Show Me the Money: Side‐Payments and the Implementation of International Agreements in Federal Systems

Federal systems face specific challenges in fulfilling their international commitments. In cases of shared jurisdiction, the federal government needs the sub‐federal level to contribute to the implementation process. Both Canada and the EU have used side‐payments to bring and keep on board reluctant and opposing provinces and member states in the implemen‐ tation of international agreements. However, both cases have experienced the limits of this strategy. This article aims to make a theoretical contribution by identifying the causal conditions and processes that help explain the success and failure of using side‐payments to encourage sub‐federal support for the implementation of an international agreement. Based on the study of the implementation of the Paris Agreement in Canada and the EU, I develop a two‐fold argument. First, side‐payments can be an effective tool to persuade sub‐federal governments if they are generally interested in contribut‐ ing to implementation. They do not work for governments of powerful entities that are unwilling to implement. Second, sub‐federal governments react to other actors’ conduct. Side‐payments can keep reluctant governments of weak entities on board only as long as no alliance of powerful sub‐federal entities is formed that resists the implementation of an inter‐ national agreement.


Introduction
Federal and decentralized political systems have generally been considered less capable than unitary and centralized states of fulfilling their international commitments (Jacobson & Brown Weiss, 1995;König & Luetgert, 2009;Levy et al., 1995;Mbaye, 2001;Raustiala & Victor, 1998).Federal systems face specific challenges in fulfilling their international commitments.Especially in cases of shared or sub-federal jurisdiction, the federal government depends on the sub-federal level to contribute to the implementation process (Gordon & Macdonald, 2014;Macdonald, 2014;Paquin, 2010).
As part of the 2015 Paris Agreement, Canada and the EU committed to reduce substantially their greenhouse gas (GHG) emissions.Both parties now face the challenge of keeping sub-federal authorities, their provinces and member states, respectively, on board with implementation.In cases of sub-federal resistance, federal governments need to find ways to ensure lower-level compliance with the Paris Agreement obligations.Forms of resistance include sub-federal refusal to adopt the necessary policies within their own jurisdiction and attempts to obstruct the intergovernmental implementation process or initiatives launched by federal institutions.
To counteract such instances of sub-federal resistance and to keep and bring sub-federal governments on board with implementation, both Canada and the EU have used "side-payments," i.e., instruments to induce actors to take actions that they consider to be a deterioration in the status quo (Cappelletti et al., 2014;Kabir, 2019;Scharpf, 1988).In general terms, the implementation approaches of Canada and the EU allow for differentiated effort, i.e., sub-federal entities that are less capable of climate action have been expected to contribute less to the implementation process than others.In addition, means that provide financial support for climate action measures have been established, such as the EU's Modernisation Fund and Canada's Low Carbon Economy Fund.Lastly, Canada and the EU have used forms of bilateral concessions, including Nova Scotia's exemption from Canada's coal-phaseout plan and additional financial support or special treatment regarding the energy structure of EU member states in Central and Eastern Europe.Sub-federal resistance to implementation and the use of side-payments thus occur in both fully-fledged federations and federalized international organizations.
Despite these multiple attempts to encourage subfederal support, several provincial and member-state governments have continued their resistance, including Alberta under Kenney, New Brunswick under Higgs, and Poland under Morawiecki.This observation suggests that the effectiveness of the side-payment strategies of Canada and the EU is limited.This article asks under what conditions side-payments are successful in keeping or bringing sub-federal governments on board with the implementation of an international agreement.
As side-payments may be particularly necessary in situations where reluctant actors have the right to veto a collective decision or the autonomy to refuse to cooperate or act (Scharpf, 1988;Taylor, 1980), their study has a firm place in international relations and federal studies literature.Scholars have addressed side-payments as a strategy to entice states into international cooperation arrangements and build alliances (Davis, 2008;Kabir, 2019;Poast, 2012;Saelen, 2016), and to promote certain policies in developing countries (Brandi et al., 2022).Others have also studied how side-payments are used to buy domestic support for international agreements that are thought to create intra-state losers (Hays et al., 2005;Mayer, 1992).Similarly, existing literature in the fields of comparative federalism and EU politics has examined how side-payments have been used to persuade sub-federal entities to accept modifications in the division of tasks between the two levels of government (Anand & Green, 2011;Cappelletti et al., 2014), as well as EU policies and decisions towards greater integration (Carrubba, 1997;Moravcsik, 1993;Scharpf, 1988;Taylor, 1980;Thielemann, 2005).Research on international cooperation has found that side-payments are particularly effective in cases of strong asymmetry between the actors involved (Barrett, 2001(Barrett, , 2005;;Fuentes-Albero & Rubio, 2010;Saelen, 2016).With respect to EU integration, it has been argued that only small, weak member states can be bought off (Moravcsik, 1991, pp. 25-26;Moravcsik & Vachudova, 2003, pp. 27-30).
This article contributes to this literature by identifying the conditional configuration under which side-payments are effective in federal systems.I study the implementation of the Paris Agreement to explore the causal conditions and processes that help explain the success and failure of side-payments used to persuade Canadian provinces and EU member states to contribute to the implementation process.Based on this analysis, I develop a dynamic, twofold argument.First, side-payments can be an effective tool to persuade the sub-federal governments if they are generally interested in contributing to implementation.However, they do not work for governments of powerful entities that are unwilling to implement.Second, sub-federal governments react to other actors' conduct.Side-payments can keep reluctant governments of weak entities on board only as long as an alliance of powerful sub-federal entities that resist the implementation of an international agreement has not formed.
In the following sections, I first present my analytical framework before examining the developments on both sides of the Atlantic since the adoption of the Paris Agreement.I then develop a theoretical argument on the causal conditions and processes for side-payments to be effective.In the last section, I summarize my contributions and suggest future avenues of research.

