Skip to main content
  • Loading metrics

The role of institutional design in mobilizing climate finance: Empirical evidence from Bangladesh, Brazil, Ethiopia, and Indonesia

  • Rishikesh Ram Bhandary

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing

    Affiliation Boston University Global Development Policy Center, Boston University, Boston, MA, United States of America


International climate finance is a crucial component of the response to climate change. This paper examines how national-level funding vehicles mobilize finance from international sources. Based on interviews with policymakers and various actors involved in the negotiation and design of four major early national climate funds, the Amazon Fund, the Bangladesh Climate Change Resilience Fund, Ethiopia’s Climate Resilient Green Economy Facility, and the Indonesia Climate Change Trust Fund, this paper identifies design features of national climate funds and highlights the trade-offs that developing countries face in their pursuit of climate finance. These design features have significant bearing on the overall effectiveness of the funds themselves. The findings from this study suggest that developing countries seek to maximize control over the funds even though it means that the design features do not minimize costs, as efficiency-oriented perspectives would suggest. The experience of these early national climate funds could be instructive to those governments and stakeholders considering establishing their own national climate funds or improving features. Three policy lessons are noteworthy: the importance of demonstrating commitment to climate policy through transparent data and results, instilling robust fiduciary standards and safeguards, and the virtuous cycle mobilizing climate finance and acquiring a track record on climate programming.

1. Introduction

Many developing countries have sought international assistance to implement climate change-related programs and projects. Estimates suggest that emerging markets and developing economies will need to mobilize an annual $2.4 trillion by 2030 towards climate change goals [1]. The role of national-level institutions will be key in attracting finance from international sources and channeling it towards national priorities. Yet, the role of national-level climate finance vehicles has not received extensive attention [2]. Much of the climate finance scholarship has primarily focused on estimates of climate finance needs, the international architecture of climate finance, and the role of multilateral and bilateral aid agencies [36]. As a number of countries have indicated their interest in establishing national climate funds, examining the experience of the early adopters also provides us with an opportunity to generate policy-relevant insights that can be incorporated into newer funds.

National climate funds are dedicated vehicles designed to access and mobilize financial resources to support the attainment of climate change objectives. These funds allow countries not just to purposely seek international climate finance, but also exercise some control over the manner in which climate finance programming is done. As these funds solicit contributions from donors and international aid agencies, their design and operations involve negotiations with these partners. Therefore, this paper examines how donors and climate finance recipients navigate the choices and trade-offs in their process of establishing a design model that is mutually agreeable. In other words, this paper views the institutional design of national climate funds as an outcome of a bargaining exercise. The paper adopts an inductive approach to understand how institutional design features come about, thereby departing from the rationalist institutional design literature that attempts to explain design features based on functionalist premises. As Swedlund has argued, a narrow focus on effectiveness, at the cost of the larger political dynamics governing aid, would not provide an accurate portrayal of how policymakers actually make the decisions on aid delivery [7].

The findings underscore how developing countries seek to maximize control over the governance of national climate funds rather than optimizing on the absolute value of aid received. Countries that are in a stronger bargaining position are able to institutionalize their preferences more effectively. In instances where countries need to make substantial concessions to the donor governments, host countries seek to minimize the loss of sovereign control. For example, this includes by allowing earmarks even if such features raise the transaction costs. Conversely, countries in weaker bargaining positions tolerate higher levels of transaction costs in exchange for international aid. Taken together, the evidence from the case studies highlights how relative bargaining strength and actor preferences lead to design features that may depart significantly from features that would maximize efficiency.

The primary contributions of this paper are four-fold. First, this paper takes developing country agency seriously and in doing so fills a major gap in the existing literature on the political economy of climate finance [8]. This paper puts the focus on how the bargaining power of host governments gets translated into the design of national institutions for climate finance. In particular, we identify how factors such as having climate policies in implementation increase the leverage of developing countries. Second, this paper bridges the scholarship on development finance and international institutions to better understand the choices facing developing countries in the context of climate finance. Development finance has extensively explored themes such as country ownership, design of pooled funds, and the advantages and disadvantages of various aid delivery channels [7]. It also builds on the literature on the design of international institutions, such as the volumes by Koremenos et al. [2001] and Hawkins et al. [2006], while departing from their strong assumptions about transaction costs and efficiency [912]. In doing so, this paper carries forward the scholarship on regulatory agencies in the global south by Dubash and Morgan [13] on governance and institutional innovation [13].

2. Climate finance and institutional design

2.1 Climate finance and policy literature

This paper investigates the design features of national climate funds. As national-level vehicles designed to attract international climate finance, two major strands of academic scholarship are most relevant. First, there is a burgeoning literature on climate finance and climate finance policy. Second, the literature on development finance, especially focused on governance and institutions, is particularly pertinent.

International climate finance has long been occupied with estimating the scale of climate finance from the global north to the global south and identifying the sources of climate finance. Peterson and Skovgaard have examined how domestic bureaucratic preferences in climate finance supplying countries shapes the actual provision of climate finance to developing countries [14]; which agency leads impacts which recipients are prioritized. Scholars have also sought to answer if climate finance does actually flow to the most vulnerable communities [1517].

