energy sectors. Amplifying risks within the financial and energy sectors can erode financial stability, hindering the net zero transition. Mitigating the broader risk of derailment, arising from interacting factors that can divert energy and political support for climate action and amplified by ongoing changes in the Earth system (Laybourn et al., Reference Laybourn, Evans and Dyke2023), requires wide-ranging policy measures including transformational adaptation to cope with risks (Pörtner et al., Reference Pörtner, Roberts, Tignor, Poloczanska, Mintenbeck, Alegría, Craig, Langsdorf, Löschke, Möller, Okem and Rama2022). Such interactions can have cascading effects on economies as well as the net-zero transition due to fossil fuel lock-ins and by inducing paucity of green finance (Hoffart et al., Reference Hoffart, D'Orazio, Holz and Kemfert2024). Additionally, there are several channels through which climate change can adversely impact sovereign debt, ranging from depletion of natural capital to international trade and capital flows (Zenios, Reference Zenios2024). Cross-border risks can be especially prevalent during the mid-transition during which fossil-fuel and low-carbon energy production co-exist (Espagne et al., Reference Espagne, Oman, Mercure, Svartzman, Volz, Pollitt, Semieniuk and Campiglio2023). In this wider context of various risks facing energy transitions, public understanding is critical to risk mitigation, with an urgent need to advance communication about systemic risks and their assessments as well as change mindsets and mental models (ASRA, 2024). Furthermore, successful net-zero transitions require forward-looking flexible funding mechanisms and strategies for the physical transitions as well as supporting policies (Kruczkiewicz et al., Reference Kruczkiewicz, Klopp, Fisher, Mason, McClain, Sheekh, Moss, Parks and Braneon2021). Also important is the need to grow public support of integrated solutions to climate change and inequality (Millward-Hopkins, Reference Millward-Hopkins2022).

Institutional strategies and communication framing are also crucial, requiring new long-term thinking about multilateral action as well as new approaches to understand and prepare for these systemic risks and their cascading effects across sectors, in the context of climate policy. The United Nations Development Program (UNDP) advises that for governments to accommodate the long term, they need to deliberately relinquish short-termism in their policymaking as well as governance. The UNDP suggests several strategies to achieve this goal: (i) Incorporating long-term thinking into political mandates and structures; (ii) Applying long-term and anticipatory thinking with strategic foresight methods for long-term planning; (iii) Examining how the future impact of major policies and programs can continue to enhance long-term thinking; and (iv) Promoting long-term thinking in the private sector (UNDP, 2022). The context of polycrisis calls additionally for multidimensional action and transformation across sectors, capacity building, and honest hope to enlist transformational possibilities (ASRA, 2024).

3. Systems mapping of net-zero risks amidst the polycrisis

To explore dynamics of the net-zero transition amidst the polycrisis, it is important to map it in a structured way as elements within a broader array of social, environmental, political, and technological systems. As illustrated here, doing so reveals that there are several proximate risks to the net-zero transition (Figure 1). Most prominent in recent years has been concern about the potential social and political backlash resulting from job losses in the economic sectors damaged by the transition (Gambhir et al., Reference Gambhir, Green and Pearson2018). This includes, most notably, fossil fuel-related sectors and those jobs and communities that depend on them, either directly or indirectly (Stark et al., Reference Stark, Gale and Murphy-Gregory2023). Such concerns have contributed to development of just transition policies and programs, such as the Just Energy Transition Partnerships (Kramer, Reference Kramer2022) undertaken in coal-dependent regions including South Africa, Indonesia, Vietnam, and Senegal. There is evidence that domestic inequality can induce voter preference for populist political movements (Pastor & Veronesi, Reference Pastor and Veronesi2021) and there is a causal effect of economic insecurity on populist resurgence (Scheiring et al., Reference Scheiring, Serrano-Alarcón, Moise, McNamara and Stuckler2024). Furthermore, populist resurgence can exacerbate political backlash against the energy transition (Wanvik & Haarstad, Reference Wanvik and Haarstad2021).

Figure 1.

A stylized systems map showing causal links between the net-zero transition and other systemic drivers. Note that a ‘+’ sign denotes a causal connection wherein if the first increases, so does the second; if the first decreases, so does the second. A ‘−’ sign denotes an inverse relationship: when the first factor increases, the second decreases; when the first factor decreases, the second increases. The blue coloring denotes proximate, or first order, risks FROM the low-carbon energy transition TO different factors (fossil fuel jobs; non-energy sector economic output; mineral mining); whereas the red coloring denotes proximate risks TO the low-carbon energy transition FROM different factors (political resistance to the transition; crop damage; energy infrastructure damage). Black arrows denote important causal relationships outside of the most proximate relationships indicated by blue and red arrows (Source: authors).