Analytical Approach
I understand side-payments in the broadest sense as instruments to induce actors to take actions that they consider to be a deterioration of the status quo (Cappelletti et al., 2014;Kabir, 2019;Scharpf, 1988).This conceptualization thus entails multiple ways of incentivizing sub-federal governments to contribute to the implementation process, which I categorize into three strategies (Table 1).Federal systems can persuade sub-federal governments to implement by explicitly supporting sub-federal implementation measures, for instance, by providing funding, by offering concessions to sub-federal governments in return for their contribution to implementation, or by making a political trade-off regarding expected contribution to implementation.As the empirical analysis below demonstrates, Canada and the EU have used all three strategies in the context of the implementation of the Paris Agreement.A side-payment strategy is considered effective if it succeeds in keeping or bringing sub-federal governments on board with implementation.
To study the implementation of the Paris Agreement in Canada and the EU, I conduct a structured, focused comparison (George & Bennett, 2005).Combining in-depth analysis with a comparative approach is particularly fruitful in identifying relevant causal conditions.Unlike static comparisons, it is sufficiently sensitive to dynamic processes within the cases.Due to the lack of a comprehensive theoretical framework on the effectiveness of side-payments, I pursue an inductive approach.Literature in the areas of international compliance, comparative federalism, and Canadian and EU politics Federal support of pipeline extension.
Watering down of rule-of-law mechanism.
Burden-reducing measures Instruments that relieve sub-federal governments of burdens, including exemptions from implementation policies or burden-sharing solutions.
Equivalency agreements, exemptions from coal phase-out, burden-sharing approach.
Exemptions from coal subsidies phase-out, free ETS allowances, effort-sharing decision.
provide clues about potential explanatory conditions, which serve to formulate questions to structure the analysis of the two cases (Table 2).I pay particular attention to the sub-federal willingness to implement in terms of policy preferences (Jensen & Spoon, 2011;Treib, 2003) and implementation incapacities and obstacles (Chayes & Chayes, 1998;Chayes et al., 1998).Moreover, besides the power argument introduced before, other research areas have also referred to power as an important condition to understand sub-federal conduct (Anand & Green, 2011;Börzel et al., 2010;Raustiala & Victor, 1998;Watts, 1996, pp. 57-60).
With regard to the implementation of the Paris Agreement, I operationalize these three conditions as the general willingness of sub-federal governments to engage in climate action, domestic implementation obstacles such as the social and economic relevance of hard-to-decarbonize industries or lack of financial capac-ity, and the relative power within the federal system resulting from a sub-federal unit's economic wealth or size in terms of population (see Supplementary File).More specifically, I coded the party platforms of the sub-federal governments in power since the negotiation of the Paris Agreement in terms of their climate action agenda (Figure 1).Concerning implementation obstacles, I take into account the economic relevance of polluting industries in the sub-federal entities and their financial capacity in terms of GDP per capita (Figures 2 and 3).To account for their power position, I created a combined indicator considering the sub-federal entity's size in terms of GDP and population (Figure 4).