While the UN-led process through the UN Framework Convention on Climate Change and associated instruments led to important outcomes and milestones, such as the commitment by developed countries to mobilize $100 billion by 2020, the climate finance landscape remains fragmented with multiple multilateral and bilateral institutions and varying standards and practices. There is strong consensus amongst scholars that the lack of a standard definition of climate finance has impeded transparency and accountability in international climate finance [18]. Skovgaard et al. [19] have documented how the attempts to coordinate climate finance at the national and international levels have thus far been limited because coordination happens amongst those that are like-minded actors. The lack of a strong centralized model for governing climate finance has also opened up the space for transnational climate networks to possibly play a coordination role [20]. Studies have also examined how the divergent preferences of donors and recipients impact climate finance coordination. Funder and Dupuy find that host governments can shape implementation and are better able to tilt practices towards their own preferences which may depart from the original agreement [21]. Why climate finance governance is different from development finance governance is a topic that Browne [22] has analyzed [22]. She identifies three principles that are unique to climate finance: ‘restitution not aid’; allocation by developing countries; and adaptation and mitigation funding. Like Browne who identifies opportunities for redefining principles governing climate finance, Ciplet et al. find that while climate finance often reproduces power asymmetries, from a world systems perspective, climate change also offers an opportunity to rectify it [23].

There is also growing literature on the effectiveness of national-level climate finance policies. For example, Bhandary et al [2021] examine the effectiveness of nine climate finance policies and assess their impact and quality [24]. Within this applied climate finance scholarship, green banks have also received attention. For example, Geddes et al. examine the domestic political processes and negotiations that went into the design of green banks in the UK and Australia [25]. As public finance is often expected to mobilize or catalyze private finance, there has been some work investigating the extent to which multilateral climate finance can indeed bring along private finance. Michaelowa et al. [2021] have examined the role of multilateral climate finance in helping to mobilize private finance in a range of contexts such as Ethiopia, South Africa and Madagascar [26]. Others have sought to analyze how financial tools can lower the risk for the private sector, making climate investments more promising. The role of guarantees is one example [27].

More broadly, the climate finance scholarship has also expanded the treatment of climate finance purely as financial transfers to developing countries to the re-direction of financial flows into climate-positive activities [28]. These studies often examine how macroeconomic and financial policies can help mitigate climate risks. This includes incorporating climate risks into macroeconomic models. Scholars have analyzed how climate finance policies have financial implications by influencing investment decisions. For example, Dunz et al. examine how carbon taxes could lead to a scaling up of green investments or may introduce financial instability [29]. Giglio et al. examine how climate risks are integrated across asset classes [30].

Second, the development finance literature provides a longer arc against which questions surrounding the institutional design of climate finance can be situated. Development finance had come to an understanding about the importance of country ownership in aid programming and the role that global institutions can play in supporting governance reforms in developing countries [31]. Since the national climate funds examined here are designed jointly by host countries and contributors, how governments delegate policy functions to agencies are also important. In the literature on the design of regulatory agencies and central banks, policies are considered more credible when governments delegate policy functions to these autonomous agencies. Therefore, this body of work has focused on measuring the extent of delegation, and the autonomy of the bodies to which the governments delegate [3236]. By voluntarily “tying ones’ hands,” a government is able to display its commitment to a course of action [37 shapes the preferences of donor]. A delegation-centric view of national climate funds would suggest that a national climate fund is credible because a government has invested a fund with policy functions to tackle climate change. In turn, donors would consider a fund to be more credible directly in proportion to the extent to which the host government has ‘tied its hands’ from the fund. To identify the specific design features of this delegation, Koremenos et al. [2001] have articulated a framework that helps to identify how institutional design is function of a series of independent variables such as the nature of the problem, preferences of parties doing the bargaining and more.

2.2 Analytical framework

As developing country governments seek to raise international climate finance, they are confronted with the need to make credible commitments. They need to assure donors that funds will be spent for their intended purposes. National climate funds can serve as commitment devices that lock in policy commitments, reduce fiduciary risks, and increase information about the actions of the government. The institutional design features of national climate funds are markers of how sound the organization or agency really is. Table 1 captures the independent variables examined in this study and the indicators that are used for measurement.

The most relevant actors in the design of national climate funds are national governments (both recipients and donors). This paper examines how their preferences shape institutional design decisions. International organizations, such as the World Bank and UN Development Program, are also relevant as they are potential hosts or trustees of national climate funds and may even lobby to be selected as the trustee. We can divide the actors into two sets: the lead actor from the host developing country, and fund contributors. The lead actor from the host country leads the negotiations with donor counterparts. This actor strives to maximize climate finance contributions, but it also wants to minimize sovereignty costs. In other words, it has to weigh the benefits of obtaining climate finance with the autonomy the country loses over how the money is spent. The exact manner in which these actors resolve this dilemma, or even how strong the dilemma really is, is left open for empirical investigation. The case studies illuminate how developing countries navigate this trade-off. For recipient states, the primary benefit lies in obtaining funding to implement projects, and improving coordination by corralling donors to support the host country’s preferred policies and reducing the fragmented approach to tackling climate change. The lead actor, usually from the environment ministry or executive office (prime minister/president), may encounter resistance from other domestic actors. For example, the preferences of the environment ministry may diverge from the preferences of the finance ministry. For this reason, the status of a country’s policies is important. When a government has already resolved policy clashes between ministries and begun policy implementation, the rivalry between these actors will matter less. When policies are formulated as the fund is designed, we can expect the rivalry to be acute.

For fund contributors, a national climate fund is one channel of several that they can use to supply climate finance. For example, donors can also route aid via organizations like the World Bank or the Green Climate Fund or their own bilateral aid agencies. As a result, the use of a national climate fund as a vehicle has to be compelling. This study suggests that two types of costs can deter donors from engaging with the national climate funds: loss of donor visibility, and information acquisition. First, donors balance the loss of their own visibility when donors pool funds (losing the ‘plant your flag’ effect) with the value gained in supporting the national climate fund. Donors will also have reduced say over how their contributions are spent when they intermingle their contributions (unless they are earmarked). As a result, without a value proposition on how the national climate fund is different from what donors are already doing, contributions to a national climate fund can be a tough sell. Second, gathering information is expensive. Evidence from the case studies indicates that that unless there is a premium attached to coordination processes, donors and government officials are not naturally inclined to coordinate. Furthermore, coordination can also be challenging because it requires staff to acquire a substantial amount of information about the policies of other donors. High staff turnover within donor missions given the short duration of diplomatic assignments, can mean that significant knowledge depreciation occurs. Thus, donors are likely to coordinate only when there is a pressing need to do so. Such situations persist despite the potential for national climate funds to reduce information-gathering costs for everyone. These costs may be counteracted if a national climate fund offers expertise that other similar vehicles do not. Yet, as national climate funds have so far been newly created at the time of mobilization, the expertise of the fund in climate change programming has not yet been a selling point.

