Applicability of leadership modes outside the negotiation framework: insights from China

Drawing insights from a qualitative content analysis of China’s national climate reports between 2016 and 2019, this article examines the extent to which classical leadership typologies introduced by Oran Young, Arild Underdal, and Raino Malnes in the early 1990s have explanatory power outside of international climate negotiation frameworks. Mode by mode, we assess the strengths and weaknesses of four classical leadership modes—directional, ideational, instrumental, and structural—to grasp the manifestation of international climate leadership in a domestic context. While the analysis points out some substantial weaknesses in classical leadership modes, it indicates that China has taken consistent efforts to offer climate leadership in a domestic context. Given the huge gap between the leadership literature and the planetary reality; however, the article concludes that the key shortcoming of the leadership literature is that it tends to focus exclusively on the negotiation phase of international climate politics. Therefore, prospective studies on climate leadership have to pay more attention to the locus of leadership.


Introduction
There is a large gap between the states' nationally determined contributions (NDC) to the United Nations Framework Convention on Climate Change (UNFCCC) and the achievement of the Paris Agreement goals. With the current NDCs, the world is very far from meeting the 1.5 °C target and heading instead toward a 2.7 °C temperature rise with catastrophic consequences (UNEP 2021). Undoubtedly, enhanced domestic emission reductions are necessary to prevent dangerous climate change from happening. Acknowledging a significant leadership vacuum in climate politics, a wide range of academic literature on international climate leadership has emerged (for an overview, see Kopra 2020). Yet the bulk of these studies tend to investigate rhetoric and pledges of aspiring climate leaders in international negotiations, and it is not always clear whether they investigate leadership or simply observe the behavior of the most powerful players in those negotiations. As Parker and Karlsson (2014: 13) point out, much less attention has been paid to the "role and importance leadership does or does not play concerning regime/agreement operationalization, implementation, compliance, and overall effectiveness." Acknowledging this key shortcoming of leadership literature, this article studies how international climate leadership manifests outside of international negotiation forums. This understudied dimension of leadership is of utmost importance: without domestic operationalization of leadership, efforts to offer leadership may remain empty rhetoric and fail to materialize in ambitious emission cuts in real life.
Like the majority of leadership studies, we build on classical leadership typologies introduced by Oran Young (1991), Arild Underdal (1994), and Raino Malnes (1995): directional, ideational, instrumental, and structural leadership. Although Underdal noted almost thirty years ago that leadership modes differ in their locus, the bulk of climate leadership literature has applied all four classical modes of leadership as analytical lenses to examine states' (and other agents') efforts to offer leadership in early phases, namely, the agenda-setting, pre-negotiation, and negotiation phases of international climate politics. Against this backdrop, this article examines whether the classical leadership modes also have explanatory power during the latter phases of those processes in general and in domestic implementation in particular. We ask: To what extent are the four classical leadership modes applicable to analyze the ways in which climate leadership manifests outside of international negotiations? While many international and national actors and organizations take part in climate change mitigation actions outside the negotiation framework, we concentrate on state-level climate communications. Empirically, we draw insights from a qualitative content analysis of China's annual national climate reports titled "China's Policies and Actions for Addressing Climate Change" for the post-Paris period of 2016-2019. Given China's authoritarian governance structure, we therefore deliberately focus on climate documents produced by the party-state and do not consider the role of other domestic actors in the implementation of China's NDC. Ultimately, we seek to encourage an academic debate concerning the kind of leadership that states (and other agents) can offer in a domestic context. Clearly, the successful implementation of China's climate policies at a national and local level is of utmost importance for the mitigation of the global climate crisis due to China's large economy, huge population, and status as a rising power. In 2017, China's President Xi Jinping declared that the state would take the driver's seat in international climate politics, this being the first time that China had formally claimed leadership in the climate negotiations (Zhang and Orbie 2021). As a sign of the role shift, in September 2020, the President announced China's enhanced pledge to achieve carbon neutrality by 2060 (Xi 2020); and in 2021, China published an action plan to achieve that target as a part of the party-state's "1 + N policy framework." These mark significant steps for global climate efforts. In this article, however, we focus on the implementation phase of China's climate policy-a key reason why we do not analyze these forward-looking policies. That said, an analysis of China's annual accounts of the implementation of climate policy undoubtedly also hints at what the future implementation of the party-state's climate leadership might look like.
The article proceeds as follows. The next section provides a brief overview of the climate leadership literature. Section 3 introduces our research materials and method, and Section 4 scrutinizes the applicability of leadership modes in an analysis of China's climate reports. We will demonstrate that leadership modes offer a systematic and unbiased lens for examining China's leadership strategies-a finding that confirms our hypothesis that leadership modes have explanatory power outside the framework of international negotiations. Finally, we conclude that the leadership literature tends to take too narrow an approach to leadership by focusing only on the negotiations phase of international climate politics, and therefore, we argue that prospective studies should pay more attention to the various loci of leadership.