Despite the overall large economic benefits of a net-zero transition, there could be economic losses during the transition that are not always offset by near-term benefits of avoided impacts from slower global warming or wider co-benefits such as air quality and energy security (Lou et al., Reference Lou, Hultman, Patwardhan and Qiu2022). In any case, the bearers of costs and benefits, in terms of geographical regions, countries, economic sectors, demographic groups, and individual households, can be very different, further stoking social and political backlash. Whilst there are convincing arguments and analyses of the potential for the transition to stimulate and crowd in new investment (Köberle et al., Reference Köberle, Vandyck, Guivarch, Macaluso, Bosetti, Gambhir, Tavoni and Rogelj2021), thereby resulting in widespread direct economic gains rather than losses even in the near term, there is nevertheless a serious risk of adverse distributional effects. If not well managed through transition assistance policies, these effects could slow or even reverse the transition.

A further, relatively less well-explored risk stems from a warming world itself, with direct impacts on low-carbon energy technologies' efficacy, as well as demand for energy. For example, energy demand for cooling could increase in warming scenarios, as heatwaves become more frequent and intense (Yalew et al., Reference Yalew, van Vliet, Gernaat, Ludwig, Miara, Park, Byers, De Cian, Piontek, Iyer, Mouratiadou, Glynn, Hejazi, Dessens, Rochedo, Pietzcker, Schaeffer, Fujimori, Dasgupta and van Vuuren2020), whilst output from hydro power and low-carbon conventional power such as nuclear or gas with carbon capture and storage may suffer during hydrological droughts (Cronin et al., Reference Cronin, Anandarajah and Dessens2018). Further less well-understood impacts include lower solar photovoltaic output as a result of shading from wildfire smoke or increased dust from dryland expansion (Gilletly et al., Reference Gilletly, Jackson and Staid2023; Li et al., Reference Li, Mauzerall and Bergin2020), effects of warming on batteries for grid-scale energy storage (Hou et al., Reference Hou, Yang, Wang and Zhang2020), and even the effects of crossing climate tipping points on atmospheric systems (Armstrong McKay et al., Reference Armstrong McKay, Staal, Abrams, Winkelmann, Sakschewski, Loriani, Fetzer, Cornell, Rockström and Lenton2022; Wang et al., Reference Wang, Foster, Lenz, Kessler, Stroeve, Anderson, Turetsky, Betts, Zou, Liu, Boos and Hausfather2023) instrumental in renewable electricity production. A warming world has also been shown to lower potential crop yields, which could compromise bioenergy potential (Xu et al., Reference Xu, Wang, Gasser, Ciais, Peñuelas, Balkanski, Boucher, Janssens, Sardans, Clark, Cao, Xing, Chen, Wang, Tang and Zhang2022). Each of these adverse impacts could make the transition more costly, more contested, and thereby riskier. Of course, the faster the transition proceeds, the lower these latter risks from warming would become.

Energy transitions, whilst being at risk from a number of impacts, create risks of their own, which could proliferate through other systems and exacerbate existing societal and ecological vulnerabilities. Of high importance in near-term planning is the mineral demands of the transition (Bazilian, Reference Bazilian2018). In sheer material terms, the overall estimated impact of mineral demand is relatively small compared to current fossil fuels extraction (Bullard, Reference Bullard2023). Nonetheless, the concentration of mining in particular regions (e.g. lithium in Chile) risks creating considerable environmental harms if not well managed, which could then drive further societal and political resistance to the transition (Bartlett, Reference Bartlett2024). Countries rich in minerals important to the energy transition, especially when shaped by extractive institutions, can experience multiple and interacting dimensions of vulnerability. For example, Democratic Republic of Congo, while being vulnerable to climate and ecological change, is challenged by low human development and institutional capacity. Cobalt mining has contributed to occupational and environmental hazards as well as violent conflict and death, aggravated by the political economy of resource extraction (Sovacool, Reference Sovacool2019; UNEP, 2022). Moreover, the rapid spread of the mpox outbreak in eastern mining provinces of the DRC (WHO, 2024) points to multiple and interacting dimensions of vulnerability in a region that will remain crucial to the low-carbon transition. Additionally, concentration of supply chains for critical minerals, including in tightly coordinated recycling economies, can also induce new geopolitical risks (Blondeel et al., Reference Blondeel, Bradshaw, Bridge and Kuzemko2021).