Studying Canada and the EU in parallel strengthens the causal inferences we may draw from the empirical analyses.Both federal systems have extensively used side-payments as a strategy to keep and bring sub-federal governments on board with implementation.Also, as the figures indicate, the governments of the Canadian provinces and EU member states differ in the conditions that can be expected to make side-payments necessary and potentially also affect their effectiveness.In more general terms, both Canada and the EU are characterized by a system of intergovernmental relations in which executives are key players in decisionmaking processes, operating based on consensus-based decision-making and the possibility of non-participation and opt-outs (Bakvis & Skogstad, 2020;Fabbrini, 2017;Fossum, 2018).While the EU is not a fully-fledged federation, it can be understood as a federal system (Fossum & Jachtenfuchs, 2017;Kelemen, 2003).In the case of the Paris Agreement, the commitment to reduce GHG emissions was formulated at the EU level-not at the member state level, meaning that the EU as a whole is responsible for effective implementation.Furthermore, focusing on an international climate agreement implies Canada EU  a relevant role for sub-federal governments since climate policy requires action across a wide range of policy areas, including environmental protection, energy, natural resources, transportation, and industrial and economic development.Thus, it represents a policy field in which sub-federal governments cannot be ignored.
While the article focuses on identifying the relevant causal and contextual conditions, and not the causal mechanisms per se, grasping the processes at play is essential to understand the dynamics and interactions between the actors and the conditions.For data collection and analysis, I thus adopt tools from process-tracing methodology (Beach & Pedersen, 2013).When collecting and analyzing my data, I focus on traces, accounts, and sequences of events (Beach & Pedersen, 2013, pp. 99-100), which helps deduce the relevant causal conditions and processes.As this article is interested in effectiveness, sequences are particularly important to  trace effects back to their causes.For example, empirical fingerprints, such as instances where sub-federal governments exhibit support for implementation when offered side-payments or cases where their support diminishes upon their discontinuation, serve as compelling evidence highlighting the significant impact of these side-payments.
In order to ensure the internal validity of the analysis, I rely on triangulation using three different data sources (see Supplementary File).First, I studied official documents, including agreements, communications, conclusions, and communiqués of Canadian and EU intergovernmental meetings and press releases of executives on both sides of the Atlantic.I considered a total of 46 documents.Second, I searched Factiva and Google News for news articles on the implementation processes in Canada and the EU and the conduct of the multiple sub-federal governments.I applied a data saturation strategy (Morse et al., 2002), i.e., I collected articles until I could not find additional information.In total, I studied 510 articles.Third, I conducted eight semi-structured interviews and three background talks with officials from provincial and member state ministries working on climate action, energy, and intergovernmental relations, as well as practitioners from the federal and EU levels.Several interview partners had worked for other sub-federal entities before their current positions, or had experience on both levels of government, i.e., they could provide insight beyond their current jurisdiction.
I study the collected data by focusing on the key decisions and frameworks that have led to particular subfederal resistance and for which Canada and the EU have used side-payments to bring sub-federal governments on board.For the Canadian case, I concentrate on the Pan-Canadian Framework on Clean Growth and Climate Change (PCF) and the adoption of a carbon pricing mechanism.As for the EU, I examine the decisions on the EU's roadmaps for 2030 and 2050.My study thus centers on pan-Canadian and pan-European schemes rather than policies and measures adopted by the sub-federal governments within their jurisdictions.My period of interest ranges from December 2015, i.e., the adoption of the Paris Agreement, to December 2021.

