Introduction

Climate scientists often assert that attaining net zero global greenhouse gas (GHG) emissions by 2050 requires an unprecedented “acceleration” of climate action. [1,2] Since its introduction by the Paris Agreement of 2015, acceleration has become a buzzword in policy documents published by the United Nations, the Assessment Reports of the Intergovernmental Panel on Climate Change (IPCC), and other international climate organizations such as the Clean Energy Ministerial [3]. When inspecting the use of the term in policy documents in particular, we detect definitional ambiguity. This ambiguity is puzzling, given that acceleration has a clear mathematical definition: the second derivative of a quantity. If this quantity were, for example, the global average temperature, then the change in temperature over a given time would be known as the rate of change (first derivate), and the change in the rate of change over time would be the acceleration (second derivative) [4]. Conceptually, one could think of either larger changes in a given time period or a shortening of the time period in which the changes in temperature take place.

However, unambiguous terms in science can become ambiguous in governance contexts [5]. This can result from negotiations involving complex issues or the divergent interests of those participating in them, not to mention that negotiations entail time constraints. In addition to such “unintended” ambiguities, policymakers, for instance, might use ambiguous language in order to reach an agreement during the negotiation of a legal text, which is sometimes discussed under the term “constructive” ambiguity [6].

To some extent, ambiguity can be conducive to policy processes. For instance, the vagueness of the term “sustainability” has succeeded in forming large coalitions of actors advocating for sustainability-related action [7]. However, the formation of such coalitions has only seldom resulted in the adoption of tangible sustainability policies, since these actors hold contrasting views on how to achieve sustainability through policy design. [8] Consequently, the academic literature has warned that ambiguity could undermine progress by allowing parties to claim that they have taken forward steps, even though there has been no significant policy development [9].

We therefore contend that actors partaking in governance processes should aim to establish precise meanings of key concepts. This precision is particularly crucial given the urgent need to reduce GHG emissions. We believe that a clearer conceptual understanding of the acceleration of climate action will lead to higher-quality policy advice. This Essay proposes a policy-focused approach to accelerating climate action [10,11], drawing inspiration from scientific definitions of acceleration. Specifically, we argue that the concept of acceleration varies depending on whether we focus on policy outputs, outcomes, or impacts. By distinguishing between these dimensions, we can provide a more nuanced and actionable understanding of acceleration in climate action. We hereby complement the literature on accelerating net-zero transitions, which tends to focus on whole (socio-techno-ecological) systems [12,13].

Scientific perspective on acceleration

In science, velocity is the rate of change of displacement (position) with respect to time, and acceleration is the rate of change of the velocity (speed) with respect to time. Mathematically, velocity is expressed by the first derivative of the displacement and acceleration by the second derivative of the displacement. There are many ways that these terms are applied in climate science, such as the rate of increase of temperature [4] or, in climate mitigation, as the deployment rate of clean technologies (leading to S-curves) [14]. To put this in a climate policy context, we first focus on thinking of the “displacement” as emissions.

There are many potential pathways going from today’s GHG emissions to net zero, but all exhibit the characteristics shown in Fig 1: an initial transition from today’s emissions growth into emission reductions, followed by sustained maximum reductions in emissions, before a deceleration phase leading to zero emissions [16]. We demonstrate this with a symmetric S-curve for emissions. The S-curve exhibits the necessary properties for conceptualizing motion, speed, and acceleration in the context of climate action.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. A stylized depiction of GHG emissions trending toward net zero, showing a) absolute emissions levels (“GHG Emissions”), b) rate of change (“Speed”), and c) change in the rate of change (“Acceleration”).

There are three phases: acceleration, a constant change (only a single point in 2050 in this example), and deceleration. These historical data are from the UNEP Emissions Gap Report [15], while future pathways is a logistic function designed to reach net zero GHG emissions around 2080.

https://doi.org/10.1371/journal.pclm.0000626.g001

Starting from the current day, there needs to be an initial period (2020–2050) in which GHG emission reductions “speed up” (Fig 1a). This should then lead to an increasing rate of GHG emission reductions (Fig 1b), followed by an increasing acceleration (Fig 1c). This acceleration must eventually slow down to zero, otherwise continuous reductions will come to exceed physical bounds, creating a maximum speed when acceleration is zero again (black dots in 2050 in Fig 1). In the “slow down” period (2050–2080), there is a decreasing rate of emission reductions (Fig 1b) and a decreasing acceleration (Fig 1b). Interestingly, the acceleration has to become negative (deceleration or retardation) as the emission reductions slow down (declining speed) and eventually settle to zero or stable emissions. In principle, it is possible to allow the emissions to become negative, but eventually, the emissions would probably stabilize at a constant value, otherwise all GHGs would be removed from the atmosphere.

