Co-Operation or Co-Optation ? NGOs ’ Roles in Norway ’ s International Climate and Forest Initiative

This paper investigates non-governmental organisation (NGO) involvement in policy processes related to Norway’s International Climate and Forest Initiative (NICFI) comparing four countries: Norway, Brazil, Indonesia, and Tanzania. Based on documents and interviews, NGO involvement is mapped using a conceptual framework to categorise and compare different roles and modes of engagement. NGOs have co-operated with government in policy design and implementation, albeit to varying degrees, in all four countries, but expressed relatively little public criticism. Funding seems to have an influence on NGOs’ choices regarding whether, what, when, and how to criticise. However, limited public criticism does not necessarily mean that the NGOs are co-opted. They are reflexive regarding their possible operating space, and act strategically and pragmatically to pursue their goals in an entrepreneurial manner. The interests of NGOs and NICFI are to a large extent congruent. Instead of publicly criticising a global initiative that they largely support, and thus put the initiative as a whole at risk, NGOs may use other, more informal, channels to voice points of disagreement. While NGOs do indeed run the risk of being co-opted, their opportunity to resist this fate is probably greater in this instance than is usually the case because NICFI are so reliant on their services.

Since REDD+ (Reducing Emissions from Deforestation and forest Degradation) was introduced on the international climate policymaking scene in 2003 [1][2][3], and officially entered the United Nations Framework Convention on Climate Change (UNFCCC) agenda in 2007 [4], this policy instrument for curbing tropical deforestation and degradation has evolved considerably [5].According to Angelsen and McNeill [6], the most important changes to REDD+ are the following: "(i) the focus has moved from carbon only to multiple objectives; (ii) the policies adopted so far are not only, or even primarily, directed at achieving result-based payments; (iii) the subnational and project, rather than national, levels are receiving a large share of resources; and (iv) the funding to date is mainly from international aid and the national budgets of REDD+ countries, and not from carbon markets" [6] (p. 31).All these factors have, to varying extents, combined to increase the role of civil society and non-governmental organisations (NGOs) in REDD+ policy and practice.Social safeguards, where local participation plays a major part, have become established as important for REDD+ [7], and it has been argued that NGOs can play important roles in mediating between different actors and interests with diverse motives [8].
However, scholars have also expressed some scepticism with regard to concerns such as participation: "it is unclear to what extent these are mere rhetoric or whether they represent genuine motivation to address such issues [in] the context of REDD+" [9] (p.25).
While there is only a limited literature on the relationship between NGOs and REDD+ there is, however, a more substantial body of social science literature studying NGOs roles and modes of involvement in environmental policymaking and implementation more generally [3,[10][11][12][13][14][15].While this literature has examined a number of cases in different countries [3,12], and sometimes compared similar cases across countries [15], most of the comparative work has been rather general, comparing environmentalism across countries as a whole [11,16,17].Little attention has been given to comparative studies of NGOs' role in designing and implementing one and the same policy initiative simultaneously in different countries [18].This is why O'Neill [19] (p.135) asserts that there is a need for further comparative study of how environmental movements and organisations address political challenges and opportunities within and across borders.
This paper will make a contribution to filling that gap, by comparing the roles of NGOs in Norway's International Climate and Forest Initiative (hereafter NICFI).Norway is the world's largest contributor to the global REDD+ initiative, accounting for about 73% of pledged funds globally [20].In addition to exploring the interface between NGOs and the Norwegian government domestically, we will use as case studies the REDD+ schemes supported in Brazil and Indonesia, being the two largest bilateral initiatives in the NICFI portfolio, and Tanzania, the first country to host NICFI initiatives as well as being the African country outside the Congo Basin with most subnational REDD+ initiatives [21] (p.219).