Canada
In order to achieve Canada's climate target effectively, Prime Minister Justin Trudeau initiated a process of intergovernmental cooperation between the federal and provincial governments.At the First Ministers' Meeting in March 2016, federal, provincial, and territorial government heads adopted the Vancouver Declaration (Office of the Prime Minister, 2016).They committed to meeting Canada's GHG mitigation target and agreed to strengthen intergovernmental coordination and cooper-ation in climate action.Based on the Vancouver Declaration, the federal and provincial environment ministers drafted the implementation strategy over the following months in the Canadian Council of Ministers of the Environment.In December 2016, the first ministers adopted the PCF (Government of Canada, 2016), designed as the collective basis for coordinated and effective Canadian climate action.Carbon pricing is a critical element of the PCF.Provinces were asked to introduce either a carbon tax or an emission trading system with a minimum price of 50 CAD/tonne.Alternatively, the federal government would introduce a pan-Canadian carbon price that would cover the provinces that do not have their own pricing mechanism.Furthermore, provinces formulated concrete provincial climate targets in the PCF.
The approach that asked provinces to define their climate targets independently allowed the challenging baselines of the energy-intensive provinces, namely Alberta and Saskatchewan, to be accommodated.Provinces that are able to do more, do more; those that face domestic challenges to implementation do less.This differentiated strategy was widely accepted.Besides signing the PCF, climate-progressive provinces, including British Columbia's Premier Christy Clark, have publicly spoken out in favor of such a differentiated approach.This procedure can be understood as a form of horizontal side-payment among the provinces.
Several provinces, such as British Columbia under John Horgan and Ontario under Kathleen Wynne, did not have to be persuaded.These provincial governments had a clear climate agenda and did not face significant internal implementation obstacles (Figures 1-3) and were, therefore, natural allies in the implementation process (Interviews 2 and 3).While the federal government managed to incorporate most provinces and territories in the pan-Canadian plan, Manitoba and Saskatchewan did not sign the PCF and consequently did not commit to any climate targets.However, Manitoba's Premier Brian Pallister decided to join the PCF in February 2018, leaving Saskatchewan under Premier Scott Moe, the only province outside the framework.
The federal government and parliament adopted several policies to support provincial implementation measures and incentivize the provincial leaders to support the implementation of the Paris Agreement.Funding was especially important.Several provinces had requested financial support to contribute to the Paris Agreement implementation, for instance, to promote renewable energies within their jurisdiction (Interview 7).As one interviewee put it pointedly, "the only way the federal government can compel provinces to do something the federal government wants them to do is to throw money at them" (Interview 5).Accordingly, the federal level created instruments such as the Low Carbon Economy Leadership Fund and the Low Carbon Economy Challenge.However, only provinces signed on to the PCF have access to the Leadership Fund, i.e., Saskatchewan has not been eligible for funding since the beginning.When the Manitoban government decided to join the PCF in 2018, it explicitly stated its wish to access the conditional funding mechanisms as its key motivation for joining the PCF (Government of Manitoba, 2018), indicating the effectiveness of this side-payment tool.
Multiple provinces that face structural challenges to implementation have received compensation from the federal government or have been exempted from federal provisions.For instance, the federal government negotiated equivalency agreements with Alberta, British Columbia, Nova Scotia, and Saskatchewan on exemptions from the federal coal phase-out plan or concerning the release of methane from the oil and gas sector in order to accommodate provincial peculiarities.
The federal government has also used exchanges across policy fields to obtain provincial support.A politically particularly relevant example of such bilateral sidepayments has been the federal support for oil pipelines for Alberta.Notably, the approval of the Trans Mountain Pipeline expansion project in 2016 was a crucial concession by the Trudeau government in return for Rachel Notley's Alberta Climate Action Plan, which included a cap on from the oil sands sector and a carbon price (Interview 8).In 2018, the federal government even acquired the pipeline system to ensure the completion of the expansion and to secure Alberta's support for the federal climate plan (Interviews 7 and 8).
