Protecting Future Generations Through Minilateralism: Climate Clubs and Normative Legitimacy

Despite three decades of global climate negotiations and high expectations for the 2015 Paris Agreement, global emissions continue to grow. To protect future generations from severe harm, scholars, environmentalists, and politicians alike explore potential supplements to the United Nations Framework Convention on Climate Change process. One potential supplement is climate clubs of a type where a small number of “enthusiastic” countries embark on ambitious mitigation efforts while encouraging other, more “reluctant” countries to join. Previous research has shown that this club type possesses a significant potential for expanding membership and eventually becoming highly effective in reducing global emissions. A common criticism of climate clubs, however, is that they lack legitimacy. Assessing this criticism, we argue that climate clubs of the type considered here can be normatively legitimate. The main challenge for normative legitimacy concerns climate clubs’ use of incentives, particularly negative incentives, to attract members. However, we argue that even negative incentives for participation can be legitimate, assuming they meet a set of relevant legitimacy criteria—including that the club respects human rights, provides a comparative benefit, maintains institutional integrity, implements only proportional incentives, and fulfills a requisite set of epistemic criteria. We also argue that the normative legitimacy of climate clubs’ use of incentives for compliance is less challenging than the normative legitimacy of their use of incentives for participation.


Introduction
Three decades of climate negotiations under the auspices of the all-inclusive United Nations Framework Convention on Climate Change (UNFCCC) have yet to produce effective international climate cooperation.The consequences are devastating-especially for children and future generations.According to Save the Children (n.d.), nearly 160 million children are already suffering from increasingly severe and sustained droughts.Moreover, some 90% of diseases caused by climate change will likely affect children under the age of five.By 2040, one in four children will be living in areas with extreme water shortages.By 2050, the climate crisis could cause an additional 24 million children to be undernourished and the number of migrants to increase by more than 140 million.Thus, climate change is a problem that particularly affects future generations.This is all the more true the longer it takes for the international community to develop an effective response to reduce greenhouse gas emissions.For the sake of the welfare of future generations, we thus need international institutions that last over time, which likely requires both effectiveness and legitimacy (in both a descriptive and a normative sense; see, for instance, Tallberg & Zürn, 2019).
The ineffectiveness of the UNFCCC negotiations, which operate under a consensus rule, has triggered various initiatives to address the climate problem in smaller-and possibly more effective-forums.However, such initiatives have been challenged.First, concerning effectiveness, scholars have deemed existing minilateral associations as little more than "forums for political dialogue" (Weischer et al., 2012) or "discussion clubs" (Andresen, 2014).Second, regarding descriptive (sociological) legitimacy, scholars have found that minilateral associations command (at best) modest support from elites and the public (Falkner et al., 2022;Gampfer, 2016;Zelli & van Asselt, 2013).Finally, concerning normative legitimacy, Eckersley (2012, p. 26) argues that "exclusive minilateralism" is "elitist, procedurally unjust, self-serving and likely to thwart the justice principles of the UNFCCC."Yet, Eckersley (2012, p. 28) acknowledges that if "more expeditious ways" exist for reducing the dangers of climate change, "they ought to be carefully considered in the name of promoting substantive justice." One potentially "more expeditious way" is a climate club of the type depicted by Victor (2011).This type of club would be initiated by a group of enthusiastic countries that would then seek to incentivize other, more reluctant countries to join.While clubs of this type "face the highest hurdles to implementation" (Falkner et al., 2022, p. 480), the longer it takes for the international community to effectively deal with the climate problem, the more relevant and urgent the establishment of potentially effective climate clubs will become.
One important function climate clubs may serve is to prevent "carbon leakage," which may occur when greenhouse gas regulations are not global.Multinational companies' incentives for relocating to areas without greenhouse gas regulations would be weakened by the risk of being sanctioned by an effective climate club.Recent efforts include the G7 climate club as well as the EU-based Carbon Border Adjustment Mechanism, which has been termed a "de facto climate club" (Szulecki et al., 2022).Both efforts bear similarities-yet also differences-to the climate club design we consider here.
If successful, a climate club would create a snowball effect, whereby adding new members would make membership increasingly attractive for remaining outsiders.Should a club succeed in establishing such a process of membership growth, it could end up including states responsible for a large share of global greenhouse gas emissions.Recent research using agent-based simulations shows that a climate club of this type can indeed grow and eventually become very effective in reducing global emissions.To succeed, however, the club must be properly designed and initiated by the "right" constellation of enthusiastic countries (Hovi et al., 2019).