Environmental NGOs (ENGOs) cannot be treated as one homogeneous mass, because organisations differ substantially on a number of variables [19], e.g., with regard to organisational form (e.g., grassroots, professional foundation), financing (members, public, private sector, project-based), relationship to government and private sector (close vs. distant) as well as tactics and choice of levers for action (confrontation vs. co-operation).NGOs are generally associated with civil society and are often expressions of broader social movements.However, environmental NGOs may only represent the views of small interest groups comprised mostly by experts and/or scientists.For this reason, Yearley [12] argues that neither a definition that focuses on organisational characteristics nor a definition that focuses on the organisations' issues and purposes (e.g., working for societal transformations of different kinds) is adequate.Rather, he asserts that only an essentially descriptive definition is acceptable [12] (p.25).In other words, Yearley directs attention to the organisations' actual practices and suggests that these should be studied on a case-by-case basis.
In this paper we will follow Yearley's maxim and conduct an empirically driven descriptive comparative analysis, studying NGOs' involvement in NICFI-related policy processes in Norway, Brazil, Indonesia, and Tanzania.A central idea in the paper is to "follow the funding", and see how the size and relative share of NICFI funding contributes to shaping the organisations' roles and practices in the respective countries.We focus on the organisations most involved in Norway and Brazil, Indonesia, and Tanzania, and discuss the choices and implications of organisations taking on different roles ('insider' as policy designers, policy advisors and implementers; 'outsider' as critics; or in combinations) in different stages of the REDD+ policy processes.To compare across countries, we use a conceptual framework to map the roles played by organisations in different stages of the policy processes.
As NICFI perceives 'civil society organisations' (to use NICFIs own terms) to possess knowledge, expertise, and networks vital to REDD+, broad involvement of such organisations has been an integral part of their strategy from the very start.A vast variety of organisations receive civil society funding from NICFI, from broad grassroots social movements working on civil and indigenous people's rights, to highly specialised organisations working on e.g., forestry and biodiversity, to high-profile international organisations performing professional advocacy and lobbying.In this paper, beneficiaries of NICFI civil society funding will all be treated under the broad common category of non-governmental organisations (NGOs).Further distinctions, for instance between environmental and development NGOs, for instance, will be made when relevant.
NGOs can be seen as an expression of public concern over issues, serving to give civil society a voice in public policymaking [3].Numerous scholars in the social sciences have often encouraged public involvement in policymaking in order to arrive at sound, democratic, and legitimate decisions, especially in controversial issues [22][23][24][25][26][27].However, commentators have also warned that public involvement strategies are sometimes designed to reduce or circumvent public resistance, rather than taking public concerns seriously [28][29][30].In her highly influential paper on public involvement in policy issues, Marres maintains that "empirical case studies in STS make clear that we cannot expect that 'public involvement' will be easy to distinguish from less authentic forms of lobbying or 'public window dressing'" [31] (p.771).We agree with Marres on this point, and see this as a strong argument for more empirical studies that can improve our theoretical understanding by looking into the intersections of public involvement, lobbying, and 'public window dressing'.A pertinent question is how many different roles NGOs can take on, and whether different roles are complementary or conflicting.The case of NICFI is particularly suited for such inquiry since we are here dealing with rather different publics relating to different issues (for instance general nature conservation, bio-diversity, indigenous peoples, and human rights), though in this case co-operating closely in relation to one issue, namely REDD+.
In summary, we will describe the nature and extent of NGOs involvement in four countries (Norway, Brazil, Indonesia, and Tanzania), analyse their roles and relations to NICFI in each case (ranging from co-operative to confrontational), and draw some explanatory conclusions.
In the remainder of the paper, we first review relevant literature and operationalise this into an analytical framework.Next, we present the empirical material from the four countries.After that, we map the cases into the analytical framework and draw out some comparative observations.We conclude that NGO involvement in REDD+ policymaking in all countries has been distinctly entrepreneurial.Although the roles adopted by NGOs in the four countries have varied considerably, a common finding across the cases is that choices seem to have been guided by a pragmatic focus on results and policy impact.Norway stands out as the case where a mixed strategy of criticism and co-operation has been most clearly apparent.Giving up some public visibility in exchange for continuing to have direct contact with policymakers (both bureaucrats and elected officials) and "a place at the table" has been the price that NGOs have paid to achieve policy impact-which in this instance has been very substantial.