These three phases (acceleration, constant speed, deceleration) are familiar to most people taking any form of transport, and for similar reasons, they also apply to the evolution of GHG emission reductions over time. Fig 1 equally applies to a vehicle that accelerates to a maximum speed, then decelerates to a stop again: The emissions are the distance traveled, though the sign would be different, as the conventional starting point would be the location of zero.

These concepts are also used in the literature on innovation, where technology deployment is often represented with an S-curve: Initially, the rate of deployment is exponential, but after reaching maximum speed (zero acceleration), it slows down (deceleration) and eventually levels out as resource constraints or maximum capacity is reached [17–19]. While the exact shape of any curve will vary depending on country, application, etc., the three phases will remain.

Acceleration in climate policy documents

In the Paris Agreement to the United Nations Convention on Climate Change (UNFCCC) of 2015, the term “accelerating” is mentioned once, in Article 10.5 on technology development and transfer: ‘Accelerating, encouraging and enabling innovation is critical for an effective, long-term global response to climate change and promoting economic growth and sustainable development.’ Later, acceleration became a more common part of UNFCCC texts.

The 26^th Conference of the Parties to the UNFCCC (COP26) Cover Decision from the summit in Glasgow in 2021 takes the acceleration language further and uses the term five times. It is mostly used in the section on climate change mitigation, where it describes the need for more ambitious climate action before 2030 and states what needs to be done in relation to mitigation and climate action generally.

The Proposal by the President for the First Global Stocktake at COP28 in 2023 mentions acceleration 18 times [20]. The most important section in this regard is point 28 in the proposal’s part on collective progress toward achieving the purpose and long-term goals of the Paris Agreement, as six of the eight subpoints use the word “accelerating”.

In addition to the UNFCCC policy documents, we must consult those produced by the IPCC, which provides the most authoritative scientific, technical, and socioeconomic information on climate change for the UNFCCC Secretariat [21]. Chapter 17 in the Sixth Assessment Report of Working Group III of the IPCC has ‘accelerating the transition’ in the title and seeks to offer a conceptual clarification of the term [1]. In contrast to the UNFCCC documents, this report explains that acceleration involves ‘expediting the pace of change (speed) while also removing the underlying drivers of vulnerability and high emissions (quality and depth), and aligning the interests of different communities, regions, sectors, stakeholders and cultures (scale and breadth).’ Interestingly, while chapter 17 explicitly mentions “speed”, it is less explicit than the UNFCCC regarding what needs to be changed or “sped up.” The remainder of this chapter emphasizes the need for multi-level governance and cross-sectoral policy coordination and integration.

Our review of key policy documents from the UNFCCC Secretariat and the IPCC reveals that they employ a broad understanding of acceleration in relation to climate action—encompassing any measures taken by individual or collective actors to address climate change [22]. The most ambiguous uses in the consulted documents refer to ‘expediting the pace of change’ (Working Group III of the IPCC), acceleration of ‘efforts’ or ‘action’ (policy documents from COP26 and COP28), and ‘sectoral action’ (COP26 documents).

The COP26 Cover Decision demonstrates a particularly comprehensive understanding of ‘acceleration,’ referring to it in conjunction with ‘policies,’ ‘innovations,’ and ‘technologies.’ While the Paris Agreement also connects innovations with acceleration, this represents its sole conceptualization. Similarly, the President’s Proposal for the First Global Stocktake at COP28 mentions technologies in relation to acceleration, but it also references the acceleration of GHG emissions—in fact, this is the only document in which this is done.

Table 1 provides an overview of the various terms used in conjunction with acceleration. Based on our review, we contend that while all policy documents contribute to conceptualizing accelerated climate action, they need synthesis and connection with a policymaking framework informed by the scientific definition of acceleration if they are to guide policymakers effectively—a task we address in the next section.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Acceleration-related terms in policy documents.

https://doi.org/10.1371/journal.pclm.0000626.t001

A framework for guiding policymaking to accelerate climate action

Despite the diverse terminology used in the climate policy documents of the UNFCCC and IPCC, basic concepts from policy analysis can help to systematize the actions that policymakers must take to accelerate climate action. For our purpose, it is important to differentiate between policies themselves and the effects they produce. Policies are the direct “outputs” of the policy process, involving politicians and public officials who propose and adopt legally binding acts such as laws. The policies themselves comprise a set of policy instruments (e.g. regulation, direct provision, economic or information instruments) [23–26], which are calibrated at a given level [27]. Once implemented, these policies are expected to change the behavior of target groups, which the policy literature refers to as policy “outcomes.” The aggregate effects resulting from these behavioral changes are the “policy impacts” [28,29].