We argue, in response to Eckersley's criticism cited above, that climate clubs of the type considered here can indeed be normatively legitimate, but only if they fulfill certain criteria.The main challenge for their normative legitimacy is not exclusiveness but rather their (potential) use of negative incentives to induce reluctant countries to join.Applying a suitably modified version of Buchanan's (2018) framework for normative legitimacy, we claim that climate clubs can be normatively legitimate if they respect (and generally do not endanger) human rights, provide a substantial benefit (in terms of climate mitigation), maintain an adequate level of institutional integrity, have a reasonable system for internal deliberation and bargaining, and impose only proportional sanctions on nonmembers.If so, they should, in virtue of their legitimacy, be supported or at least not interfered with (Buchanan & Keohane, 2006).Conversely, illegitimate institutions lack this moral right to noninterference and support.In addition, and more generally, we agree with Buchanan (2018) that democracy is legitimacy-engendering, but that this criterion is subject to a feasibility constraint, and that democracy is infeasible in the international realm, at least for the time being.
The existing climate club literature has been concerned with functions, effectiveness, and descriptive (sociological) legitimacy.In contrast, it has paid far less attention to normative legitimacy.This article aims to help fill this void.
The article is organized as follows.In Section 2, we define climate clubs and review major contributions to the climate club literature.In Section 3, we introduce a set of criteria for assessing international institutions' normative legitimacy.In Section 4, we apply this set of criteria to assess the normative legitimacy of climate clubs' use of incentives to expand membership.In Section 5, we briefly explain why the use of incentives to enhance compliance is less problematic for climate clubs' normative legitimacy than the use of incentives to enhance participation.Finally, in Section 6, we conclude by summarizing the criteria climate clubs of the type considered here must fulfil to be normatively legitimate.

Climate Clubs
Climate clubs could come in various forms and shapes.As already mentioned, many clubs are simply "discussion clubs," with no membership fee, no legally binding commitments, and few (if any) incentives for enhancing participation and compliance.In many cases, membership is based on invitations, preexisting membership (e.g., the G7 climate club), or some other mechanism that restricts potential participation.Few existing clubs pursue an open-membership policy, which would seem a precondition for initiating significant membership growth and thus having a substantial impact on global emissions.
Because of the diversity in club design, clubs might vary considerably in terms of normative legitimacy.
Therefore, any assessment of normative legitimacy must focus on a particular type of climate club.This is exactly what we do in this article.Our choice of club type is not random; rather, we focus on a type shown by previous research to have the potential to become very effective in reducing global emissions.Building on Victor (2011) and using agent-based simulations, Hovi et al. (2019) show that to be effective, a climate club must include the possibility of using incentives if it is to recruit reluctant countries to become members and enforce compliance.However, using such incentives (especially to induce other countries to participate) raises normative legitimacy concerns.The purpose of this article is precisely to map the criteria climate clubs need to satisfy to overcome these concerns.
In our view, it is particularly interesting and relevant to consider the legitimacy of a club design with the potential for becoming effective.We therefore base our discussion of normative legitimacy on the design considered by Hovi et al. (2019).We also invoke their definition, according to which a climate club is an "international actor (country) group that (1) starts with fewer members than the UNFCCC has and (2) aims to cooperate on climate change mitigation" (Hovi et al., 2019(Hovi et al., , p. 1072)).Further, the founding members are assumed to be "enthusiastic," that is, willing to undertake emissions reductions beyond what is in their (narrow) self-interest.Finally, their design includes an "open-membership policy," meaning that any country can become a member, provided it is willing to "pay the membership fee," that is, to fulfill the club's membership criteria.For example, in the model used by Saelen (2016) and Hovi et al. (2019), members must commit to spending 1% of their GDP on mitigation yearly.Thus, the climate clubs they-and we-study may be termed semi-exclusive: Any country could in principle join; however, countries that decline to accept the membership fee cannot become members (i.e., they are excluded).Because the type of climate club we consider uses incentives (including negative incentives) to induce membership growth, the criteria for normative legitimacy are likely to be more challenging for this type than for many other types of climate clubs.Thus, if the criteria for normative legitimacy are fulfilled by climate clubs that use incentives, they will also most likely be fulfilled by many other club types, for example, clubs that do not include any incentives for participation or compliance.