Study Design, Methods and Materials
In the latest report from Norad (the Norwegian Agency for Development Cooperation) about NICFIs support scheme for civil society, it states: "The non-governmental organisations play different roles.They serve as advocates, watchdogs, and independent verifiers, as well as knowledge and service providers" [32] (p. 6).Thus, official Norwegian policy documents portray these roles as complementary, exemplifying what by many is seen as "the Norwegian government/NGO model".However, as will be set out in the subsequent section, different strands of literature disagree whether, how, and to what degree NGOs actually can combine different roles; for instance close co-operation (including financially) with government, and public confrontation.Here, "the Norwegian government/NGO model" (elaborated below) is particularly interesting.A central objective of this paper is to compare and analyse how the NICFI funding scheme-where "the Norwegian government/NGO model" is implicitly embedded-plays out in the different countries.
The study is designed as a 'least likely' comparative case study [33] (p.121) as Norway and the other countries we compare are very different across a range of dimensions.Put differently, we hypothesise that we will not find patterns resembling "the Norwegian government/NGO model" in the other countries.The countries do, however, have in common one important independent variable, namely being recipients of substantial NICFI funding.This makes it relevant to trace and seek to explain variations on the dependent variables, i.e., the roles the NGOs take in different countries, and feed that back to theory testing [33] (pp.115-119).
The study uses process-tracing, a method which draws on a number of empirical sources to trace the (intervening) independent variables and possible variations on the dependent variable (i.e., what roles NGOs take), as well as uncovering and discussing explanations for that possible variance [33] (p.206).The empirical basis consists of written documents, media coverage, and interviews (both formal and informal) as well as participant observation in meetings with NGOs and government officials.The time period studied extends from 2005 until 2016, and data collection was conducted from 2008 until 2016.Political and administrative documents have been analysed together with evaluations and other reports from both the public sector and NGOs.Moreover, a systematic analysis of Norwegian media coverage of NICFI has been carried out.This analysis consisted of a search combining the keywords "rainforest" and "climate" in the database "Retriever", which contains all published articles in the Norwegian media.A similar approach was also used in Brazil to follow the media coverage of three of the country's main outlets.The justification for focusing especially on Norway is twofold: first, it is by far the largest REDD+ donor globally (73% of the funds [20]); second, as our references to the literature show, Norway has been identified as a rather unusual case with regard to the relationship between NGOs and policymakers.
In Norway, the formal interviews consisted of 11 semi-structured interviews and one telephone interview conducted with NGOs, bureaucrats, politicians, and academics.In the case of Brazil 19 interviews were conducted with representatives from the Ministries of Environment, Science and Technology, and Foreign Affairs and three of the main NGOs involved in the creation and implementation of the Amazon Fund.Informants were asked about the background for REDD+ generally and NICFI specifically, how policies had been designed and evolved in the different countries, funding schemes for NGOs, NGO involvement, criticism expressed by NGOs-both public and private-as well as prospects for the future.All interviewees gave their informed consent before the interview.
In September 2012, as part of this research, a workshop was organised in Norway with senior representatives from NGOs, public administration and academia, in total 10 persons.The purpose of the workshop was to discuss the background, status and future of REDD+ generally and NICFI specifically, including the relations between the different actors in Norway.As a follow-up to the workshop, the authors in October 2013 organized a conference in Norway (attended by about 100 people) about deforestation in Brazil, attended by politicians (at the level of state secretary in Brazil and Norway), senior officials, leading academics, NGOs, and civil society.The authors had all followed the progress of Norway's involvement in REDD+: participating in and arranging seminars and conferences, advising policymakers (mainly bureaucrats), and supervising and commenting on students' doctoral and masters theses on the topic.In addition, they had had more than two dozen informal conversations in the course of their active research involvement in REDD+ in all four countries and collected in the context of participant observations in meetings between NGOs and officials from the Brazilian and Norwegian governments.The study was conducted in accordance with the guidelines set out by the Norwegian Centre for Research Data (Norsk Samfunnsvitenskapelig Datatjeneste-NSD) [34].

Analytical Framework: Inside, Outside-or Both?