Two specific events challenged the federal government's strategy to keep the provinces on board and the generally broad consensus among the provinces regarding the intergovernmental implementation process.With the election of Doug Ford over Wynne in Ontario in June 2018, Trudeau lost a strong advocate of his climate action and implementation strategy.In addition, a federal court halted the pipeline expansion project in Alberta.The election of Ford and the court ruling led to the governments of Alberta and Ontario deciding to withdraw from the PCF in the summer of 2018 (Interview 3).As a result, the largest province in terms of population and economy, Ontario, and the two main oil-producing and polluting provinces, Alberta and Saskatchewan-taken together responsible for three-quarters of Canada's GHG emissions-were no longer part of the PCF.Alberta, in particular, stated publicly the power position the province holds regarding the implementation process: So today I am announcing that with the Trans Mountain halted, and the work on it halted, until the federal government gets its act together; Alberta is pulling out of the federal climate plan.[...] And let's be clear, without Alberta, that plan isn't worth the paper it's written on.(Notley, 2018, as cited in Tasker, 2018) Notley's statement further indicates that the degree to which provinces have strong leverage in the Paris Agreement implementation context results not only from their size and economic power but also from their contribution to Canada's GHG emissions (Interviews 6 and 8).
With the materialization of this new group of resistance against the intergovernmental implementation plan, Manitoba's government also decided in October 2018 to leave the PCF.The election of Jason Kenney in Alberta in April 2019 further strengthened the group of opposing provinces, which became a veritable block of resistance against Trudeau's Paris Agreement implementation plan.These opposing governments have publicly discredited and attacked the Trudeau government and its climate policies, with Alberta emerging as the leading force of opposition.Open tensions between Alberta and the federal government had already begun at the end of Notley's tenure, despite her general willingness to contribute to implementing the Paris Agreement, and were exacerbated when Kenney came to power.Both premiers distanced themselves from Trudeau and his climate agenda, aware of the federal government's unpopularity in Alberta (Interviews 5 and 6).Besides public criticism and the lack of climate action within their jurisdictions, the "resisting" governments also actively challenged federal implementation measures.The strongest manifestation of this joint resistance occurred when Alberta, Ontario, and Saskatchewan contested the federal Greenhouse Gas Pollution Pricing Act at their respective provincial courts of appeal starting in 2018.
Unwilling governments of large provinces, especially Kenney's in Alberta and Ford's in Ontario, became lost causes for the Paris implementation (Interview 3).As a result, following the government changes, Prime Minister Justin Trudeau halted the multilateral intergovernmental implementation process with provincial premiers and focused on bilateral negotiations to bring reluctant provincial governments on board or to collaborate with willing provincial leaders (Interview 6).
Although generally less aggressive, after the provincial elections in 2018, New Brunswick also joined the resistance block under the new government of Blaine Higgs (Interview 5).Only after the federal elections in the fall of 2019 that confirmed Trudeau's government in power and resulted in a strong result for the Green Party in New Brunswick did the provincial government start distancing itself from the resistance club.Hence, the Higgs government's abandoning its opposition to implementing the Paris Agreement was not a consequence of Canada's side-payment strategy.Rather, strategic considerations regarding elections led the government to become more willing to engage in climate policy.
Table 3 outlines the key implementation decisions, the side-payments instruments, and the moments of sub-federal resistance.Generally, we could observe an emergence and stabilization of the group of resisting provinces, which advanced substantially when the large provinces of Alberta and Ontario joined Saskatchewan in its opposition.Consequently, when Canada decided to