Leadership in international climate politics
As mentioned above, most-if not all-studies on leadership build on definitions of leadership offered by Young, Underdal, andMalnes in the early 1990s. Young (1991: 285) defines leadership as the "actions of individuals who endeavor to solve or circumvent the collective action problems that plague the efforts of parties seeking to reap joint gains in processes of institutional bargaining." Underdal (1994: 178), in contrast, offers a more encompassing definition of leadership as "an asymmetrical relationship of influence, where one actor guides or directs the behavior of others toward a certain goal over a certain period of time." As both Young and Underdal focus on the significance of leadership in multilateral negotiations, it is unsurprising that classical leadership typologies identified by Young, Underdal, and Malnes, and leadership studies drawing on those typologies, largely concentrate on states' efforts to pursue leadership in making and shaping international environmental regimes. While the terminology varies among researchers, the modes of leadership refer to directional, ideational, instrumental, and structural leadership.
Directional leadership engages in leading by example. It involves taking unilateral actions and making the first move to promote the feasibility of particular policy solutions to collective problems. Trust-building is an important dimension of this: a forerunner can demonstrate its commitment to a common goal instead of empty words or promises (Underdal 1994: 183;Malnes 1995: 92;Grubb and Gupta 2000: 21;Karlsson 2014: 6, Parker andKarlsson 2018: 522).
Ideational leadership is based on the ability to foster and uphold visions, ideas, and framings of a problem or a policy solution concerning collective problems. Hence, it covers efforts to influence international agenda-setting, to discover joint solutions, and to mold the perspectives of others in a way that serves the leader's needs (Young 1991: 298-302;Karlsson 2014: 7, Parker andKarlsson 2018: 522).
Clearly, these modes are complementary ideal types; and in practice, leaders deploy some kind of combination of them (Young 1991;Underdal 1994). In addition to aspiring leaders' own political, cultural, and institutional backgrounds and perceptions regarding leadership needs, the selection of specific strategies depends on the context, such as the conference agenda, timing, and other participants' positions in the negotiations. What is more, as Underdal (1994: 183) puts it, leadership modes "differ in their locus: leadership through unilateral action is exercised outside the negotiation framework, instrumental leadership within, while coercion can take place within as well as outside the negotiation game." For the time being, however, most empirical studies on climate leadership do not address this factor. Instead, they utilize all four classical modes of leadership to study aspiring leaders' conduct within international regime formation in general and the phases of agenda-setting, pre-negotiations, and negotiations in particular (Parker and Karlsson 2014: 13). This raises a question: To what extent do leadership modes actually differ in their locus or can we assume that all the modes have explanatory power in the analysis of climate leadership outside of the international negotiation framework?
Although leadership can hardly be efficient without credible operationalization (Gupta and Ringius 2001;Parker and Karlsson 2010), there is a gap in the research concerning the domestic performance of would-be leaders and the actual environmental impact of leadership efforts. Some would presumably argue that the underlying cause of this research gap in leadership performance depends on leadership typologies' failure to engage with the domestic operationalization of leadership. As a response, Andersson and Mol (2002: 50) introduce the concept of environmental leadership. According to them, "Environmental leadership refers to both (1) the ambition level of (domestic) environmental policy in varying degrees of legality and (2) the actual (domestic) implementation." We, in contrast, build on a hypothesis that classical typologies have explanatory power outside of international negotiation forums as well (cf. Underdal 1994: 183;Karlsson 2010, 2014). Hence, we assume that the key reason for the research gap concerning the operationalization of leadership is that only a few studies have investigated international climate leadership after the negotiation phase (Parker and Karlsson 2014). Prospective studies should therefore pay more attention to leadership beyond the negotiation framework.
Another open question in the leadership literature concerns the relationship between leaders and followers. Recall Underdal's definition of leadership as a social relation: without followers, there is no leadership (e.g., Sarasini 2009;Parker et al. 2012;Torney 2019;Eckersley 2020). At the same time, some regard self-declaration as a sufficient aspect of a leadership position (e.g., Saul and Seidel 2011;Urpelainen 2011) in general and in the context of unilateral leadership in particular. As Underdal (1994: 183-186) explains: "Leadership by unilateral action is exercised by seeking to solve a collective problem outside of the negotiation framework via one's own efforts and thus by setting an example for others". At first glance, such a definition seems to neatly capture efforts to offer climate leadership in a domestic context. Yet it fails to acknowledge that legitimate operationalization of international climate leadership is closely linked to international negotiations. Even if practical measures to implement climate strategies are decided and carried out at a national level, it is an international context where common goals specifying the ambition level of those actions are debated and agreed on. A legitimate leader builds on these collective goals, values, and interests (Underdal 1994: 179;Malnes 1995: 94)-a dimension that recent literature on climate leadership has addressed (e.g., Eckersley 2020; Kopra 2022).