Finance-providing countries also want to see their committed funds get disbursed. When donor officials are able to get money out the door, they are viewed to be effective. Similarly, if committed funds are not spent, finance ministries may not be convinced of the need for subsequent rounds of funding. Donor countries can also consider international norms on development assistance, for example, providing development assistance to meet the target of 0.7% of gross national income (GNI). For climate change, there is also a quantitative target of mobilizing public and private climate finance by $100 billion by 2020 (with the interim target of $30 billion in fast-start funds by 2012). For all of these reasons, there is a “disbursement imperative” that shapes the preferences of donor officials [38, 39].

Non-governmental stakeholders are involved with national climate funds in at least two ways. First, they help to provide expertise and technical support to governments as countries design and implement projects through these funds. Second, NGOs can serve as implementing entities for the national climate funds. A number of national climate funds have dedicated NGO windows. In some funds, they may also be included in the governance arrangements.

The situational characteristics considered here involve the salience of the country’s profile in terms of climate change and the general policy context. By presenting their vulnerability to climate impacts or their ability to reduce greenhouse gas emissions, developing countries may be able to gain international interest. Scholars, such as Narlikar, have argued that developing countries have exercised agency by deploying persuasive arguments about their “powerlessness” and impoverished states [40, 41]. Similarly, in the context of peace and security, Fisher has argued that states like Uganda have highlighted their “fragility” as a way to garner international support and increase their leverage with donors [42]. Building on these insights, this paper takes a closer look at the exercise of agency by developing countries and how they are shaped and constrained by domestic political contexts.

The status of climate policies in the country and the ability to monitor implementation are also important considerations. By having climate policies under implementation, a government is not only more transparent about its intentions on addressing climate change but it also suggests that the country has also undergone the necessary homework to move to implementation. Likewise, the availability of data that allows results to be tracked inspires more confidence in how implementation is progressing than in contexts where such data is not available and the value addition of the fund is more difficult to ascertain.

3. Methodology

This paper examines the design of Amazon Fund (Brazil), Bangladesh Climate Change Resilience Fund (Bangladesh), Climate Resilient Green Economy Facility (Ethiopia) and Indonesia Climate Change Trust Fund (Indonesia). In order to understand how policymakers designed the funds, interviews with those directly involved in the fund design process was vital. These elite interviews allowed me to “acquire information and context that only that person can provide about some event or process” [43]. Therefore, interviews form the primary empirical base of evidence and helped to conduct event history analyses. Reconstructing the decisions taken and timelines of events helped to identify the processes at work.

To collect data, fieldwork was carried out in Bangladesh, Brazil, Ethiopia and Indonesia. We interviewed policymakers and stakeholders associated with the national climate funds. Through snowballing, we identified officials that were involved in the negotiations on the design of the funds. The interviews were semi-structured and lasted between 40 minutes to 90 minutes. The interviews were conducted in English. To encourage the respondents to speak freely, identifying details have been withheld. A summary of the number of interviews by country and role are provided in Table 2. The research protocol was approved by the Tufts Institutional Review Board. Oral consent was obtained from the interviewees in line with the protocol. Apart from interviews, primary documents such as fund design documents, annual reports, operations manuals, national policies, aid evaluation reports, and reports of donors to their respective parliaments provided additional information. I have tried to heed the advice of Bennett and Checkel “to be relentless in gathering diverse and relevant evidence” [44]. The case studies provide “a plausible interpretation to pull all threads together” while acknowledging the “noise” in the data as semi-structured interviews yield high volume of information which will not have the coherence of structured surveys [45].

To ensure that the evidence elicited through the interviews was reliable, I employed three strategies. First, learning the contextual background was vital. As Hochschild has advised, the interviewer needs to know “as much as possible about the context, stance and past behavior of the interview subject” [43]. Having strong command over the contextual information is a key part of building trust with the interviewees as well [43, 45]. Second, I compared information elicited from the interviews against what other elite interviewees said and likewise compared interviewee recollections with other sources of information such as public reports, newspaper article, and so on. This process of triangulation meant that in some cases I interviewed the same person more than once. Third, to further account for the possibility of interviewees offering recollections that were not genuine, I paid attention to the potential audience costs of their statements. If the audience cost is high, it is likely that the reported beliefs are more accurate.

This paper analyzes four national climate funds as case studies. An interview-based, qualitative approach with the key actors involved in the fund design was deemed to be the most appropriate research methodology given the motivating research question. To understand actor preferences and to trace the process on how the design features came about, semi-structured interviews were carried out with the key officials and stakeholders.

The four cases selected are: Amazon Fund, the Bangladesh Climate Change Resilience Fund, the Climate Resilient Green Economy Facility of Ethiopia, and the Indonesia Climate Change Trust Fund. The Amazon Fund was established as a partnership between Brazil and Norway in 2008. Norway committed to contribute $1 billion in results-based financing as an incentive to reduce deforestation in the Amazon. The Amazon Fund was established as the financing instrument invest the financial receipts from this partnership. The two sides agreed to make the Brazilian national development bank–BNDES–the holder of the fund.