Climate clubs can serve at least two important functions (Falkner, 2016): First, while the UNFCCC's consensus rule enables any country to block collective decisions, a climate club could start with limited and enthusiastic membership, thereby circumventing the resistance of unambitious UNFCCC members (cf.Underdal, 1980).Second, and partly related, climate clubs might find it easier to agree on incentives for participation and compliance.Positive incentives might take the form of club goods, conditional commitments, or side payments, whereas fines, penalties, or sanctions are examples of negative incentives (Hovi et al., 2019;Nordhaus, 2015;Saelen, 2016).
In short, climate clubs of the type considered here aim for a snowball effect: A coalition of the willing forms and then invokes incentives to expand and develop without requiring all UNFCCC countries' consent.Could such a coalition satisfy the basic requirements of normative legitimacy?The answer depends not only on what type of club we consider but also on what criteria for normative legitimacy we use.

Legitimacy
"Legitimacy" is used in a variety of ways.We take "legitimacy" to refer to political institutions, as well as to their acts, decisions, rules, regulations, laws, and so on (Peter, 2017).Thus, while climate clubs (or other political institutions) may be legitimate or illegitimate as such, their acts, decisions, rules, regulations, laws, and so on may also be legitimate or illegitimate.
The legitimacy of an institution affects how we ought to relate and respond to it.According to Buchanan and Keohane (2006, p. 407), "We should support or at least refrain from interfering with legitimate institutions."Further, legitimacy implies "the right to rule" (Buchanan & Keohane, 2006, p. 405; see also Buchanan, 2018, Politics and Governance • 2024 • Volume 12 • Article 7674 p. 56; Christiano, 2012, p. 381).We generally agree with both claims.Although "rule" might not be the most compelling term for the ambitions of a small, newly formed climate club, even such a club aims to shape and guide the rules and norms within a certain domain.Hence, the question is whether it has the right to do that.Thus, we understand the term "rule" in a quite flexible manner.
We rely here on Buchanan's (2018) metacoordination view of institutional legitimacy but adjust it in certain aspects.According to this view, an institution is "legitimate if and only if it is morally worthy of our support and an institution is morally worthy of our support only if the benefits of empowering it outweigh the risk of doing so" (Buchanan, 2018, p. 55).In the metacoordination view, institutions are valuable in facilitating the coordination needed to solve problems or provide goods (Buchanan, 2018, p. 56).Further, coordination is needed to ensure that we "converge on some particular institution" (Buchanan, 2018, p. 56).In this sense, the metacoordination view acknowledges sociological or descriptive legitimacy (Peter, 2017;Tallberg & Zürn, 2019;Weber, 1964), but without confusing this kind of legitimacy with normative legitimacy.The latter, as noted, relies on whether an institution is morally worthy of our support.Here, we focus squarely on "normative legitimacy" and leave aside legal, sociological, or descriptive understandings of the term (Buchanan, 2018, p. 58;Buchanan & Keohane, 2006;Peter, 2017).One reason not to take sociological legitimacy into account is that such legitimacy is in principle detached from normative legitimacy.
A normatively legitimate institution may lack sociological legitimacy, and an institution that enjoys sociological legitimacy may lack normative legitimacy (Tallberg & Zürn, 2019, p. 587).This is not to deny that the two forms of legitimacy can be interrelated.Normative legitimacy may sometimes contribute to sociological legitimacy to the extent that the relevant actors are motivated by normative concerns (which of course is oftentimes not the case; Tallberg & Zürn, 2019, p. 587).Furthermore, sociological legitimacy may, as a matter of fact, help institutions in reaching their goals, insofar as support facilitates coordination, and in that sense, sociological legitimacy may strengthen normative legitimacy, albeit indirectly.Since this connection is intrinsically contingent, we do not consider sociological legitimacy to be a (presumptively) necessary criterion for normative legitimacy.Further, whether international legitimacy requires domestic or global democracy is contested (Peter, 2017).Here we assume that international institutions can be legitimate despite the absence of global democracy, and even if the participating collective agents, especially states, are not internally democratic.Although democracy may contribute to an institution's legitimacy, democracy (like sociological legitimacy) should not be a presumptively necessary criterion for normative legitimacy.As Buchanan (2018) notes, democracy is at present not feasible on the international level.
If democracy nevertheless is set as a criterion of legitimacy, the upshot would be that no international institution could be legitimate.This would not be a reductio, because it is entirely possible that no international institution is normatively legitimate.However, on reflection, it seems at least possible that some institutions should be considered legitimate even if their procedures are not democratic and even if those institutions comprise states that are not internally legitimate.