Empirically, most leadership studies focus on the role of the European Union (EU) and the US in international climate negotiations (e.g., Bäckstrand and Elgström 2013;Bang and Schreurs 2016;Wurzel et al. 2016;Parker and Karlsson 2017;Jänicke and Wurzel 2019). As China undoubtedly also has significant structural power in international climate negotiations (cf. Grubb and Gupta 2000: 19), there is extensive literature on China's contribution to international climate politics (e.g., Gao 2016;Hilton and Kerr 2017;Engels 2018;Kopra 2019). Yet the number of studies utilizing the concept of leadership has grown only recently (e.g., Qi and Dauvergne 2022;Yang 2022;Hurri 2023). In addition, some studies have investigated expectations for and recognition of China's leadership in Page 5 of 16 31 international climate negotiations (e.g., Karlsson et al. 2011;Burzo and Li 2018;Hurri 2020). Apart from Jiang (2022), there are hardly any systematic studies concerning leadership modes adopted by the Chinese government in international climate politics. This gives rise to a question: Does this research gap mean that classical leadership typologies are somehow Eurocentric in nature and therefore not applicable when it comes to studying China's leadership? Or does the research gap simply imply that leadership scholars do not (yet) perceive China as a leader in international climate negotiations? While we leave it to prospective studies to analyze China's leadership strategies in the negotiations under the UNFCCC, we seek to provide insights into these questions by examining the applicability of modes to grasp the manifestation of China's leadership outside of the negotiation framework. To this end, we examine whether China's national reports encompass acts of leadership that do not conform with the classical typologies.

Materials and method
Given our focus on the post-Paris era, we collected annual "China's Policies and Actions for Addressing Climate Change" reports for the 2016-2019 period. This period covers the first years of the post-Paris era, when nations are expected to nationally define their climate contributions. In addition to NDC and the five-year plans, these documents can be viewed as cornerstones of China's climate policy. They not only outline the party-state's efforts to mitigate and adapt to climate change but also integrate those efforts with China's overall policy goals (Kwon andHanlon 2016: 1181). The 2016 and 2017 documents were published by the National Development and Reform Commission (NDRC) and the 2018 and 2019 reports by the Ministry of Ecology and Environment (MEE). This change was caused by China's 2018 institutional reform when the Department of Climate Change was transferred from the NDRC to the established MEE (Zhang et al. 2019: 136). In our material, this institutional reform was identifiable only as a shift in the document producer, and the structure of the reports, for instance, remained the same throughout 2016-2019. We analyzed documents published in English, but they can also be found in Chinese. This bilingual setting hints that the documents are largely targeted at an international audience.
Clearly, the documents represent government-produced narratives of the central government. Therefore, it cannot be presumed that there is no gap between the narratives and the actual implementation of the policies. Despite China being a one-party authoritarian system, implementation is shaped by various actors having potentially competing interests and is influenced by, for instance, the changing role of environmental agencies in China's bureaucratic system (cf. Mertha 2009; Ding 2020). However, the material provides a perspective on implementation outside of the international negotiations framework. While studies examining leadership in international climate negotiations may end up simply observing the behavior of the most powerful players in those negotiations, an important advantage of our choice of material is that it enables us to overcome one of the key shortcomings of studies focusing on leadership modes: the failure to analytically distinguish between domination and leadership (Eckersley 2020).
We analyzed the material with deductive qualitative content analysis, which offers a fruitful lens for reviewing how the existing theory on climate leadership can be validated outside the negotiation framework (e.g., Elo and Kyngäs 2008). According to this approach, we utilized the predetermined categories as a starting point and classified the text into the four leadership modes: directional, ideational, instrumental, and structural. Moreover, we simultaneously examined whether it was possible to identify acts of leadership that could not be included in any of the predetermined categories (cf. Hsieh and Shannon 2005).

Leadership modes under critique: insights from China's climate reports
In this section, we draw insights from China's climate reports to assess the applicability of the leadership modes to grasp climate leadership outside of the international negotiations framework. Mode by mode, we assess the strengths and weaknesses of the classical leadership modes-directional, ideational, instrumental, and structural-in an analysis of how climate leadership manifests in a domestic context. In particular, we demonstrate how modes, when applied uncritically, can be used to showcase leadership even in contexts where it does not exist. In this way, we not only offer empirical knowledge of China's leadership strategies outside of international negotiation frameworks but also demonstrate the merits and demerits of leadership modes as analytical tools.