The Bangladesh Climate Change Resilience Fund was established as a multi-donor trust fund to support the implementation of the Bangladesh Climate Change Strategy and Action Plan. Given the high vulnerability of Bangladesh to the impacts of climate change, the focus of this fund has been on climate adaptation and resilience. The fund was established as a pooled fund that received contributions from the various funders. The governing board, supported by a technical committee, was ultimately responsible for allocations. The Ethiopian Climate Resilient Green Economy Facility was established to support the government’s Climate Resilient Green Economy vision and strategy. This Facility was set up to be jointly hosted by the Ministry of Finance and Economic Development and UNDP through the Multi-Partner Trust Fund office. Likewise, in 2009, the Indonesia Climate Change Trust Fund was established under the Ministry of Planning (BAPPENAS) to support the implementation of Indonesia’s climate change plans and policies (Rencana Aksi Daerah Penurunan Emisi Gas Rumah Kaca–RAN/GRK). The ICCTF created was UNDP support and its management was fully handed over to BAPPENAS with Bank Mandiri serving as the fiduciary agent.

This set of national climate funds, while not representative of the larger universe of national climate funds, was selected for two primary reasons. First, these funds were a part of the first wave of national climate funds and countries often looked to these funds to design their own financing instruments. Therefore, studying these funds is important from a policy experimentation perspective. Second, since these funds have been in operation for at least ten years, there is a stronger base of evidence to examine alongside the possibility of tracking the evolution of these funds over time.

4. Case illustrations: Design features of national climate funds

4.1 Design features of the national climate funds

Drawing from the experiences of Bangladesh, Brazil, Ethiopia, and Indonesia, the following section discusses how governments and their donor counterparts, along with other actors, made decisions on the four design features. Across the four cases, the following institutional design features have been the primary foci for discussion and debate: the trustee, the governing board, financial contributions, and the fund’s scope. The claim is not that policymakers negotiate these institutional design features independently. Instead, distinguishing design features allows us to disentangle an otherwise highly complex decision-making process. These design features are outcomes of the negotiations between host countries and potential fund contributors. Institutional design features vary from fund to fund because the preferences of the actors involved in the fund design vary, and the contextual factors, both domestic and international, also differ.

4.1.1 Selecting the trustee.

One of the most contentious issues in the design process is the selection of the fund’s trustee. Both the host country and the fund contributors are invested in the selection of the trustee because they do not want to lose control over the financial resources. While recipient countries usually prefer domestic trustees, such trustees may not always be available. Financial institutions that have international recognized financial and fiduciary standards, especially with a track record of working on climate change, are challenging to find domestically. Fund contributors worry about host governments not being able to meet fiduciary standards and safeguards. Furthermore, donor countries may be more disposed to international organizations over host country trustees as donors international organizations may be perceived as being more reliable, although little empirical evidence was found for this theory in my research. Negotiating parties have two options: selecting an existing entity to house the fund (either domestic or international) or creating a new domestic fund. When selecting an existing entity, countries can choose from international organizations, such as the World Bank or UNDP, or national ones. Countries can also compromise by awarding interim trusteeship to an international organization until a national trustee becomes ready. Table 3 lists the trustees of the four trusts examined.

A domestic financial institution with sound governance can help to bolster the case for selecting a domestic trustee. The availability of an internationally recognized domestic agent helped Brazil make a case for a domestic agent to hold the fund. As the Brazilian National Development Bank (BNDES) is a major development bank with sound fiduciary standards, convincing Norway to accept BNDES as the trustee was not difficult. Bangladesh established its national trust fund (Bangladesh Climate Change Trust Fund, BCCTF), as a way to pre-empt the negotiations. Donors were wary of vesting their contributions to a purely domestically-managed fund given their concerns about fiduciary risks, however. Donor countries eventually agreed to ask the World Bank to manage the multi-donor Bangladesh Climate Change Resilience Fund. Across the cases, the presence of fiduciary risks substantially eroded the ability of developing countries to argue for a domestic trustee. Similarly, policymakers in Ethiopia decided to split the difference and established the CRGE Facility with two windows–a UNDP-managed fund and a government-managed account.

When there is no available trustee that countries can agree on, governments negotiated stopgap measures. For example, as a compromise with the Bangladeshi government, fund contributors agreed to limit the World Bank’s duration as trustee by inserting a ’sunset clause’ regarding its involvement. During this interim phase, the World Bank would help to bolster the fund management capabilities of the government to enable an eventual handover. In Indonesia, UNDP had a more active role in the initial stage of the ICCTF before the responsibility was passed over to BAPPENAS and Bank Mandiri. As the government-managed window of the CRGE Facility has received most of the funds, as opposed to the UNDP-managed one, the question of how to transition fund management has not been a matter requiring political attention.

Channeling funds via international organizations is not without cost to fund contributors. First, they lose the visibility of their contributions when the international organization’s brand outshines that of the fund contributors. Second, they are subject to the expertise and programming modalities of the international organization in question. As a result, when fiduciary standards can be maintained, donors may favorably weigh channeling contributions directly to the national fund, as Ethiopia’s experience suggests. Third, the novelty of climate programming creates space that host countries can leverage. Policymakers generally accepted that they were making decisions based on limited information on the efficacy of these funds. Therefore, rather than a focus on the expertise the trustee could offer, there was an overwhelming consideration of fiduciary risks as the experiences of Bangladesh and Indonesia illustrate.

The skepticism of developing countries about handing over fund management to an international organization often arose from a fear of losing easy access to the fund because the government would be subject to the standards and practices of the international organizations. An Indonesian government official captured this sentiment saying, “If we had given it to the World Bank, it would not be our money anymore” (Interview I19, 2018). Given their long history of working with the World Bank and similar multilateral institutions, developing countries were looking for a break from business as usual when they established national climate funds. Furthermore, as the funds are a component of a broader set of policies on climate change, policymakers viewed domestic management of the fund as a vote of confidence from the international community.