EU
During the implementation process, the European Council, the institution of the EU's heads of state or government, has, in several instances, underlined that the EU and its member states have to develop solidarity mechanisms.Such mechanisms should consider the different starting points of each member state and their capacities to contribute to the EU's overall commitment (for instance, European Council, 2020).
Based on guidelines adopted by the European Council, the European Commission launched a process that has entailed both the definition of climate targets and the adoption of concrete legislation to set the EU on track to fulfill its 2030 climate commitment.Relevant communications of the Commission have been related to the goal of climate neutrality, the European Green Deal, and the increase of the EU's 2030 target from 40% to 55%.Also, regarding legislation, the Commission has proposed the relevant legislative acts, such as the new effortsharing regulation, the Clean Energy for All Europeans package, including the regulation on Governance of the Energy Union and new renewable energy and energy efficiency directives, and more recently, the European Climate Law.
The EU has adopted several measures to implement its Paris Agreement target that consider the different national capacities and provide financial support to regions in need.Member states that are more economically developed and have already moved towards a more climate-friendly economy have been willing to support other member states in transitioning towards a more sustainable economic system.This assistance has been possible because several member states not only follow a climate action agenda but also face little internal structural obstacles to implementation, such as the governments of Xavier Bettel in Luxemburg and Stefan Löfven in Sweden (Figures 1-3).For instance, the new trading period of the EU's Emission Trading Scheme (ETS) includes the establishment of a Modernisation Fund and an Innovation Fund, both of which are financed by the ETS and aim to support the modernization of the energy systems of low-income member states and innovation in the area of low-carbon technologies, respectively.In addition, the Just Transition Mechanism, including the Just Transition Fund, was established to support regions most challenged by a transition to climate neutrality.To benefit from the fund, member states have to develop territorial just transition plans.Furthermore, as part of the new effort-sharing regulation, which addresses the reduction of emissions not covered by the EU-ETS, the member states agreed to mitigate their GHG emissions targets by considering each member state's capacity.This approach can also be understood as a form of side-payment for member states with lower levels of economic development.Such mechanisms have enabled member states that are generally willing but lack financial resources, such as the governments of the three Baltic states or the Portuguese government under António Costa, to contribute to the implementation of the Paris Agreement by helping them to bear the implementation costs.
Poland has been a resistant member state from the beginning of the implementation process.The country saw a change of government right before the Paris Agreement negotiation with the PiS party taking power.On several occasions, the governments under Beata Szydło and Mateusz Morawiecki have attacked the European Commission's implementation strategy and the former Polish government that had agreed to the EU's Paris Agreement target.In the context of the implementation of the Paris Agreement, the public discourse of the Polish government, but also other executives such as Hungary's, has become increasingly politicized.
Regarding multiple EU decisions in the European Council and the Council of the EU, Poland was joined in its opposition by other member states, including the governments of Boyko Borisov in Bulgaria, Andrej Babiš in Czechia, and Viktor Orbán in Hungary.While several member states have regularly attempted to water down specific pieces of legislation, the resistance alliance did not hold regarding the landmark decisions, such as the target for 2030 or climate neutrality.Most opposing member states have tied their support for decisions at the EU level to specific conditions and have asked for financial compensation at every implementation step.Specifically, the creation of the Just Transition Fund was fundamental for Orbán's and Babiš's consent to the 2030 climate targets.As a result, the European Council adopted the new 2030 climate targets and endorsed the Just Transition Fund in its meeting in December 2020 (European Council, 2020).Besides recurring demands for funding, the Polish government has successfully insisted on maintaining the existing free allowances from the ETS and on an exemption clause regarding the phase-out of coal subsidies.These member state governments have also repeatedly urged the European Council to underscore the freedom of member states to determine their energy mix, including the demand to explicitly include nuclear energy as a climate-neutral technology or gas as a transition technology (European Council, 2019, 2020).
The endorsement of the increased 2030 climate target and the Just Transition Fund in December 2020 was part of the adoption of the EU's Multiannual Financial Framework for 2021 to 2027 and the Next Generation EU package (European Council, 2020).The Hungarian and Polish governments had blocked the EU's budget and recovery plan as the use of EU funds was to be conditional upon the respect of the rule of law.The adoption of both financial schemes was of major importance for the implementation of the Paris Agreement, with 30% of the expenditure being dedicated to climate action.The blockage by the Hungarian and Polish governments could be overcome through two concessions that watered down the new rule of law mechanism.The European Council decided that the mechanism cannot be triggered in general breaches of the rule of law, but only when those breaches have an unambiguous and direct negative effect on the EU's financial interests.Moreover, the heads of state or government agreed to delay the mechanism's actual application.These concessions represented relevant side-payments that compelled the Hungarian and Polish governments to consent to the financial frameworks, including funding for climate action.
The accommodation of the multiple demands for funding and exemptions has thus substantially helped to keep or bring member states on board with implementation.In addition, the German government under Angela Merkel played an essential part in the stability of the alliance of resisting member states.While the German government did not become an active opponent of the implementation process, it was a reluctant actor in multiple instances and delayed substantial decisions.For instance, Chancellor Angela Merkel was one of the heads of government who prevented the endorsement of the 2050 climate neutrality objective in the European Council meeting in March 2019.Once Germany had decided to support this target after months of reluctance, smaller member states, such as Bulgaria, Czechia, and Hungary, followed suit and gave their consent at the European Council meeting in December 2019.Only Poland opted out (European Council, 2019).In other words, the combination of side-payments in the form of funding and a German change of heart caused the collapse of the resistance club with regard to the 2050 objective.
Table 4 summarizes the EU implementation process, including implementation measures, side-payments, and instances of member-state opposition.In contrast to the governments of Kenney and Ford in Canada, Szydło and Morawiecki could not establish a strong group of member states to support their opposition.The smaller hesitant member states with low capacity or willingness were brought back on board through financial incentives or gave up their resistance when large member states became advocates for an implementation measure.