Directional
As China's annual climate reports can be described as stocktaking the party-state's climate action during the previous year, it is rather unsurprising that the directional mode was in a dominant position in our analysis. The documents encompassed explicit descriptions of successful implementation in various sectors related to climate change adaptation, mitigation, and low-carbon piloting. For example, the Chinese government had "achieved positive results through a series of proactive actions, including adjusting the industrial structure, optimizing the energy structure, conserving energy and improving energy efficiency, controlling GHG emissions from non-energy activities and increasing carbon sinks" (2018). Noteworthy in the analysis was the broad scope of policies and actions that were defined as acts for addressing climate change. For example, adjusting the industrial structure by developing the service industry and nature conservation was framed as climate action-a tendency that highlighted how deeply integrated these documents are with China's fiveyear plans. In light of this finding, the directional mode has explanatory power beyond the negotiations as it explains the state's climate action comprehensively. That said, the coverage of the directional mode can easily be questioned. Due to the complexity of climate change, a major part of unilateral action can be justified as climate action. During the analysis, much of decision-making power is left to the researcher to decide which of the claims can truly be viewed as climate action. In the scope of this research, it is difficult to evaluate the extent to which the prominence of the directional mode depended on our material being limited to national strategies and the extent to which it depended on the high inclusivity of the directional mode.
China's climate reports introduced the party-state's efforts to be a forerunner in climate change mitigation-an important aspect of directional leadership (Parker and Karlsson 2018). In 2016, for example, China's actions were praised for having made remarkable progress, and China played an exemplary role by being the first developing country to submit its NDC to the UNFCCC. In addition, the BASIC (Brazil, South Africa, India, and China) coalition was described as leadership whereby China "joined hands with the large developing countries to make voice and promote the multilateral process" (2018). Although this Page 7 of 16 31 aspect seemed strong in the material, leading by example was justified only through China's own claims, and the material omitted the response of the followers.
The use of the past tense indicated that China had already begun or undertaken all the commitments introduced in the documents-a notion that simplified the recognition of the directional mode. These claims were reinforced by showcasing various targets China had exceeded, often ahead of schedule. Regarding China's NDC for 2020, the target of reducing carbon intensity by 40-45% compared to 2005 was achieved ahead of schedule in 2017, and the goal of increasing forest stock was accomplished in 2016. Achieving one's targets diminishes uncertainty and distinguishes achievements from cheap talk. Hence, these figures can be categorized as China's directional leadership. However, these targets are also set domestically in China, and fulfilling these promises cannot, therefore, be considered to explicitly demonstrate forerunner ship.
Furthermore, the directional mode enables categorizing China's many pilot projects as demonstrations of feasibility, value, and superiority of particular policy solutions (e.g., Parker and Karlsson 2018). These pilots were related, inter alia, to the emission trading system, carbon capture technology, low-carbon cities, and low-carbon vehicles. The pilot programs aimed to "form a series of duplicable and transferrable pilot experiences and play a guiding and demonstrative role" (2018). The documents emphasized continuous efforts to develop and improve these pilot projects to make them even more successful. As mentioned previously, China's case is especially interesting because these local and national actions have such great international significance. While we regarded many examples of China's unilateral action as directional leadership, the modes do not offer tools to assess whether these unilateral acts have any global significance in reality.

Ideational
Ideational leadership appeared rather strong in our material as we found evidence of agenda-setting effort, problem naming and framing, discovery and promotion of policy solutions, and attempts to shape the content of negotiation outcomes (e.g., Parker and Karlsson 2018). In fact, the underlying motive for delivering China's climate reports could be interpreted as efforts to claim ideational leadership; in the material, they are explained as having been prepared "to enable all parties to fully understand China's actions and policies and accomplishments in addressing climate change" (2016). Concerning the problem framing, China's national climate documents defined climate change as a developmental question-a reason why climate policies and actions were integrated into the overall national development objectives. The priority of economic development was emphasized with win-win discourse, which appeared in the material frequently: China was deemed to "treat the tackling of climate change as a significant opportunity to transform the development mode, as well as continue to explore and take the low-carbon development path in line with China's national circumstance" (2016). Furthermore, China introduced a large number of pilot projects developing and promoting Chinese solutions. The ones that we considered to discover globally significant solutions to climate change included, inter alia, developing energy-saving and emission-reducing technologies, natural-gas-fueled vehicles, satellite observation systems, low-carbon cities, carbon capture, utilization, and storage technologies. In addition, China proposed new ideas concerning the Chinese South-South ideology, which China described as an innovative solution that could be advanced in the UN system. Although these practices and policies could be argued to only present ideas and innovations at the national and local level, they can be classified as ideational leadership because they were promoted internationally and, given China's size, emissions, and substantial role in the global economy, many of the solutions would be, at least to some extent, globally significant.