The dichotomous debate over a domestic or international trustee, however, oversimplified the finer details about how the institutional arrangements, such as the governing board, would be set up and the financial instruments available for the fund. When host countries have managed to convince contributors to select a domestic trustee, under medium to high governance risk situations, fund contributors have used tools such as earmarking to mitigate those risks.

4.1.2 Governing board.

As the governing board has overall responsibility for the fund, the design of the governing board is of great interest to designers of the national climate fund. The governing board structures the relationship among the various actors associated with the fund. For example, funds differ by the extent to which the governing board retains responsibility or delegates functions to the trustee. Koremenos et al. [2001] refer to this variable as “control.” Apart from the balance of responsibilities between the board and the trustee, a key question that fund designers have to consider is the board’s composition. Countries often debate over the balance of seats between host country representatives and donor countries. Some actors prefer to include non-governmental organizations on the board while others do not. Similarly, the governing board also collects information on the performance of the fund and keeps track of progress.

The relationship between the governing board and the trustee sheds light on the extent to which the host government and the contributing countries have delegated authority to the fund. Table 4 identifies the governing structures broken down by the trustee, the governing body and the secretariat for the four funds. The Brazilian government and the fund contributors have delegated many functions to BNDES. For example, the governing board of the Amazon Fund has overall authority and sets the fund’s policy direction while BNDES staff who manage the Amazon Fund have the authority to allocate funds. Governments vested less authority to the World Bank as the trustee of the Bangladesh Climate Change Resilience Fund. The management committee and the technical committee retained powers to allocate funds. While the division of responsibilities proved to be fragile, it serves to highlight the variation in the level of delegation.

In designing the governing board, host governments face a significant trade-off. On the one hand, the government can concede some autonomy and invite donors to join the board. Such close engagement with donors may encourage them to contribute funds directly into a common pot where contributions are intermingled for projects (as opposed to each donor contribution generating stand-alone projects). On the other hand, the country can maintain a tight grip over the fund with a board comprised of recipient government representatives alone. Such a model, however, is not attractive to contributors who want some say in how the contributions will be programmed. A compromise solution has been to use earmarks. For example, in Indonesia’s new general-purpose environment fund (the BLU), REDD+ is only one part of the fund’s scope. As a result, the government did not wish to open up the board to donors. Rather, Indonesia had compartmentalized Norway’s involvement to the management of the Norwegian contributions alone. Furthermore, as Norway wanted to see stakeholders such as indigenous peoples and local communities included in the fund’s management, the government agreed to include these stakeholders for the earmark specific to Norway’s alone.

In addition to the governing board, management or technical committees may also play a vital role in the fund’s operations. For example, in Bangladesh, while donors had only one representative on the fund’s supreme governing body, the technical committee, which was open to all of the fund contributors, played a significant role in shaping policies of the fund. Apart from these visible and formal means of governance, the nature of financial commitments also sheds light on the opportunities and constraints the fund faces.

4.1.3 Financial commitments and modalities.

The conventional approach to providing climate finance has been in the form of grants and/or loans as inputs for interventions. Under an input-based financing model, funders release finance when the national climate fund can show that it has reached specified milestones. There is a built-in assumption that meeting the milestones will mean attaining results. This model allows for actors to have greater input in how climate interventions are designed by the fund. It also allows for and requires intensive monitoring of the numerous steps in project design. A second modality of disbursing finance is results-based financing. Results-based funds provide fund contributors with the assurance that they are providing funds only for the results attained. Output-based funds trigger a payment claim only when results are achieved. It spares fund contributors the need to keep a close eye on how funds are being used. The advantage for recipient states is that they have leeway in designing the most well suited interventions to achieve the results. Freeing up policy choices for developing countries has been a major break from past practices that tried to limit attention towards inputs (for example, structural adjustment programs that were heavily input based).

Results-based funds, such as the Amazon Fund, claim resources upon verified performance data, such as avoided deforestation or reduced emissions. For example, when the government can show that it has reduced deforestation levels below specified thresholds, payments are unlocked from the contributor, Norway. Such a model is attractive for Norway as it minimizes risks that its finances are not obtaining the results that it would like to see. For the host countries, the output-based model is also attractive because the government retains control over the decision-making about how to reduce deforestation and associated GHG emissions. In other words, Norway does not get involved in the domestic policymaking process. The experience of Brazil suggests that as long as the fundamental elements of a results-based payments program are in place (an acceptable trustee to manage the funds, and reliable indicators to monitor results), implementation can move forward. In fact, the Amazon Fund has used a significant portion of its money to improve monitoring and measurement techniques in the Amazon, including the coupling of land-use change data with property demarcation information.

For input-based funds such as the BCCRF, donors have greater uncertainty about how their contributions are utilized to support desired outcomes. Given such risks, fund contributors have taken a more engaged approach to minimize performance risk, primarily through two means: earmarks and tailored reporting. While the BCCRF did not have explicit earmarks and allowed the World Bank and the government the latitude in project design, the CRGE Facility, by comparison, has had a lot less freedom in project design. DFID, one of the leading fund contributors, sought to minimize fiduciary risk by directing funds towards projects that had already been vetted by DFID. Similarly, to increase and improve the information available to contributors, both the CRGE Facility and the Indonesia Climate Change Trust Fund provide maximum flexibility to the donors to set up parallel reporting structures. Under such a setup, donors benefit from receiving the kind of information they would like from the fund while the host country receives funds to implement programs.