A critical risk (and opportunity) facing society that is also shown in Figure 1, albeit somewhat speculatively, is that from pervasive use of artificial intelligence, which could be instrumental in driving low-carbon technology innovation. For example, machine learning could vastly accelerate energy research, for example, in devising new combinations to improve the performance and lower the cost of key technologies such as batteries, transportation fuels, and low-carbon building materials (Debnath et al., Reference Debnath, Creutzig, Sovacool and Shuckburgh2023a; Jin et al., Reference Jin, Ocone, Jiao and Xuan2020). In addition, electricity grid management could be enhanced to enable higher penetrations of variable renewables like wind and solar, while limiting costly redundancies (Boza & Evgeniou, Reference Boza and Evgeniou2021). By contrast, geo-political tensions exacerbated by AI, for example, through cyber-attacks (Guembe et al., Reference Guembe, Azeta, Misra, Osamor, Fernandez-Sanz and Pospelova2022) or pervasive disinformation, could have direct consequences on cross-border energy innovation as well as mineral mining and trade. In addition, increasing energy demand from ever-more powerful and demanding AI computation could accelerate global warming, though the extent to which this could be offset by improving energy efficiency or advances in computing technologies requires further research (Luers et al., Reference Luers, Koomey, Masanet, Gaffney, Creutzig, Lavista Ferres and Horvitz2024).

There are wider risks across countries arising from ongoing climate change as well as poorly conceived climate policies. Slow-onset climate changes could play a growing role in migration pressures in many parts of the world (Kaczan & Orgill-Meyer, Reference Kaczan and Orgill-Meyer2020). Coupled with the rise of fiscal austerity driven by various factors including demographic shifts, ideological pressures, supply shocks, or financial crises, migration pressures could fuel the rise of populism, political polarization, and fiscal conservatism. Especially whilst internal political competition is divisive, these trends can contribute to demoting internal cooperation as well as democratic processes (Lawrence et al., Reference Lawrence, Homer-Dixon, Janzwood, Rockstöm, Renn and Donges2024; Levin et al., Reference Levin, Milner and Perrings2021). Such factors are subject to threshold effects and can diminish domestic support for international coordination as well as green investments at crucial moments in the net-zero transition (Perrings et al., Reference Perrings, Hechter and Mamada2021). Additionally, exposure to inflation drivers (energy prices, currency depreciation) as well as boom-bust cycles in the economy can induce compensating monetary policy (e.g. higher interest rates to control inflation) that hinders low-carbon investments, but these non-equilibrium dynamics are not well understood or modeled (Pollitt & Mercure, Reference Pollitt and Mercure2018). Amidst these systemic risks, it remains a central challenge to stabilize virtuous cycles of expanding cost competitive low-carbon energy, such as renewable energy sources that are now amongst the cheapest available, as a global public good.

This tour of the low-carbon transition, rough and partial though it is, only covers those interconnections between the transition as a system and other systems of relevance. The net-zero energy system is itself a highly (and possibly increasingly) interconnected system, potentially at cross-continental scales, consisting of numerous new and hitherto untested technological combinations, for example, high voltage direct current electricity links, connecting power supply and demand across continents such as North Africa and Europe (Benasla et al., Reference Benasla, Allaoui, Brahami, Denaï and Sood2018), and combinations of continuous, intermittent, variable, and batch processes across the energy system, which have to be combined to run smoothly (Davis et al., Reference Davis, Lewis, Shaner, Aggarwal, Arent, Azevedo, Benson, Bradley, Brouwer, Chiang, Clack, Cohen, Doig, Edmonds, Fennell, Field, Hannegan, Hodge, Hoffert and Caldeira2018). There will also be numerous linkages between electricity networks, fuel and materials cycles, and heterogeneous networks of communication and control, giving rise to new failure modes (Gao et al., Reference Gao, Liu, Li and Havlin2015). Such a vast array of new technological combinations and interconnections also creates a heightened potential for ‘normal accidents’ (Perrow, Reference Perrow1984) within the energy system itself, quite apart from any risks cascading from or to other systems. This makes systemic risk assessment in energy transition planning all the more vital.