For illustration, consider the following (unlikely) case: Suppose nuclear disarmament is highly beneficial to world peace and security long term.Suppose further that Russia, China, and Iran engaged in a minilateral cooperation of the club type and committed themselves to substantial disarmament, while encouraging other nuclear powers to join, in exchange for, say, preferential trade agreements, and imposed mild sanctions on nonmembers.It would be odd to deem this disarmament club illegitimate, just by virtue of being nondemocratic.While their being nondemocratic might make the member states of this club illegitimate qua states, this does not necessarily stain all their international efforts if these efforts would likely provide great benefits.It would seem wrong not to support this imaginary club and even worse to interfere with its efforts.
To be sure, in this example, the nature and importance of the benefit-nuclear disarmament-is likely to do quite a lot of the normative work.A lack of democracy may be legitimacy-undermining in many circumstances, but not all.As we will return to, combating climate change is of course also a vital (potential) benefit, and in our view, the importance of this benefit makes it hard to conclude that a climate club is illegitimate just because one or more of its members are nondemocratic.This problem does not necessarily arise on the level of the climate club itself because such clubs may have internal deliberative procedures that may be sufficient to secure some sort of democratic legitimacy at the international level.We return to this below.
A further benefit of the metacoordination framework of legitimacy is that it is flexible in the sense that the criteria are applied and interpreted in ways that are sensitive to the function and value of the institution in question (Buchanan, 2018, p. 53).Of the criteria listed below, only the second criterion is strictly necessary.The others will be judged in light of the "noninstitutional alternative" (Buchanan, 2018, p. 60).
The specific criteria against which we assess the legitimacy of climate clubs build on, as noted, Buchanan's work (2018; see also Buchanan & Keohane, 2006), with some crucial adjustments and elaborations: 1. Minimal moral acceptability (most significantly, respect for basic human rights); 2. Provision of a comparative benefit; 3. Institutional integrity; 4. Proportionality; 5. Sound procedures (Buchanan, 2018, pp. 59-60;Buchanan & Keohane, 2006, pp. 432-433).This list differs from Buchanan's in two respects.First, we have dropped a criterion requiring "acceptable origination (or sound pedigree)" (Buchanan, 2018, p. 60).The reason is that it is hard to see that the pedigree of an institution matters to its legitimacy if it currently provides a substantial benefit, is rights-respecting, etc. Suppose a climate club that contributes significantly to mitigation came about historically through force and conquest.The history would be unfortunate, yet arguably not sufficient to give us reason to withdraw our current support.Further, we add the criterion of proportionality, well known from the just-war tradition, and highly relevant to legitimacy assessments-especially regarding negative incentives (e.g., see Reichberg & Syse, 2018).Plausibly, the use of disproportionate incentives would limit a climate club's legitimacy.
We also accept Buchanan and Keohane's (2006, p. 412) claim that legitimacy requires less than justice does.
Climate clubs need not be morally immaculate to be legitimate.We assume, then, that climate clubs are legitimate to the extent that their existence, acts, decisions, policies, and so on meet the above set of legitimacy criteria.To the extent that climate clubs meet these criteria, they have the right to rule, in the relevant sense, even if their acts, decisions, laws, etc., fall short of perfect justice.For nonmembers, the proper response to a legitimate climate club is noninterference and a real openness to cooperation.
Finally, legitimacy may occur at different levels (Peter, 2017), raising the question of whether and how these levels interact.In our context, the most pressing issue is whether a climate club's legitimacy depends on the legitimacy of the individual participating states and if not, whether the way such clubs can permissibly behave towards nonmembers depends on the legitimacy of the members.Although we cannot address this question in full here, we consider it briefly when discussing institutional integrity.

The Normative Legitimacy of Incentives for Participation
To grow and eventually become successful in reducing global emissions, a climate club must entice reluctant countries to become members.Such enticement may consist of positive incentives (club goods, conditional commitments, side payments), negative incentives (penalties, fines, sanctions), or both.
Incentives for participation may be challenging from a legitimacy viewpoint, because they may have negative consequences for those choosing to remain nonmembers.Even positive incentives may entail negative effects for nonmembers.For example, a preferential trade agreement for climate club members might divert trade between members and nonmembers to trade between members.Hence, nonmembers would suffer.
Nevertheless, negative incentives will be particularly problematic.We therefore focus on negative incentives, which typically take the form of imposing costs directly on persistent nonmembers (or threatening to do so).What criteria must be fulfilled if negative incentives for participation in a climate club are to be normatively legitimate?