Across the four national climate funds, two attributes of the financial instruments they deploy are worth highlighting: the time horizons of the projects, and grant-based finance. Project time horizons have a significant bearing on what national climate funds can deliver. Across all of the cases, fund managers were under pressure to design projects with short time horizons that were also “transformational” in nature. The emphasis on short time horizons to demonstrate "transformational" impact may, in turn, inhibit the potential for national climate funds to deliver on their value.

National climate funds have primarily limited their instruments to grants, even though some have blended their grant resources with larger loan packages. As the targets of the national climate funds were primarily sectoral ministries, the use of grants is not surprising. Furthermore, as these national climate funds were designed when there was a push to get cross-sectoral buy-in to the climate change agenda, such a focus was indeed warranted. The legal form of a fund has a direct bearing on the diversity of financial instruments that it can utilize. Offering instruments such as loans requires higher risk appetites for both host governments and fund contributors, and government housed funds such as the CRGE Facility or the ICCTF have explicit no-loss provisions.

4.1.4 Scope.

The scope of the fund is the domain of the fund’s operation (geographic or policy domains). Some funds can have a tightly defined scope. For example, the Amazon Fund’s scope was restricted to the Amazon biome alone. Other funds can be more comprehensive in their scope. The Ethiopian Climate Resilient Green Economy Facility is economy-wide in its ambition and includes all aspects of climate change.

The scope of the fund is likely to be broad when the actors involved have diverse preferences. This finding is consonant with studies of institutional design such as Koremenos et al. 2001. The fund’s scope suggests an interesting trade-off. For example, in the case of the Amazon Fund, the government faced a tension between obtaining the UK as an additional funder in exchange for expanding the Amazon Fund’s scope to include the Cerrado region, which the government was hesitant to do. The original design of the Amazon Fund had not included the Cerrado region but only the Amazon biome. Ethiopia also displayed a similar trade-off. The CRGE-Facility’s climate-first approach was not widely appreciated by all of the donors who felt Ethiopia had other more pressing priorities, such as food security.

Apart from the preferences of the fund contributors, the scope of the fund also reveals domestic actor preferences. We can distinguish between narrow funds that have regulatory mandates like limiting deforestation, such as the Amazon Fund from general-purpose funds like the CRGE Facility and the BCCRF, which target a variety of sectors. More focused funds, such as the Amazon Fund, support regulatory actions such as limiting deforestation. The Ministry of Environment and Forests serves as the fund’s host and is the target of its actions as well. Such overlap has enabled the Amazon Fund to finance significant enforcement measures in areas that are under the Ministry’s control. The fund’s slow movement towards positive incentives, as opposed to regulatory measures, is in part due to the need to engage a broader set of actors beyond the Ministry of Environment and Forests, which is more challenging to do.

The potential for multiple framings of climate change, however, complicates what we mean by the fund’s scope. This cognitive element is, therefore, quite crucial. In Ethiopia, several donors, such as USAID, have been engaged in matters that have significant climate components, yet they may not define their actions in climate terms. Actors viewed the Productive Safety Net Program as an agriculture intervention as opposed to a climate change one. USAID did not participate in the CRGE Facility, despite having an extensive aid portfolio in Ethiopia, as the climate orientation of the fund did not match with the food security framing.

5. Discussion

The case studies highlight the trade-offs that governments face when designing the governance features of national climate funds. A core finding is that developing countries seek to maximize control over the fund even at the risk of raising transaction costs for the fund. How these trade-offs are managed depends on the bargaining strength of the host country government, among other factors. The findings of this study lend support to Swedlund’s contention that a focus on the effectiveness of design features is too restrictive and far greater attention needs to be paid to the political dynamics shaping fund design even though the actual design outcomes may be less than what is theoretically optimal.

The paper examined decisions on four major aspects of fund design: scope, governing board, trustee and membership. In terms of scope, host governments face the challenge of framing the fund in a manner that resonates with donor governments but also aligns with their own priorities. As the Amazon Fund’s experience illustrated, sometimes donors may want more substantial changes, such as, changes in the fund’s scope.

In terms of the governing board, host governments could include donors and sectoral ministries in the boards of national climate funds as a way to obtain buy-in. Yet, retaining host country control over the fund can mean that donors insist on earmarks, thereby raising transaction costs. The sovereignty costs of including funders in narrowly defined funds is lower. For large multi-purpose funds, host governments are likely to minimize the areas where donors could have influence. One way in which host governments give donors the control they wish is by creating parallel project management units and allowing for earmarked allocation of resources. These features raise the transaction costs for the fund, and for the donors, but the host governments continue to exercise control over the fund’s governance. This finding is in agreement with Graham and Serdaru’s results that earmarks are an option that donor governments implement as a ‘design substitute’ in exchange for not being represented on the board or through voting powers [46].

The question of the fund’s trustee has been particularly contentious in all of the cases examined. Donors have paid heavy emphasis on of fiduciary standards and safeguards in fund design. This is largely due to the fact that these funds were all created de novo. As a result, the question of experience in fund management was secondary. Furthermore, given the general novelty of climate change programming, expertise on climate change was not the dimension on which potential trustees really distinguished themselves. When countries like Brazil had agencies like the Brazilian National Development Bank, a bank with a major track record, it was acceptable to donors even though BNDES itself had little experience with grant-based programming to stop deforestation.

Funds such as the CRGE Facility and the BCCRF managed to increase their membership as they proceeded with implementation. A growing track record of projects helped to convince otherwise skeptical donors. Similarly, those unable to contribute at the moment of founding due to funds already being committed to other programs were able to eventually switch over to the fund. The findings from this paper are consonant with Koremenos et al on the trade-off between membership and scope as well. Obtaining additional fund contributions, that is, increasing the membership, will entail broadening the fund’s scope. Whether the host government actually carries through with this or not depends on the specific contextual considerations.