From Empirical Insights to Theorization
In both Canada and the EU, several governments, which have shown political commitment to climate action and do not face domestic implementation obstacles, have supported the implementation of the Paris Agreement from the beginning.Examples include British Columbia and Sweden.Such cases did not require that they be incentivized through side-payments to support the implementation process and have contributed to the creation of mechanisms to bring other reluctant governments on board.We have also witnessed on both sides of the Atlantic sub-federal governments that have been hesitant or even actively resistant to support the implementation of the Paris Agreement.This opposition has generally resulted from a sub-federal government's lack of willingness to engage in climate action or implementation obstacles.Implementation obstacles include the lack of financial strength (especially in the Central and Eastern European member states of the EU), the economic or social relevance of industries that are difficult to decarbonize, energy-intensive (such as the coal sector in Poland, the oilsands industry in Alberta, manufactory industries in Germany, or agriculture in the Canadian Prairies), or involve carbon-dependent energy production (as seen in Nova Scotia and Poland).Canada and the EU have launched systems of side-payments to keep or bring on board these reluctant governments that either lack the willingness or capacity to implement.Based on the empirical observations, I propose a two-fold argument regarding the effects of side-payments.The argument is dynamic and configurational as it accounts for how sub-federal actors react to a changing context, such as other actors' behavior, and how explanatory conditions jointly explain the effectiveness of side-payments. Figure 5 illustrates the causal conditions and the process, including their empirical manifestations, that help explain the success and failure of side-payments.First, side-payments appear to work less effectively or not at all for large, powerful sub-federal entities whose governments lack the willingness to contribute to implementing the Paris Agreement (path A in Figure 5).For example, Ontario and Alberta, major economic powers within Canada, could not be persuaded to abandon their resistance to the implementation of the Paris Agreement under the new Ford and Kenney governments, which have no political interest in climate action.The fact that Ontario does not face any relevant domestic implementation obstacles, such as highly polluting economic structures, suggests that capacity issues trigger the launch of side-payment strategies but do not condition the effectiveness of side-payments.  to the implementation of Saskatchewan, a rather small province in terms of economy and population, and its resistance to side-payments demonstrates that having a high GDP per capita can also be a source of power to resist implementation and dismiss financial incentives.
In the EU, Poland, the EU's fifth most populous member state, governed by an unwilling government and facing domestic obstacles to implementation, is a player constantly impeding the implementation process, despite the EU's repeated attempts to bring member states with lower levels of economic development on board.
From the power perspective, it has been argued that governments of powerful entities can bear costs resulting from non-implementation, such as losses in reputation or non-access to financial instruments, and thus resist pressure to implement more easily than weak entities (Börzel et al., 2010).The empirical observations complement this power-based argument by pointing to situations in which powerful governments, such as Alberta, can actually gain reputational benefits from non-implementation.The governments of Alberta have strategically decided to oppose implementation to avoid being sanctioned by their electorate for cooperating with the Trudeau government-one that is unpopular in this province at this time.Such a calculus related to political capital regarding credibility (Bourdieu, 1991;Jentges, 2017) is a privilege for powerful sub-federal governments that can more easily resist social or material pressures from federal institutions or other provincial and national executives, including positive incentives such as side-payments.Similar dynamics could be observed in Poland, where the PiS government not only rejects an ambitious climate policy but also publicly positions itself as unwilling to cooperate with the EU institutions and member states.
In contrast, side-payments can help overcome low implementation capacity and keep or bring governments of both weak and powerful sub-federal entities on board with implementation as long as they are generally willing to act (path B).For instance, side-payments have been an effective instrument for EU member states facing implementation challenges, such as low economic capacities or energy-intensive economies.Examples include the governments of Costa in Portugal and Jüri Ratas in Estonia.Similarly, Nova Scotia's opposition under the government of Stephen McNeil lessened after the federal government exempted the province from the coal phase-out plan.The Notley government in Alberta illustrates that side-payments can also work in cases where powerful entities face implementation obstacles, and the sub-federal government is generally willing to act.Alberta exited the implementation process as soon as the federal government's key side-payment, i.e., the Trans-Mountain Pipeline extension, was under threat of being withheld.This observation points to the importance of side-payments in keeping the Notley government on board, and that the role of sub-federal governments in the implementation process is dynamic and responsive to a changing context.
The second pattern concerns unwilling sub-federal governments of weak entities whose role in the implementation process is subject to a more complex chain of causal conditions and processes (path C).Several provincial and member state governments that have generally shown no interest in climate action have been kept on board and effectively engaged in the implementation process, or, if they deviated from the implementation process, regularly re-engaged.For instance, Croatia under Andrej Plenković or Czechia under Babiš agreed to the increase of the EU's 2030 emission mitigation target in line with the Paris Agreement in 2020 once their condition of financial compensation had been met by the European Council through the creation of the Just Transition Mechanism.However, while side-payments appear necessary to incentivize weak entities' governments that are reluctant to contribute to the implementation process, they do not represent a sufficient explanation as Manitoba under Pallister or New Brunswick under Higgs indicate.Based on the empirical evidence and the sequence of events presented above, the conduct of the powerful entities appears to impact the governments of weak member states and provinces substantially.The effectiveness of side-payments for small, reluctant sub-federal governments broke down as soon as a group of powerful governments resisted implementation (path C1).For instance, Palliser's government in Manitoba followed a back-and-forth strategy regarding its role in the implementation process.But once Alberta and Ontario had withdrawn from the intergovernmental implementation process, Pallister's government also permanently joined the alliance of resisting provinces, i.e., side-payments, especially financial incentives, became ineffective.In the EU, we can also witness how the change of heart of a large member state towards support affected governments of small member states.Shortly after Germany under Merkel decided to no longer block the climate-neutrality objective in the European Council, small member states such as Hungary under Orbán or Bulgaria under Borisov also gave up their opposition and were persuaded by means of financial assistance (path C2).
If costs for implementation are neutralized through side-payments, other costs become important to consider, such as reputation losses.Small provinces or member states might have a harder time bearing these costs than powerful sub-federal entities or resisting pressure from their peers when they act alone.However, once an alliance of powerful entities that oppose implementation is formed, it becomes easier for governments of weak entities that are critical of the international agreement to manifest their opposition openly.They are then shielded by powerful entities, which can absorb much of the reputational damage and resist pressure from other actors.