However, the global significance of some of the ideas promoted in the material cannot be evaluated in the scope of this study. Two concepts, in particular, low-carbon development and ecological civilization, were promoted throughout 2016-2019. Although lowcarbon development can be considered a rather universal concept, in the material, it was often presented as a particular model that China seeks to promote both at the national and the international level. Besides low-carbon development, the concept of ecological civilization-a term that was added to the constitution of China's Communist party in 2012-was explicitly framed as an ideal that the Chinese seek to achieve at the domestic level and promote in the international context. Regardless of the active promotion of these concepts, for them to be categorized as successful ideational leadership, they would have to influence the international agenda or the perspectives of other actors.
Before the analysis, we expected that efforts to shape the negotiation outcomes might prove difficult to identify, not least because the material lacked social interaction similar to the negotiations. However, the ideational mode was surprisingly distinctive. In particular, separate sections were dedicated to China's basic standpoint and propositions for each year's UN climate conference. These frequent propositions included a call for finalizing the Paris Agreement rulebook in accordance with the principle of common but differentiated responsibilities, concern about the insufficient pre-2020 climate action by the developed countries, and a demand for the developed countries to honor their commitment to mobilize support for developing countries to enhance the trust among countries in post-2020 climate actions. The demands for developed countries to scale up support intensified, particularly in 2018 and 2019. However, an analysis of China's national reports does not provide tools to assess whether these agenda-setting efforts remained as propositions on paper or whether China was truly able to shape the outcomes of negotiations in practice. This reflects a broader weakness of leadership analyses. It is difficult to identify and assess which aspiring leaders have succeeded in establishing international agendas. For instance, while the EU has been seen to employ ideational leadership by promoting the 2 °C target for the global average temperature rise (Parker and Karlsson 2017: 450), it might be somewhat challenging solely from the perspective of leadership modes to empirically demonstrate it in practice.
To some extent, China's changing framing of climate change since the Copenhagen conference in 2009 (Zhang and Orbie 2021) can be recognized in the material. China increasingly views climate change as "a common challenge facing humanity and requires the joint efforts of each country to deal with" (2016). The discourse change and the growing acceptance of collective responsibility were particularly visible in a comment by the Chinese Premier Li Keqiang during his visit to Germany that "addressing climate change is not only the international responsibility borne by China as a big developing country but also the underlying demand for China to change the development pattern" (2017). However, regardless of these moderate claims to accept responsibility, demands for nationally determined climate policies were pushed throughout the material. For example, China emphasized its differentiated national circumstances as "a developing country with GDP per capita lower than the world average" and, regardless of its position, that it "has actively undertaken international responsibilities in line with its own development stage and national conditions and has made painstaking effort to effectively implement climate change policies and actions" (2019). Noteworthy in recognizing these framings as ideational leadership is the fact that not all proposals or agenda-setting efforts are necessarily in line with the goal of limiting the global average temperature rise below 1.5 °C. Yet, according to the definitions of the ideational mode, they can still be recognized as ideational leadership. Hence, the analysis of the modes is closely attached to the question of how we define climate leadership.
Based on our analysis, it can be claimed that the ideational mode recognizes leadership perhaps even too broadly. Nations can be granted leadership roles based on the promotion of ideas, regardless of how successful they are in pushing these ideas onto the international agenda. The risk of such overinterpretation is especially high in analyses of the manifestation of international leadership in the domestic context because other countries' competing ideas are largely absent. This absence influences other modes as well.

Instrumental
Before the analysis, we expected the instrumental leadership mode to be the weakest in the material because it is thought to be exercised particularly within the negotiation framework (Underdal 1994: 183). Nevertheless, instrumental leadership was distinguished in efforts to highlight China's diplomatic skills in consensus-building and establishing feasible coalitions and alliances at a global level. Particular space was allocated to showcase instrumental leadership as each document included sections on "Active Involvement in International Negotiation" and "Enhanced International Exchanges and Cooperation." Furthermore, China was stated as having the ability to "push mutual understanding among various parties" (2016) and "build a broad consensus on the issue of climate change, making positive contributions to advancing global governance and guiding international cooperation in addressing climate change" (2017). China's frequently communicated central role in reaching consensus exemplified strong claims for instrumental leadership. This raises an important question: Can an actor assess its own success in the negotiations?
China's contribution to the success of the UN climate conference in Paris in 2015 was framed as significant. For example, China was argued to have been "highly praised by the international community for the significant contribution to the conclusion and rapid entry into force of the Paris Agreement and the promotion of international cooperation in addressing climate change" (2017). The cooperation between China and the US before the Paris Conference has been identified as a major explanatory factor for the achievement of the agreement (Dimitrov 2016). Hence, arguments in the material supporting China's instrumental leadership can be considered credible, at least to some extent.