The understanding of credibility that shines through the cases is more diffuse than the formal and legal understanding of agency autonomy that is prevalent in most of the literature on regulatory agencies. While legal autonomy affords a degree of insulation from political interference, states negotiate credibility at every stage and element of the process. Funds may deliberately invite other ministries to their boards to draw attention to climate change and get the buy-in of the sectoral ministries. What this understanding of negotiated credibility also helps to challenge is the implicit notion that any interference is inherently inappropriate or perverse [13].

The findings here also run counter to a perspective that would simply focus on climate finance receipt maximization by developing countries. We would see far greater flexibility and far lower resistance to the demands placed by donors if the central aim was to maximize receipts alone. Even Brazil, arguably the country in the strongest bargaining position compared to Indonesia, Bangladesh, and Ethiopia, countenanced trade-offs to attract more contributions to the Amazon Fund. One way in which countries have tried to resolve the trade-off is by compartmentalizing the range over which donors have a say. In other words, through earmarks, donors can maintain control over how their funds are spent, while recipient countries can obtain the contribution while maintaining control of the fund overall. In such a model, efficiency, however, has suffered.

6. Conclusion

By focusing on the process of designing funds, this study reveals how developing countries mobilize climate finance by negotiating mutually acceptable design components with their counterparts. While the broader efforts to construct a post-Kyoto climate regime shaped the terrain over which countries bargained, the ability of countries to mobilize climate finance depended on their ability to communicate credibility via their policies and the design features of the funds. Since the cases examined here were focused on institutional creation, further research will be required to examine whether countries face similar trade offs as the funds mature and evolve and acquire longer track records of operation [21].

Three research gaps stand out. First, the national climate funds that were examined as a part of this study were created in the late 2000s or early 2010s. The climate policy landscape has substantially changed over the last decade, most notably with the entry into force of the Paris Agreement. An important research gap is regarding how the national climate funds fit within the emerging domestic climate finance architecture. How the functions and mandates of the national climate funds have evolved alongside the broader transformation merits further attention. Second, the legal form of the national climate funds, that is, mostly in the form of a trust that is managed by a domestic or an international agency, enabled the disbursements of grants. Since many policymakers viewed these national climate funds as precursors of green banks, it will be important to track the legal and functional evolution of the national climate funds.

Broader analyses of the role that institutions play in mobilizing climate finance will be important and will complement econometric analyses that have tried to identify country-level attributes to explain climate finance receipts. Finally, an important area for further inquiry is how the evolution of national climate policies affects the ability of national climate funds to mobilize finance. Since these funds were established as financing arms of national climate change plans and policies, does the enhancement of ambition translate into greater climate finance receipts or not will be important.

The findings of this study lead to three policy recommendations. First, investing in improving fiduciary standards and safeguards significantly helps to improve the bargaining position of host countries. Governments should investment in improving the governance standards of possible trustees. Not only is a no loss proposition, it would also help to strengthen their case for a domestic trustee. Second, national climate funds should focus on building expertise and a track record of implementation. Over time, the fund will be able to garner greater interest and may achieve the scale that was originally desired. Finally, since host governments should collect and disseminate data on their actions related to climate change. Since host country commitment to climate change is crucial operating context for national climate funds, governments should also regularly report on their efforts and results.

Supporting information


The author thanks Kelly Sims Gallagher, Daniel Drezner, David Victor for their valuable feedback and input during the research process. The author also benefited from input and guidance from Rachel Kyte, Agus Sari, Abay Ezra, Saleem ul-Huq, and Roberto Schaeffer.