To illustrate these three concepts, we focus on policies aimed at reducing GHG emissions from residential heating. The possible policy output for achieving this goal would be a subsidy for the installation of heat pumps to replace fossil-fuel heating devices. The policy outcome would be homeowners replacing their heating systems with heat pumps. The policy impact would be reduced GHG emissions from residential heating.

Applying this basic distinction between policies and their effects to the climate policy documents reviewed above, we can observe references to all three dimensions. Some mentions in the policy documents refer to the need for policymakers to produce policy outputs, while others focus on the desired outcomes of these policies, that is, accelerating innovation and the uptake of technology [3,14,30], and their impacts, that is, accelerating GHG emission reductions [31,32] (see last column in Table 1). From this perspective, the policy documents collectively call for comprehensive policy action. However, it is unclear what constitutes ‘comprehensive policy action’ because these documents refer to both policies and their effects without clearly distinguishing between them.

Having clarified that the UNFCCC and IPCC policy documents actually make a strong call for policy-driven climate action, we now turn to how the literature on public policy and the scientific definition of acceleration can provide guidance to policymakers for achieving this.

Based on the President’s Proposal for the First Global Stocktake at COP28, we contend that the policy-driven acceleration of climate action fundamentally refers to increasing the speed of GHG emission reductions (see also Fig 1c). While we could regard the diffusion of technologies and innovations as policy effects that should be accelerated, we view these as primarily instrumental in achieving the overarching goal of reducing GHG emissions. Consequently, policymaking needs to focus on achieving this specific policy impact; however, ‘capable policy design’ [33] is essential for producing policy outputs that stand a good chance of producing the intended outcomes and impacts.

Policy studies have shown that while policymakers maintain a high degree of control over policy outputs, this influence weakens progressively when moving toward policy outcomes, and diminishes further vis-a-vis policy impacts [28,34–36]. Fig 2 illustrates that the correlation between policymakers’ intentions (e.g. to reduce GHG emissions by 50%) and their actual implementation (e.g. a reduction of GHG emissions by 40%) varies significantly depending on whether we examine policy outputs, outcomes, or impacts. The relationship between the intentions of policymakers and the policies they propose and adopt (policy outputs) is the strongest one, but time constraints, insufficient knowledge, political compromises, and other limiting factors can hamper the quality of the decisions made [37].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Illustration of varying strength of correlation between the intentions of policymakers and the policy outputs, outcomes, and impacts.

https://doi.org/10.1371/journal.pclm.0000626.g002

Policy outcomes refer to changes in the behavior of target groups as intended by a policy. These outcomes result from both the state bureaucracy’s implementation of policy outputs and the target populations’ level of compliance with policy stipulations [36,38]. Another complicating factor is the existence of policies that provide conflicting incentives to target groups, potentially influencing their behavior in opposite directions—an aspect extensively discussed in the interdisciplinary literature on the design of (climate) policy mixes [26,39,40].

It is even more difficult for policymakers to control the aggregate change in the target groups’ behavior as a result of achieving specific policy goals (in this case, GHG emission reductions). This challenge arises because the behavior of a large number of policy addressees can be influenced by a broad range of factors, including technological, political, economic, and geopolitical developments, as well as various types of crises [3,41]. Since policy impacts encompass the aggregate effects of policies, the behavior of actors outside the target populations becomes relevant, as they may act strategically to create benefits for themselves [29].

However, it should be noted that while we depict a weakening relationship from policy outputs to outcomes and impacts (as indicated by the increasingly lighter blue color of the boxes in Fig 2), the relationship can also be strengthened if a policy performs better than expected, for example, by effectively leveraging emerging social or technical trends.

There are many potential scenarios illustrating how the climate policy outcomes may differ from the apparent policy intentions as reflected in the policy outputs. The Paris Agreement and the IPCC Special Report on Global Warming of 1.5°C arguably led to the proliferation of ambitious net-zero climate targets, both at the country and company levels [42]. However, there remains a large gap between policy targets and policies as they are implemented [36,43].