The ability to establish feasible alliances and coalitions to serve one's own needs constituted an important part of the instrumental leadership findings in our analysis (e.g., Grubb and Gupta 2000: 20). Bilateral and multilateral cooperation was described in detail and seemed to intensify and diversify from 2016 to 2019 as more partners and joint declarations were communicated. As an effective climate regime requires acknowledging North-South differences and variations within the North and South (Grubb and Gupta 2000), China's instrumental leadership seemed diverse. China bridged the North-South divide and showed the capability to build alliances with both developing and developed countries. Considering China's rather blurry position between these two groups in climate negotiations (Jinnah 2017), we considered this particularly strong evidence of instrumental leadership. Out of China's coalitions within UN negotiations, particularly the BASIC countries and Like-Minded Developing Countries were mentioned. Within the South-South cooperation, the small island countries, least developed countries, and African states were distinguished. In addition, China's ability to push forward negotiations outside the UNFCCC was claimed through active participation within the G20, the Asia-Pacific Economic Cooperation, the International Civil Aviation Organization, International Maritime Organization, the BRICS (a coalition of Brazil, Russia, India, China, and South Africa), and the United Nations General Assembly. However, it is rather difficult to distinguish between active participation in international organizations and instrumental leadership, particularly in the domestic context. In addition to building these alliances, instrumental leadership requires the ability to successfully advocate solutions in climate negotiations. Thus, mere participation is insufficient for instrumental leadership.

Structural
While the occurrence of structural leadership was low and rather implicit in the material, each document encompassed some sort of structural leadership. Instead of coercion, constructive inducements dominated the structural mode. The South-South incentives explained the structural leadership of China. In particular, fund mobilization, capacitybuilding, and technology transfer were mentioned frequently. Financial and technological aid was considered to "support developing countries' participation in the international climate change negotiation, policy planning and personnel training" (2016). This example illustrates how China frames incentives as measures seeking to encourage particular behavior by other states. In addition to financial incentives, China signed a memorandum of understanding with 28 developing countries (2017), which could be considered a political incentive. Besides the bilateral incentives, China was outlined as active in international organizations by funding the UN to advance South-South cooperation and the Asian Infrastructure Investment Bank to help capacity-building in member countries.
Out of the four modes, we found structural leadership the most difficult to distinguish. Structural leadership requires economic or political resources to be deployed to change actors' preferences or behaviors by enforcing or applying incentives (Parker and Karlsson 2018: 522). Although there were several examples of the utilization of resources in the material, it proved difficult to pinpoint their influence on other actors' behaviors-a finding that implies the structural mode's inapplicability for analyzing the manifestation of international leadership outside the negotiation framework. Furthermore, a structural leader deploys carrots and sticks to influence other actors (Parker and Karlsson 2014: 6). We, however, were able to distinguish only carrots in the material. This is hardly surprising because the underlying motive of domestic reports is presumably to frame China's actions in a positive light. Thus, it is difficult to imagine sticks being as openly represented as carrots in this context. However, this was not the case with sticks that applied to China. For instance, in the 2019 document, the carbon border adjustment tax initiated by some developed countries was condemned for undermining the collective efforts to cope with climate change. Hence, we assume that the domestic focus might miss some features of structural leadership.

Discussion
Our empirical analysis indicated that four classical leadership modes offer a systematic and comprehensive approach to analyzing states' leadership outside of the international negotiation framework. In the case of China's climate reports, all four modes proved to have some explanatory power. Our analysis also demonstrated that leadership modes are rather neutral analytical lenses that are not embedded in European values, ideas, or traditions concerning leadership. We did not identify the type of leadership that could not be categorized in accordance with the classical leadership modes of directional, ideational, instrumental, and structural. This is not to say that leadership analyses would be somewhat free of Eurocentrism: leadership modes can be used to advance biased views against non-Western leadership. For the time being, leadership research often fails to recognize leadership offered by non-Western countries, although the burden-sharing between developed and developing countries is changing into collective responsibility. According to our analysis, China has great potential to become a climate leader should it succeed in the efficient implementation of its 2060 target. Given the Eurocentric postulates of contemporary world politics and speculations about the normative threat to the existing normative order posed by the rise of China and other non-Western powers (Pu 2012;Jinnah 2017), it remains to be seen whether this potential leadership will be recognized by global international society in the future.