  1. 1. Songwe V, Stern N, Bhattacharya A. Finance for cliimate action: scaling up investment for climate and development, (2022).
  2. 2. Ballesteros A, Nakhooda S, Werksman J, et al. Power, Responsibility, and Accountability: Re-thinking the Legitimacy of Institutions for Climate Finance. Washington, D.C.: World Resources Institute, (2010, accessed 21 April 2020).
  3. 3. Pickering J, Betzold C, Skovgaard J. Special issue: managing fragmentation and complexity in the emerging system of international climate finance. Int Environ Agreem Polit Law Econ 2017; 17: 1–16.
  4. 4. Stadelmann M, Roberts JT, Michaelowa A. New and additional to what? Assessing options for baselines to assess climate finance pledges. Clim Dev 2011; 3: 175–192.
  5. 5. Romani M, Stern N. Sources of finance for climate action: principles and options for implementation mechanisms in this decade. In: Haites EF (ed) International Climate Finance. London; New York: Routledge, 2013.
  6. 6. Haites EF (ed). Mobilizing Climate Finance. In: International Climate Finance. London; New York: Routledge, 2013.
  7. 7. Swedlund HJ. The Development Dance: How Donors and Recipients Negotiate the Delivery of Foreign Aid. Ithaca, NY: Cornell University Press, 2017.
  8. 8. Whitfield L, Fraser A. Negotiating Aid. In: The Politics of Aid. Oxford: Oxford University Press, 2009.
  9. 9. Koremenos B, Lipson C, Snidal D. The Rational Design of International Institutions. Int Organ 2001; 55: 761–799.
  10. 10. Koremenos B (ed). The Rational Design of International Institutions. Cambridge, UK: New York: Cambridge University Press, 2003.
  11. 11. Hawkins DG, Lake DA, Nielson DL, et al. Delegation and Agency in International Organizations. Cambridge University Press, 2006.
  12. 12. Hawkins DG, Lake DA, Nielson DL, et al. Delegation under anarchy: states, international organizations, and principal-agent theory. In: Delegation and Agency in International Organizations. Cambridge University Press, 2006.
  13. 13. Dubash NK, Morgan B. The Rise of the Regulatory State of the South: Infrastructure and Development in Emerging Economies. OUP Oxford, 2013.
  14. 14. Peterson L, Skovgaard J. Bureaucratic politics and the allocation of climate finance. World Dev 2019; 117: 72–97.
  15. 15. Kafle K, Uprety L, Shrestha G, et al. Are climate finance subsidies equitably distributed among farmers? Assessing socio-demographics of solar irrigation in Nepal. Energy Res Soc Sci 2022; 91: 102756.
  16. 16. Garschagen M, Doshi D. Does funds-based adaptation finance reach the most vulnerable countries? Glob Environ Change 2022; 73: 102450.
  17. 17. Islam MdM. Distributive justice in global climate finance–Recipients’ climate vulnerability and the allocation of climate funds. Glob Environ Change 2022; 73: 102475.
  18. 18. Weikmans R, Roberts JT. The international climate finance accounting muddle: is there hope on the horizon? Clim Dev 2019; 11: 97–111.
  19. 19. Skovgaard J, Adams KM, Dupuy K, et al. Multilateral Climate Finance Coordination: Politics and Depoliticization in Practice. Glob Environ Polit 2023; 23: 125–147.
  20. 20. Kawabata T. Climate finance governance through transnational networks. J Sustain Finance Invest 2023; 13: 1624–1643.
  21. 21. Funder M, Dupuy K. Climate Finance Coordination From the Global to the Local: Norm Localization and the Politics of Climate Finance Coordination in Zambia. J Dev Stud 2022; 58: 1899–1916.
  22. 22. Browne KE. Rethinking governance in international climate finance: Structural change and alternative approaches. WIREs Clim Change 2022; 13: e795.
  23. 23. Ciplet D, Falzon D, Uri I, et al. The unequal geographies of climate finance: Climate injustice and dependency in the world system. Polit Geogr 2022; 99: 102769.
  24. 24. Bhandary RR, Gallagher KS, Zhang F. Climate finance policy in practice: a review of the evidence. Clim Policy 2021; 21: 529–545.
  25. 25. Geddes A, Schmid N, Schmidt TS, et al. The politics of climate finance: Consensus and partisanship in designing green state investment banks in the United Kingdom and Australia. Energy Res Soc Sci 2020; 69: 101583.
  26. 26. Michaelowa A, Hoch S, Weber A-K, et al. Mobilising private climate finance for sustainable energy access and climate change mitigation in Sub-Saharan Africa. Clim Policy 2021; 21: 47–62.
  27. 27. Hourcade J-C, Dasgupta D, Ghersi F. Accelerating the speed and scale of climate finance in the post-pandemic context. Clim Policy 2021; 21: 1383–1397.
  28. 28. Steffen B, Michaelowa A. Policies for climate finance: Status and research needs. PLOS Clim 2022; 1: e0000083.
  29. 29. Dunz N, Naqvi A, Monasterolo I. Climate sentiments, transition risk, and financial stability in a stock-flow consistent model. J Financ Stab 2021; 54: 100872.
  30. 30. Giglio S, Kelly B, Stroebel J. Climate Finance. Annu Rev Financ Econ 2021; 13: 15–36.
  31. 31. UN. Addis Ababa Action Agenda of the Third International Conference on Financing for Development, (2015).
  32. 32. Berger H, Haan JD, Eijffinger SCW. Central Bank Independence: An Update of Theory and Evidence. J Econ Surv 2001; 15: 3–40.
  33. 33. Hanretty C, Koop C. Measuring the formal independence of regulatory agencies. J Eur Public Policy 2012; 19: 198–216.
  34. 34. Elgie R, McMENAMIN I. Credible Commitment, Political Uncertainty or Policy Complexity? Explaining Variations in the Independence of Non-majoritarian Institutions in France. Br J Polit Sci 2005; 35: 531–548.
  35. 35. Bodea C, Hicks R. Price Stability and Central Bank Independence: Discipline, Credibility, and Democratic Institutions. Int Organ 2015; 69: 35–61.
  36. 36. Cukierman A, Lippi F. Central bank independence, centralization of wage bargaining, inflation and unemployment:: Theory and some evidence. Eur Econ Rev 1999; 43: 1395–1434.
  37. 37. Reinhardt E. Tying Hands without a Rope:: Rational Domestic Response to International Institutional Constraints. In: Drezner DW (ed) Locating the Proper Authorities. University of Michigan Press, pp. 77–104.
  38. 38. Lipton M. Introduction: Aid-effectiveness, Prisoners’ Dilemmas, and Country Allocations. IDS Bull 1986; 17: 1–6.
  39. 39. Monkam NF. International donor agencies’ incentive structures and foreign aid effectiveness. J Institutional Econ 2012; 8: 399–427.
  40. 40. Narlikar A. The Power of the Powerless. 2015, (2015, accessed 13 March 2020).
  41. 41. Narlikar A. Poverty narratives and power paradoxes in international trade negotiations and beyond. New York: Cambridge University Press, 2020.
  42. 42. Fisher J. When it pays to be a ‘fragile state’: Uganda’s use and abuse of a dubious concept. Third World Q 2014; 35: 316–332.
  43. 43. Hochschild J. Conducting Intensive Interviews and Elite Interviews.
  44. 44. Bennett A, Checkel JT. Process tracing. In: Process Tracing: From Metaphor to Analytic Tool. Epub ahead of print November 2014.
  45. 45. Harvey WS. Strategies for conducting elite interviews. Qual Res 2011; 11: 431–441.
  46. 46. Graham ER, Serdaru A. Power, Control, and the Logic of Substitution in Institutional Design: The Case of International Climate Finance. Int Organ 2020; 74: 671–706.