The same holds true for the relationship between policy outputs and policy impacts. Even as climate policies seem to strengthen [23–25], GHG emissions continue to grow [44]. One reason for this can be seen in the adverse incentives climate policy may produce for non-target groups. For example, the adoption of an Emissions Trading System (ETS) will have a direct effect on companies within the ETS, but it could also lead to activity or emissions shifting to areas outside the ETS. This shift could be negative, leading to “carbon leakage” [45]. To illustrate this point, the European Union’s (EU) ETS provides incentives to EU-based airlines to reduce their GHG emissions, but it has also created incentives to redirect traffic flows to countries not covered by the system but close to EU territory, such as Turkey [46]. Thus, a capable policy design would not only affect airlines based in the EU but also in neighboring countries. Conversely, the effects on non-target groups can also be positive, leading to “spillovers” [47].

Thus, to accelerate GHG emission reductions, policymakers must recognize their significant control over policy outputs and therefore prioritize the careful design of climate policies and effective policy mixes [26,39]. Furthermore, based on the scientific definition of acceleration, policymakers should understand that accelerating (to which we refer as “stepping up” when talking about policymaking) climate action must also be accompanied by maintaining it and, when appropriate, decelerating (to which we refer as “stepping down” when talking about policymaking) certain measures.

When considering policy outputs, one might initially equate stepping up with merely increasing the quantity of climate policies and the speed by which they are adopted. However, this approach –adopting more climate policies faster – can provide an inadequate framework for effective climate action, as explained in the subsequent paragraphs. Instead, we advocate for a more nuanced understanding: Stepping up should refer not to the rate of policy production, but rather to enhancing the incentives that these policies provide for accelerating technological change and reducing deployment barriers, which then translate into reducing GHG emissions more quickly [32,48].

Importantly, as technologies mature and become self-sustaining, policy support may need to undergo strategic reduction and an ultimate phase-out once market saturation is achieved. This would represent an instance of stepping down.

For the period it takes to achieve technological maturation, policymakers should maintain the relevant policies intended to catalyze this process. This means avoiding the adoption of superfluous additional policies that might overburden the public bureaucracy and/or create contradictory incentives for the target populations [40], while also ensuring that existing and effective policies are neither dismantled nor terminated prematurely [25,49,50].

Electric vehicle (EV) policy in Norway offers a good example of a policy strategy for accelerating a reduction in GHG emissions. In fact, it follows the three phases mentioned above. Policy support was initially strong and broad until nearly all new vehicle purchases were EVs (stepping up), then policy support was reduced as it was no longer needed to ensure that nearly 100% of new car sales were EVs. The winding back of EV support (stepping down) subsequently liberated resources and capacity to focus on other climate policy priorities. Other examples exist across many technologies that are important for reducing GHG emissions, such as Germany’s introduction, maintenance, and eventual elimination of its renewables levy on consumer power prices in 2022 [48].

Taken together, our conceptual reflections on the acceleration of climate action call for an agile policy approach that undergoes regular review and evaluation, and in which policymakers are willing to learn from the insights produced by these assessments. Effective evaluation-based learning should maintain policies that accelerate GHG emission reductions while modifying or eliminating those that prove ineffective.

Implications for the policy-driven acceleration of climate action

Our integrated understanding of the policy-driven acceleration of climate action can guide climate governance across multiple levels in three key ways.

First, UNFCCC negotiators crafting agreements to support the Paris Agreement objectives should exercise precision when referencing “acceleration” and clearly articulate what this concept entails in practice. The current trajectory of increased use of this term without sufficient definition may make it hard for national policymakers to interpret what the UNFCC decisions mean. Ambiguous use of the term may undermine, rather than underpin, development of solid national governance systems that facilitate the development of effective climate policies.

Second, for national and sub-national policymakers to design effective climate policies that accelerate GHG emission reductions, they should recognize that acceleration ultimately hinges on the multiple factors that shape policy outcomes and impacts. Although impacts take time to materialize, the factors affecting them warrant immediate attention at the stage of policy design. Policies should provide appropriate incentives to target groups while discouraging strategic exploitation by non-targeted actors.

Furthermore, policy-driven acceleration should be understood as one phase in a process, followed by maintenance and eventual deceleration phases. We propose the terms “stepping up,” “maintaining,” and “stepping down” when referring to climate action that aims to achieve acceleration of GHG emission reductions.