Basic Human Rights
In averting grave harm to an immense number of people, including both present and future generations, it seems that climate clubs will not pose an immediate threat to human rights.Negative incentives, however, have an obvious material side, imposing costs on target states.Sometimes material costs translate into human costs.Thus, regarding climate change, human suffering that endangers basic human rights is plainly at stake on both sides of the equation (Caney, 2010b, p. 89).Potential basic human rights violations following from imposing penalties or sanctions on nonmembers or from mitigation efforts must be weighed against potential violations of the basic human rights of present and future generations resulting from climate change.
Violating the basic human rights of present individuals is unlikely to be necessary to avert future climate change, however.In cases where imposing costs on a poor nonmember state endangers its inhabitants' human rights, it could be remedied by letting resource-rich club members shoulder more than their (otherwise) fair share of the burden, or by imposing more than what would (otherwise) be a fair share of penalties on capable nonmembers (see Ashford, 2003, p. 286).Doing so would help ensure that climate clubs do not violate basic human rights.
It is not obvious what a "fair share" amounts to in this context, and we consider this issue further in our discussion of epistemic criteria (see Section 4.5).
As noted, all costs imposed on a state might affect individual welfare.But for rich states, such costs need not pose any threat to basic human rights if responsible politicians handle the situation with due care.Thus, whenever it is reasonably possible to prevent sanctions from violating basic human rights, climate clubs must do so to stay legitimate.
While protecting citizens in poor (or spiteful) nonmember states by diverting costs to rich and capable members and nonmembers would be most legitimate, doing so might have other, more remote negative consequences.For example, if such diversion were to undermine rich members' future motivation or bolster rich nonmembers' recalcitrance, these unfortunate effects would have to be factored into the assessment.

Comparative Benefit
To be legitimate, an international institution must provide some kind of benefit "that cannot otherwise be obtained" (Buchanan & Keohane, 2006, p. 422).Thus, an institution can lose its legitimacy if some other feasible institution can provide the same benefit more efficiently, more effectively, or both (Buchanan & Keohane, 2006, p. 422).
To justify the attempt to lay down rules in the relevant domain, the benefit concerned must be of some substance.It must also justify the use of incentives.The benefit of climate clubs is potentially very substantial.The Intergovernmental Panel on Climate Change (2023, p. 89) warns that ambitious action on climate change is now extremely urgent if countries are to achieve the collective temperature goals listed in the 2015 Paris Agreement.Failure to reach these goals would likely entail grave and lasting consequences for current and (especially) future generations.This point is worth emphasizing.Intergenerational conflicts of interest are potentially extremely challenging (Gardiner, 2004).This is problematic not only because future generations are powerless vis-à-vis the current generation, but also because future generations will likely outnumber the current generation on a potentially very grand scale.Thus, the current generation's lack of effective climate action could undermine decent living conditions for very many people in the future.
The stakes are thus very high, and the benefit that can be provided by well-functioning climate clubs is hard to overestimate.
The benefit (i.e., mitigating climate change) must not only be important in absolute terms.To be legitimate, climate clubs must provide the benefit effectively.If the provision is (sufficiently) ineffective, a climate club cannot legitimately impose sanctions on nonmembers in support of its (ineffective) efforts (Buchanan & Keohane, 2006, p. 422).
We must also consider whether the benefit is provided in a comparatively effective and efficient way (Buchanan & Keohane, 2006, p. 422).Suppose two climate clubs are unequally effective or efficient in their mitigation efforts and that the existence of two clubs (for whatever reason) is detrimental to total mitigation efforts.Only the most effective or efficient climate club can then plausibly be legitimate.To conclude otherwise would be to allow unnecessary costs, in terms of reduced goal achievement or increased rights violations, for no good reason.Furthermore, we must compare a climate club's efforts with multilateral efforts, notably those of the UNFCCC.Recent research suggests that different types of dissatisfied states tend to join different kinds of institutions outside the UNFCCC framework.Minilateral forums tend to be initiated by states dissatisfied with the UNFCCC process (Rowan, 2021).States considering that the UNFCCC process moves too fast ("fragmenters") tend to join forums that (a) focus on networking and deliberation and (b) challenge the UNFCCC by articulating rival climate goals and policy principles.In contrast, states considering that the UNFCCC process moves too slowly ("deepeners") tend to join forums aimed at developing and implementing more ambitious climate policies.Interestingly, membership in climate institutions designed to facilitate implementation is positively associated with the ambitiousness of nationally determined contributions.
In contrast, membership in climate institutions in general is unrelated to such ambitiousness.Because enthusiastic countries-the initiators of the climate club type we consider-would almost certainly be "deepeners," a (successful) climate club will arguably more likely strengthen than weaken nationally determined contributions under the Paris Agreement.