Despite their merits, our analysis also indicated that the modes have some substantial weaknesses that erode their explanatory power not only outside of negotiations frameworks but also in other phases of international climate politics. In particular, our empirical case indicates that leadership modes seem to have four principal weaknesses: they take too broad an approach to leadership and fail to analytically evaluate ambition, verification, and followership. Notably, individual modes tend to define leadership very broadly. Therefore, the four modes were not always exclusive in our case. For example, South-South cooperation encompassed elements from more than one mode, and thus, we coded it in three modes simultaneously: structural, ideational, and instrumental. Similarly, pilot projects were often categorized as both ideational and directional because the difference between the modes, in this case, was rather ambiguous. For instance, developing a carbon-trading pilot can simultaneously be considered an effort to discover solutions to a collective problem (ideational) but also a demonstration of the project's feasibility (directional) (e.g., Parker and Karlsson 2018: 522).
While we categorized pilot projects on emissions trading as China's effort to take directional leadership, others have argued that China has only followed the EU's example in developing its emissions trading system (e.g., Torney 2019: 182). This example illustrates the incapability of modes to verify leadership: leadership can be granted to several actors simultaneously. In an international context, the verification tends to be more straightforward, although the typology still lets the researcher decide to whom the leadership is granted in specific cases. In the absence of internationally agreed criteria for leadership, it is consequently very difficult to assess the extent to which some actions can be considered leadership. In national documents as well as in statements made in international negotiations, China, like other states, describes various new policy documents and practical actions to implement its climate policy. Yet it often remains unclear whether such policies have been successfully implemented, and the degree to which they have reduced greenhouse gas emissions (or will reduce them in the future). This is a valid question in China's case, as, in spite of its efficient climate plans and environmental laws, the implementation of these policies has been identified as a particular challenge for the country (Schreurs 2017;Ding 2020).
Therefore, a key shortcoming of leadership typologies is that they do not offer us tools for verifying leadership or assessing the ambition level of leadership (cf. Andersson and Mol 2002;Liefferink and Wurzel 2017). This is particularly problematic for the bottomup approach of the post-Paris era because the ambition level of climate action is nationally determined. The modes enable categorizing unambitious actions as climate leadership even though they would be insufficient for the 1.5 °C goal. This type of leadership can still be significant for the success of climate negotiations as it can accelerate other nations' ambition (Underdal 1994;Parker and Karlsson 2010). To bridge the gap between the Paris Agreement goals and states' NDC targets; however, the most important criteria for climate leadership is arguably linked to the ambition level of climate actions. This raises important questions: Can one be considered a leader based on one's pledges to take actions, or should the outcome of such actions matter in those assessments? To what extent can national climate policies be seen as minimum efforts to act out national climate responsibility to implement the Paris Agreement, and what criteria should those actions fulfill to be seen as efforts to offer leadership? Hence, the crucial shortcoming of leadership modes is that they recognize ambition in relation to other states but fail to prioritize leadership that would bridge the gap between the NDCs and the Paris goals. This is particularly evident in the domestic context; for instance, our material lacked any notion of China's great emissions levels or insufficient contributions toward the Paris goals.
Clearly, our analysis of national climate reports focuses exclusively on the supply of leadership and fails to touch upon the demand side of leadership (cf. Parker et al. 2012: 271). Hence, it cannot help us distinguish "between the desire to take leadership and the ability to take actual leadership" (Gupta and van der Grijp 2000: 70), but offers a rather one-sided picture of leadership. Yet the same shortcoming applies to the broader leadership literature. Despite the growing number of studies investigating perceptions of leadership and factors motivating actors to follow an aspirant leader, there is no consensus on the relationship between leaders and followers in the literature. As our analysis of China's climate reports demonstrated, China has introduced new ideas and solutions to tackle climate change, and from the perspective of leadership typologies, such activities can be described as directional and/or ideational leadership. Yet neither of the modes seems to notice that legitimacy and recognition are important factors in attracting followers.
As rationalist actors tend to present a favorable image in both international negotiations and national documents, there often seems to be a fine line between rhetoric and leadership that cannot be overcome with the help of the modes alone. Undoubtedly, documents produced by China's NDRC and MEE are likely to portray China's climate policies and action in an uncritical light and do not consider the relative weakness of environmental agencies in China's bureaucratic system (e.g., Ding 2020). In particular, the documents do not take into account that China's NDC targets during 2016-2019 were evaluated as "highly insufficient" for achieving the Paris goals (CAT 2021). The period of our analysis excludes China's pledge for carbon neutrality by 2060 announced in 2020, which will most likely change the evaluation of China's contributions in the foreseeable future.