The emission reduction trajectory will tend to be at different phases in different sectors, relying on how long climate policies have operated in the sectors in question. For instance, since many countries introduced climate policies in the energy sector first [48], this sector will probably reach the maintenance phase earlier than other sectors, such as agriculture [51]. Hence, the climate policy in the energy sector may be stepped down at the same time as policies are stepped up in the agricultural sector. Agile policymaking that remains attentive to the role of policy in these stages will increase the likelihood of countries meeting national GHG emission reduction targets on time.

Third, climate governance evaluators can benefit from these insights by considering the broad range of factors that facilitate or hinder GHG emission reductions. They should analyze output, outcome, and impact phases comprehensively. This multidimensional evaluation approach is especially important for national climate councils, such as the United Kingdom Climate Change Committee, which are tasked with rigorous policy evaluations [52]. These bodies must carefully assess the role of climate policy in national acceleration processes for climate action and, to the extent that acceleration is policy-driven, distinguish whether obstacles to acceleration stem from policy output deficiencies, implementation challenges, or impact factors originating from broader technological, economic, political or societal issues.

Conclusion

UNFCCC and IPCC policy documents have increasingly adopted the term “acceleration” in relation to climate action, shaping the global discourse. However, this terminology carries significant ambiguity. While we cannot determine whether this ambiguity in the UNFCCC and IPCC policy documents is unintentional or deliberately constructed [6], we believe it is essential to address the issue and provide a conceptual framework to guide both policy analysis and policymaking.

In this Essay, we advocate for differentiating between policy outputs, outcomes, and impacts when conceptualizing the policy-driven acceleration of climate action. Of these three dimensions, policy impacts are most directly connected with the acceleration of GHG emission reductions, which we consider the most effective way of mitigating climate change. However, for the intended policy impacts to materialize, the policy process must first produce adequately designed policies and policy mixes (policy outputs). Policymakers must then put these into practice in a way that changes the behavior of the targeted groups as intended (policy outcomes). The sum of any changes in the behavior of targeted and non-targeted groups constitutes the policy impacts.

To avoid design deficits that result in flaws in bringing about the intended climate policy outcomes and impacts, policymakers should invest sufficient time in producing policies and setting up governmental apparatus that facilitates their implementation [53]. From this perspective, and somewhat counterintuitively, accelerating climate action may necessitate slower policy adoption to ensure sufficient consideration of scientific knowledge and, where available, of evaluation results from previous policies.

Our more general point is that policymaking needs to become more agile if it is to accelerate the reduction of GHG emissions. It is widely accepted that there is an implementation gap between countries’ climate targets and the performance of their climate policy portfolios [54]. However, empirical research on the actual effects of climate policy—encompassing both outcomes and impacts—is in its infancy. Recently, a few studies have started to systematically assess the relationship between the adoption of policies (policy output) and emission reductions (policy impact) [32,36]. A clearer understanding of climate policy’s role in accelerating climate action, as we have aimed to provide here, will facilitate this research agenda. 

We have also stressed that a compelling conceptualization of acceleration needs to recognize that at some point the climate policy effort needs to be maintained and eventually lowered, that is, the policies need to be modified, dismantled or even terminated. Examples of stepping down climate policy effort exist in certain places and sectors (e.g. subsidies on EVs or feed-in-tariffs for renewable electricity). On a more aggregated level, we are nowhere near the point at which a stepping down of climate policy effort is needed, but governments should set up their climate governance systems in ways that could facilitate an initial stepping up in tandem with policy maintenance and later, stepping down.

Overall, we believe that fostering open, transparent discussions on the current stage of the acceleration of climate action, coupled with rigorous evaluations of policy design will substantially improve collective learning and ultimately strengthen the outcomes and impacts of climate policy. Conversely, if “acceleration” remains merely a catch-all phrase encompassing all aspects of climate transitions without precise definition or application, we risk undermining the clarity needed to facilitate the broad learning processes essential for effective climate governance. Achieving conceptual precision will enable more targeted interventions, clearer accountability, and ultimately more effective climate action across different governance levels.

Acknowledgments

The following ACCELZ project members have provided valuable feedback on draft versions of this paper: Solveig Aamodt, Robbie Andrew, Tobias Bach, Bård Lahn, Emiliano Levario, Shivika Mittal, Mette Undheim Sandstad, Ida Sognnæs, and Yves Steinebach. We further acknowledge constructive comments from Laurence Crumbie, Christin Heinz-Fischer, Jessica Jewell, Andrew Jordan, Corinne Le Quéré, Simon Schaub as well as from two anonymous reviewers on previous versions of the manuscript.