In this section we review different accounts of the "the Norwegian government/NGO model" and construct a conceptual framework to guide the analysis.Dryzek et al. [11] compared the relationship between the state and environmental NGOs in four countries (the United States, the United Kingdom, Germany, and Norway) (A weakness of the analysis of Dryzek et al. [11] is that they analyse environmentalism within states, while many environmental NGOs in fact operate across borders, as exemplified by RFN.This study contributes to the growing literature on environmental NGOs and globalization, see Rootes [35]).In the case of Norway, the authors found that the relationship between the state and NGOs was characterised by so-called active inclusion, meaning that the NGOs receive substantial financial support from the state and are actively invited to contribute to policymaking processes in a consensus-oriented manner.However, the authors also warn that NGOs run the risk of being co-opted by the state in this way: "States such as Norway cultivate groups that moderate their demands in exchange for state funding and guaranteed participation in policy making, to the extent they can hardly be called NGOs" [11] (p. 171).The authors conclude that a close co-operation between NGOs and the state might increase NGO influence on state policy, as least apparently, but might also come at the expense of co-optation, pre-empting what otherwise could have been effective outsider criticism.Basically, it comes down to a trade-off between inside vs. outside strategies.
Three years after the Dryzek et al. [11] study was released, Grendstad et al. [13] published a study where they criticise the conclusions of Dryzek et al. [11] regarding the case of Norway.According to Grendstad et al. [13], Norwegian society is much more open and inclusive than the study of Dryzek et al. [11] indicates.The Norwegian societal model, characterised by the "inclusive polity and the state-friendly society" [13] (p. 2, italics in original), has resulted in a tradition of Norwegian ENGOs being more co-operative, pragmatic, and less openly confrontational than their counterparts abroad.Norwegian NGO's have not seen the same need to distance themselves from the state and oppose it, since they do not see the state as a "threat" as such, in contrast to the situation in many other countries.Rather, the NGOs consider that they can achieve more by co-operating with the state, instead of constantly opposing it: "Rather than seeing the organizations being pressed to become moderate, we see them as being moderate from the start.It is this symmetry that makes cooperation come so naturally for both parts" [13] (p. 39).In other words, Grendstad et al. [13] maintain that it is fully possible for Norwegian NGOs to co-operate closely with the state and receive funds without being co-opted by the state.
Dryzek et al. [11] and Grendstad et al. [13] study environmental NGOs (ENGOs), but we find it relevant to refer also to the work of scholars who focus on development NGOs.This is for two reasons.One is that this gives further insight into the 'Norwegian model'.The second is that REDD+ relates, in practice, to a mix of environment, climate and development issues, and NICFI is funded from the development aid budget.
In Norway, according to Steen [36], also in the field of development aid "relations between government and NGOs can be characterised as 'near' or 'close'" [36] (p.157).Similar to that which Grendstad et al. find in the environmental field, Steen concludes that "although Norad is based on a well-regulated bureaucratic system, there are no indications that this has created any serious conflict with the NGOs" [36] (p.157).However, Tvedt [37], who for decades has been studying Norwegian NGOs working with development aid, has a rather different view.Tvedt's [37] basic claim is that development NGOs have gradually aligned with state interests in a corporatist manner, not least because of huge financial dependency on the state.According to Tvedt, "institutional relationships and patterns of interaction tendentiously eliminate distinctions between different sub-systems in society that contribute to the pluralism of the community-e.g., organisations, research and the state" [37] (p.622).The result is a consensual "aid system" where the state has the upper hand, although the corporatist system has also given opportunities for NGOs and researchers to shape policy: "The state has gradually invaded the organisations and the research institutes, while the leaders of the organisations and the researchers have broken the state's monopoly of foreign policy" [37] (p.630).
In brief, Tvedt's [37] argument is that generous funding for civil society has dampened what might have been highly relevant criticism; in other words co-opting civil society voices [37].
Summing up, we see three main positions along a continuum of relative proximity to the state.At one end of the scale we find Steen [36] and Grendstad et al. [13], who argue that Norway is a special case where a hybrid, dual strategy of co-operation and confrontation is feasible without the NGOs having to compromise.At the opposite end of the scale we find Tvedt [37], who claims that Norwegian civil society development organisations already are co-opted/silenced by the state through generous funding systems.Dryzek et al. [11] occupy a middle position, acknowledging how a close co-operation between NGOs and the state may increase NGO influence on state policy, but perhaps at the expense of co-optation, preventing what otherwise could have been effective outsider criticism.