Let us nevertheless assume that a climate club really would adversely influence the Paris Agreement's effectiveness.Would such adverse influence constitute a legitimacy problem for the club?The answer, in our view, depends on the two institutions' combined effect on global emissions.To be sure, a climate club could increase, decrease, or leave unchanged the effectiveness and efficiency of the Paris Agreement.To be legitimate, the climate club must enhance the institutions' total effectiveness and efficiency.It must even do so by a substantial margin because a new climate club will likely have less institutional resilience than an established UN institution with universal membership, such as the Paris Agreement.Thus, we must factor in the possibility that a future shock could cause the climate club to falter and leave us with only a weakened Paris Agreement.This being said, the potential conflict between climate clubs and the Paris Agreement should not be exaggerated.Because Paris is largely about collective goals and uses a framework based on self-determined pledges, monitoring, review, and verification, it could have a synergistic rather than an antagonistic relationship with climate clubs.Some readers might object that for tackling climate change, a certain minimum of collaboration within the UNFCCC is necessary and that, therefore, a climate club that reduces the level of cooperation below this minimum cannot be legitimate.However, even if one accepts this objection, the above line of reasoning remains valid unless the level of UNFCCC cooperation drops below the minimum.After all, the main purpose of a climate club is to enhance progress on climate change mitigation.
Although two (or more) institutions pursuing the same goal might both be legitimate, differences could exist concerning the extent to which they are permitted to impose negative incentives.Although we cannot pursue this question in detail here, it is plausible that the entitlement to impose negative incentives would vary with the institutions' effectiveness and efficiency, as well as with the expected long-term consequences.
To enhance efficiency and effectiveness, negative incentives must be clearly specified beforehand to make the requirements for avoiding them unambiguous.For climate clubs, this criterion could be fulfilled by including regulations for the use of incentives in the institutional structure (e.g., in a club's constitutive agreement) and by making these regulations publicly available for all members and nonmembers alike.In addition, all sanctions must be clearly contingent on the offense, so that a nonmember that accepts becoming a member will avoid further punishment (concerning participation).This requirement, too, should be easy to satisfy.Indeed, having induced a particular reluctant country to join, other club members would have no further grounds to punish it (over membership).

Institutional Integrity
The institutional integrity criterion requires the absence of a "pattern of egregious disparity between its actual performance…its self-proclaimed procedures and major goals" (Buchanan & Keohane, 2006, p. 422).In our context, this means that climate clubs must do their fair share to achieve the goals themselves (Buchanan & Keohane, 2006, p. 423).An institution's integrity might also be affected by the integrity of its members.Consider a perfectly legitimate international institution that consists only of illegitimate (say, oppressive and human-rights-violating) states (Buchanan & Keohane, 2006;Rawls, 1999).It seems that even illegitimate members of a climate club might be permitted to impose ambitious climate standards upon other states, at least if these illegitimate states are not too illegitimate and the climate club is otherwise legitimate.Although it is hard to argue conclusively for this claim, one relevant consideration is that the severity of climate change's consequences dwarfs at least minor sources of internal illegitimacy.Suppose that a climate club consisting of minimally illegitimate states were positioned to ensure emissions reductions in other countries to an extent that would make a substantial difference globally.It would be contrived to insist that the minimally illegitimate members were not morally permitted to ensure these emissions reductions, given only insignificant sources of illegitimacy and the substantial dangers of climate change.However, severely illegitimate members would more likely undermine the club's legitimacy (this point is related to our earlier discussion of democracy as a candidate criterion for legitimacy).

Proportionality
That negative incentives can be legitimately imposed in principle does not mean that all kinds and magnitudes of incentives are permissible.Further legitimacy criteria are needed, including proportionality.
Proportionality requires that incentives be proportionate to the offense.Here, the offense concerned is failure to contribute fairly to solving the climate change problem.It is not straightforward to determine precisely what incentives qualify as proportionate to this offense.However, at the very least, the punishment can reasonably equal the difference between the offender's fair share (e.g., the membership fee) and its actual contribution.The combined cost of the actual contribution and the punishment would then equal the (mitigation) cost of membership.This punishment would therefore simply ensure that the offending state must bear a cost equivalent to that of contributing its fair share of climate change mitigation.