Conclusion
When are side-payments effective at keeping sub-federal governments on board when it comes to implementing international agreements?The study of the implementation of the Paris Agreement in Canada and the EU has helped to develop a dynamic model that also addresses how the involved actors respond to each other's conduct.The comparative approach has specifically allowed for a better understanding of when sub-federal governments can be brought in through side-payments and has stimulated the development of a two-fold argument.First, if governments of powerful sub-federal entities do not want to contribute to the implementation of an international agreement, side-payments can be expected to have no effect on their opposition.On the other hand, willing sub-federal governments, whether weak or powerful, facing domestic implementation obstacles can be persuaded by means of side-payments.Second, unwilling governments of weak sub-federal entities can only be brought on board as long as there is no alliance of powerful entities resisting the implementation process.
On the one hand, this is good news for implementation.Side-payments can be an effective tool for hesitant sub-federal governments if they are generally willing to contribute to the implementation or are in a weak power situation.This limits the pertinence of the general assumption in the international compliance literature that federalism negatively affects compliance and implementation.Federalism has, for instance, allowed the Canadian government to work effectively on implementation with those sub-federal governments that are willing or that it has persuaded through side-payments.On the other hand, powerful, unwilling governments are "lost causes" that cannot be brought on board.Moreover, the support of small unwilling entities for the implementation process only holds as long as no alliance of powerful resisting governments is formed.Politically, this means that powerful, hesitant governments have a responsibility in that their behavior also affects the behavior of small sub-federal entities, as the effect of the reluctance of the German government to support implementation indicates.
In order to increase both the internal and external validity of this argument, further research is required.As a follow-up to my analysis, a second round of qualitative research should more specifically study the causal mechanisms at play, especially the calculations considering reputation, political capital, and implementation costs.Also, an analysis of additional policy fields would allow for testing the relevance of issue salience as a contextual condition and whether the argument also holds for regulatory agreements.Sub-federal resistance to the implementation of international agreements and the use of side-payments is, in fact, not specific to the Paris Agreement.For example, the Canadian government has responded to provincial opposition, especially from Québec, to the free trade agreements with the EU and the US and Mexico by creating several financial incentives mechanisms, such as the Dairy Processing Investment Fund and Dairy Direct Payment Program, to support the dairy industry against foreign competition (Government of Canada, 2022aCanada, , 2022b)).Conducting a qualitative comparative analysis would provide one possible means to test the theoretical argument proposed here across federal systems and agreements.
The observations made suggest similar dynamics on both sides of the Atlantic regarding the demand for, use of, and effectiveness of side-payments despite the differences between Canada and the EU.Institutionally, the EU, for instance, differs from Canada in the requirement of unanimity in most of its climate-policy-related decisions and in the cooperation between EU institutions and member states during the negotiation of international agreements.In contrast, intergovernmental decisions in Canada are based on voluntary cooperation and Canadian provinces are not involved in international negotiations by default.While the empirical observations made in the scope of this article indicate that these institutional features do not dismiss the validity of the argument developed here, future research should consider how such differences in the Canadian and EU federal models influence the effects of, or-more likelythe size of side-payments.
This article has aimed to contribute to the literature on side-payments specifically but also to the more general bodies of literature on comparative federalism and international compliance.Combining international relations with federal studies, an approach not new to the study of Canada (Simeon, 1972), has proved productive.The dynamic and configurational approach of this article has helped to refine the existing power argument (Börzel et al., 2010;Moravcsik, 1991;Moravcsik & Vachudova, 2003) by identifying the conditions under which powerful sub-federal governments can be persuaded, understanding the impact of powerful governments' behavior, and adding a causal mechanism surrounding political capital.I have also aimed to contribute to the debate on the "comparative turn" in Canadian political science (Turgeon et al., 2014;White et al., 2008) and, more recently, in the field of EU studies.In line with authors who have argued that studying the EU benefits from borrowing approaches and tools from comparative politics (Hix, 1994) and comparative federalism (Fossum & Jachtenfuchs, 2017;Sbragia, 1993), and from comparisons with the Canadian federation in particular (Fossum, 2018), this article provides a concrete example of the value of embedding the EU in comparative studies and abandoning the myth that has dominated EU studies for too long, namely that the EU is a sui generis organization unlike any other.

Figure 1 .
Figure 1.Climate policy preferences of sub-federal governments in Canada and the EU.Notes: The cases are labelled using the names of the heads of government; red = rejection of climate action and green = support of climate action.

Figure 3 .
Figure 3. Financial capacity of sub-federal governments in Canada and the EU.Note: Red = low capacity and green = high capacity.

Figure 4 .
Figure 4. Power of sub-federal governments within Canada and the EU.Note: Red = powerful position and green = weak position.

Figure 5 .
Figure 5. and processes explaining failure and success of side-payments.

Table 1 .
Side-payments in the Paris Agreement implementation in Canada and the EU.

Table 2 .
Guiding questions for structured, focused comparison.
Relevance of polluting industries within sub-federal entities in Canada and the EU.Note: Red = high relevance and green = low relevance.

Table 3 .
Key events of the Canadian implementation process.

Table 4 .
Key events of the implementation process in the EU.
The opposition