Conclusions
Our empirical insights gleaned from China's climate reports between 2016 and 2019 demonstrate that four classical leadership modes have explanatory power outside of international negotiation frameworks. Since different modes vary in their logic and mechanisms not only in the negotiations (cf. Parker and Karlsson 2014: 8) but also outside of them, a mode-by-mode review is necessary to grasp their explanatory power outside the international negotiation framework. According to Underdal (1994: 183), the directional mode in particular can be expected to occur outside the negotiation framework. Our empirical analysis supports this assumption: the directional and ideational modes were prominent in the findings. The directional mode enables us to explain China's efforts to be a forerunner, showcase how it has exceeded its climate targets ahead of schedule, and explore China's pilot projects as demonstrations of successful policy solutions. Although the directional mode captures unilateral action well, the international significance of these efforts is difficult to evaluate in this locus. One reason for this is the temporal dimension; for the time being, it is impossible to assess the overall significance of China's pilot projects in the 2016-2019 period. This aspect can be evaluated only in the future if and when accurate statistical data are available. In our material, the levels of directional and ideational modes were nearly in balance, and their dominance continued throughout the period under study. In the locus of the implementation phase, the ideational mode offers a useful category for recognizing China's innovations and solutions, but particularly for highlighting China's efforts to place propositions on the agenda of the annual UN climate conference. Our analysis detected numerous efforts to pursue ideational leadership, and in this sense, it had a lot of explanatory power in this locus. Yet the negotiation locus would provide a better picture of how successfully these efforts actually influence the international agenda. Therefore, if we focus only on the locus outside the negotiations, ideational leadership might be identified even too broadly since the concepts and visions promoted in the material will not all succeed in influencing the international agenda. Time will tell whether China will successfully introduce its concepts, such as the Chinese South-South ideology and ecological civilization, to international discourses and agreements.
Furthermore, the instrumental mode can be expected to occur more inside the negotiations than outside (Underdal 1994: 183). In our material, however, instrumental leadership, albeit significantly lower than directional or ideational, was substantially stronger than structural leadership. In the domestic context, the instrumental mode succeeds in explaining the diversity and broadness of China's bilateral and multilateral cooperation and in highlighting China's role in achieving the Paris Agreement. That said, we argue that in this locus, active international participation becomes an easily determining aspect of instrumental leadership, whereas the ability to advocate acceptable solutions remains in a minor role in comparison to the negotiation focus. Hence, the locus of the implementation phase might not cover the whole story of instrumental leadership efforts. Finally, the structural mode appeared weakest in the material regardless of the anticipation that it can occur both within and outside the negotiation framework (Underdal 1994: 183). The structural mode clarifies how China has attempted to encourage particular behavior among its reference groups and helps to categorize constructive inducements, which are constructed in this locus. Notably, our findings on structural leadership omitted any coercive inducements in this locus. However, the domestic context offers one a chance to defend against other states' coercive inducements: for example, in our material, China opposed border adjustment tax plans proposed by other states. Overall, although the modes have explanatory power outside the negotiation frameworks, each of the four modes seems to struggle with the same issue: it is difficult to cover all dimensions of leadership in one locus.
While our analysis pointed out some substantial weaknesses in classical leadership modes, it indicates that China has made consistent efforts to offer climate leadership in the domestic context. In particular, China assumed all four modes of leadership in the 2016-2019 period. Yet these efforts have not (yet) capitalized on international climate negotiations (cf. Eckersley 2020: 16-18). Although some stakeholders have identified China as one of the potential climate leaders (e.g., Karlsson et al. 2011), it is probably fair to say that the (Western) international community has been rather unwilling to recognize China's leadership due to its lack of soft power and the perceived "China threat" among the Western states (cf. Kopra 2019). For the time being, global greenhouse gas emissions continue to increase, and global average temperatures rise-a fact that underlines that there is de facto a leadership vacuum in international climate politics, although the climate leadership literature has identified several aspiring climate leaders.
Acknowledging this huge gap between the leadership literature and the planetary reality, we conclude that the key shortcoming of that literature is that scholars tend to take too narrow an approach to leadership. As Karlsson (2014, 2017) have pointed out, an efficient leader deploys various leadership strategies simultaneously-an aspect that our empirical analysis supports. More importantly, however, our analysis underlines that to be an efficient leader, a state (and other agents) has to exercise leadership in various loci. A genuine climate leader offers leadership and attracts followers in all phases of international climate politics: not only in the agenda-setting and negotiations phase but also in the implementation phase. Arguably, the contemporary leadership vacuum stems from the state of affairs that none of the aspiring leaders has managed to accomplish this: the EU is stuck with its internal challenges, the US has not implemented its international climate leadership at a domestic level, and China has not succeeded in establishing its domestic leadership at an international level (cf. Eckersley 2020). Prospective studies on climate leadership have to pay more attention to the locus of leadership and take a broader approach: an analysis of leadership efforts in only one phase of climate politics will fail to identify genuine leadership that meets the goals of the Paris Agreement.
Code availability Not applicable.
Funding Open Access funding provided by University of Helsinki including Helsinki University Central Hospital. This work was supported by the Tiina and Antti Herlin Foundation (Grant number 20200015) and the Kone Foundation (Grant number 202104011).
Data availability Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

Competing interests
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.