The means of punishment are also relevant.Consider trade sanctions, which worsen the offender's position in the trade system.An equally costly incentive could in principle be imposed by seizing a portion of the offender's natural resources or occupying a piece of its (unpopulated) territory.However, imposing the latter types of punishment is much more problematic in terms of legitimacy because they violate the offender's sovereignty and self-determination more directly.While we do not rule out that more direct measures could ever be legitimate, the threshold would then be significantly higher, and less intrusive means must be unavailable.As a rule, sovereignty and self-determination should be respected.Proportionality requires considering not only immediate consequences but also more remote ones.Even if some penalty level is legitimate when considered in isolation, it might also influence the motivation of other actual or would-be offenders.For example, whether an otherwise legitimate penalty makes other offenders even more recalcitrant must be considered.Although such evaluations can be complex, they are nevertheless needed to properly assess proportionality.
Must there be a substantial probability that the target will yield for a punishment to be legitimate?We think not-at least if no sanctions would make it less likely that others would do their fair share (Drezner, 2003).
Penalizing a state, even when clearly in vain, can be legitimate to increase the likelihood of effectively combatting climate change.Such penalizing is equivalent to imposing harm on one agent to realize a good for other agents; however, reducing climate change will likely also benefit the recalcitrant nonmember.

Procedures for Handling Disagreement
Two general epistemic problems are associated with legitimacy assessments of international institutions.First, there is "the problem of factual knowledge" (Buchanan & Keohane, 2006, p. 425).It is hard to establish beyond a reasonable doubt whether a particular institution is legitimate.A lot of detailed information is needed to track the institution's human rights record, its integrity, its effectiveness in providing a common benefit, the proportionality of its incentives, and so on.
The second problem is "the problem of moral disagreement and uncertainty" (Buchanan & Keohane, 2006, p. 425).The primary issue here is not information shortage; rather, it is the deep-seated disagreement about moral questions, including the extent and content of human rights and international institutions' moral responsibility to promote good in the world.The question is how to practically handle such disagreement.
Both problems are important.They are also somewhat intertwined.However, we generally accept Buchanan and Keohane's (2006, pp. 424-433) suggestion that accountability, transparency, and public justification of institutional policies are necessary (though not necessarily sufficient) factors for tackling them.Here, we focus on a narrower question of particular interest in the present context.Climate clubs must not only respect human rights and combat climate change.As highlighted by the large literature on climate justice (see Butt, 2007;Caney, 2005Caney, , 2006Caney, , 2010aCaney, , 2010b;;Gardiner, 2004;Page, 2008Page, , 2012;;Singer, 2004), some conception of fairness in burden-sharing is also necessary (Buchanan & Keohane, 2006, p. 410).This conception must be acceptable and justifiable to current members, ideally also to future members.Above we mentioned the example of a climate club using a "membership fee" of 1% of members' yearly GNP.Even if such a fee is proportional, and in that sense fairer than a flat fee, some countries might still reasonably complain.For instance, they might consider it unfair to require poor states that have emitted relatively little and benefited less than others from carbon-intensive industries to pay proportionally the same as rich and large emitters that have benefited greatly from past emissions.
The general point is that even if a fully just way of distributing climate burdens were to exist, states might well disagree over it.And climate clubs must face this disagreement head-on.A solution could be to include a deliberative procedure for establishing and justifying a cost-distribution scheme acceptable to all members.
As stressed by Rawls (1971) in describing the original position, such procedures cannot be dominated by powerful and wealthy states, because fairness does not track such features, and because these states' bargaining strength cannot be allowed to influence such deliberation.Moreover, the membership fee must be up for periodic revision so that members with (reasonable) complaints against the current fee can present their case in a not-too-distant future, and so that nonmembers will not be discouraged from entering because of disagreement over the current scheme.Finally, the scheme should be justified in an openly accessible way to encourage reason-giving and to provide transparency to all interested parties.Ideally, such justification should refer, when appropriate, to widely accepted (general) fairness principles, but also, when necessary, make plain the compromises involved.
Given the intergenerational aspect of climate change, one might also encourage climate clubs to take future generations explicitly into account in their deliberative procedures.Justice across generations is an extremely complex issue (Caney, 2005;Gardiner, 2004;Parfit, 1984), and it is far from clear what fairness entails in this context.Nevertheless, given what is at stake, it would make sense to include future generations in the deliberations.Inclusion could be done in several different ways, for instance by representing them directly in negotiations (representation in national parliaments has been suggested by Skagen Ekeli, 2009).Note that it would be too demanding to make legitimacy conditional on such representation of future generations.Because climate clubs could potentially benefit future generations immensely even without such representation, it would be highly counterproductive to withdraw support from, or worse, interfere with, an effective and efficient climate club that does not represent future generations in its deliberative procedure.Nonetheless, it is plausible to claim that such representation could contribute to strengthening the legitimacy of climate clubs.
It might appear overly optimistic to expect states to calmly deliberate their way towards a reasonable compromise.However, some response to moral disagreement is necessary, and climate club members might be more motivated than most other states to engage in deliberation about fair burden-sharing to enhance the chance of long-run success.After all, climate clubs (as defined here) comprise enthusiastic states motivated to do more than others to avert great harm.If the members nevertheless prove unable to agree on a burden-sharing scheme, a possible solution might be to combine a minimum threshold for membership with a bottom-up approach to burden-sharing based on nationally determined pledges.Such a system would resemble the one in the Paris Agreement; however, the minimum threshold would ensure that all members uphold a certain level of ambition.
Moral disagreement is, of course, not limited to fair burden-sharing principles.There might be disagreement about the legitimacy criteria themselves, which raises the question of where to draw the line for acceptable moral disagreement.There are no universally accepted answers here (apart perhaps from the claim that basic human rights are nonnegotiable); however, the general idea of requiring some firm ground rules, and then deliberating about the more detailed (yet highly important) points, is presumably familiar to most.
Finally, the deliberative procedure itself rests on fairness principles, the most important of which is that members' material bargaining power should not enter the deliberations.Here, too, one might expect disagreement about what constitutes a fair procedure.This is a general problem, but it is easier, presumably, to agree on a few fundamentals for fair deliberation than on the fairness of their outcomes.

The Normative Legitimacy of Incentives for Compliance
So far, our discussion of climate clubs' normative legitimacy has focused exclusively on their use of incentives to enhance participation.However, to become effective climate clubs might also need to include incentives for compliance.Do incentives for compliance in climate clubs raise normative legitimacy concerns comparable to those raised by incentives for participation?
We think not.The moral legitimacy of incentives for compliance is less challenging than that of incentives for participation because the former types of incentives apply only to members.By joining a climate club, a country effectively consents to the membership terms (even if these terms may change over time).This argument seems particularly compelling for the club's founding members, who design the (original) membership rules, and for other members participating in later deliberative procedures.Assuming that the club's decision-making relies on unanimity or consensus (the most common decision rules in international environmental governance), any founding member must necessarily have consented to-or at least not explicitly opposed-the club's compliance incentives.We are then dealing with a case of what Hardin (1968) refers to as "mutual coercion mutually agreed upon." This argument is somewhat less compelling for countries joining at a later stage because they may have acceded despite being opposed to the club's use of incentives for compliance.However, if compliance incentives are proportional and effective, it is reasonable that they are legitimate for reasons resembling those we discussed regarding incentives for participation.And here too, the presence of a deliberative procedure will further weaken the case for thinking that incentives for compliance are illegitimate.
For both founding members and countries acceding later, one might hold that compliance incentives should not be more intrusive or severe than necessary.Of course, it is hard to imagine states voluntarily consenting to overly harsh incentives, but if they did, moral reasons arguably exist to relax these rules even if initially adopted by mutual consent.

Conclusion
The slow progress of the UNFCCC negotiations constitutes a serious threat to future generations.Scholars, environmentalists, and politicians alike therefore continue to suggest supplements to the UNFCCC.Climate clubs represent one such potential supplement.Research using agent-based simulations has shown that climate clubs might be able to grow and eventually become very effective in reducing global emissions by invoking incentives for participation and compliance (Hovi et al., 2019).This finding raises the question of whether such potentially effective climate clubs can also be normatively legitimate.
We argue that such climate clubs can be legitimate, but only if a set of criteria is satisfied.We find that the main problem for these clubs' legitimacy is not exclusivity, as previous research has suggested (Eckersley, 2012), but rather that they must induce reluctant countries to become members if such clubs are to become effective in reducing global emissions.We contend that even negative incentives for participation can be legitimate, provided the club respects human rights, provides a comparative benefit, maintains institutional integrity, implements only proportional incentives, and fulfills a requisite set of epistemic criteria.These conditions are stringent yet not impossible to fulfill.
We also find that compliance incentives are less challenging than participation incentives because compliance incentives apply only to members.By joining the club, a member country has consented to the club's membership terms.In contrast, nonmembers have not in general consented to the club's use of incentives to enhance participation.
We emphasize that our findings refer to a particular form of climate club-one that is open to all countries willing to pay the membership fee and that uses incentives to enhance membership and compliance.Although our focus on this type of club is far from accidental, future research should consider whether the conditions for normative legitimacy are different for other types of climate clubs.