The literature we have reviewed here relates to Norwegian NGOs.A pertinent question with regard to Brazil, Indonesia, and Tanzania is thus what roles are adopted by NGOs in these countries: how, if at all, is "the Norwegian government/NGO model" (regardless of whether one initially agrees with the position of Steen [36] and Grendstad et al. [13], Dryzek et al. [11] or Tvedt [37])-embedded in the NICFI funding schemes-"exported" and adapted to these country contexts.
In this paper, we will operationalise the positions of Steen [36] and Grendstad et al. [13], Dryzek et al. [11] and Tvedt [37], by placing them on a continuum between two contrasting strategies: co-operation and confrontation.The category of co-operation is divided into two sub-categories, relating to the two roles that NGOs may play: policy design in the initial phase and implementation on the ground.The second strategy, confrontation, is manifested by public criticism of REDD+ policies.In brief, we will classify NGO involvement according to three different roles; inside as policy designers (co-operation), inside as implementers (co-operation), and outside as public critics (confrontation).
The policy design role implies that NGOs have been active in promoting and advocating REDD+ on the international, national, and/or local level, and assisted and/or advised government in designing policies and/or institutions for governing REDD+ (for instance commenting on draft documents, participating in hearings etc.)The implementation role comprises both providing policy advice in the implementation phase, for instance regarding portfolio development, as well as actual implementation "on the ground".The last role, public criticism, refers to explicit criticism by NGOs of REDD+ in general, or NICFI in particular, in the media or in public events, meetings, and workshops.Lack of public criticism in combination with a high degree of co-operation (either in policy design and/or implementation) is seen as an indicator of co-optation.To the degree in which we have data, we will also map and discuss private criticism.
We will now present relevant information about the four cases to enable us to complete the analysis.

Results: Country Case Studies
In this section, the empirics from the country cases will be set out in detail.Before delving into the details, we have in Table 1 compared the countries across some key dimensions relevant to REDD+ to give the reader a quick and comparable overview of some important factors relevant to the analysis.

Norway
In the case of Norway, ENGOs have been heavily involved from the very start.In fact, NICFI originates from a proposal from the leaders of two ENGOs: Lars Løvold from the Rainforest Foundation Norway (RFN) and Lars Haltbrekken from Norwegian Society for the Conservation of Nature/Friends of the Earth Norway (FOEN) (see [3,38,39] for thorough analyses, including timelines of major events (particularly [39] (pp.26-29) and [38] (p.41)).This heavy NGO influence on NICFIs inception has continued to be a long-lasting characteristic of the initiative and we will therefore devote some space to explain the emergence of the initiative.
In September 2007, the leaders of RFN and FOEN sent a letter to leading Norwegian politicians.The authors-"Lars & Lars"-here proposed that Norway should allocate 6 billion NOK (approximately 1 billion USD) (The NOK/USD exchange rate has varied considerably in recent years.For convenience we have used a figure of 6:1 throughout this paper, and will in subsequent cases omit the word 'approximately') annually over five years to protect rainforest as a climate mitigation measure in a payment-for-performance mechanism, very similar to what has become the REDD+ mechanism.The two ENGOs then went on to lobby their proposal, both to the sitting 'red-green' coalition government (consisting of the Labour Party, the Socialist Left Party and the Centre Party) and the centre/right opposition, arranging meetings with political parties and attending hearings in Parliament.They even flew in the prominent REDD+ proponent Marcio Santilli from the Brazilian Instituto Socioambiental (The Socioenvironmental Institute-ISA) to back their case [3].Also Brazilian Minister of the Environment, Marina Silva, attended an event organised by RFN.In order to gain broad support for their proposal, they also induced other Norwegian ENGOs to back the rainforest proposal.
Parallel to this, under public pressure, the political opposition challenged the cabinet to enter into a 'climate settlement' (a cross-party Parliamentary agreement on climate policy).The discussions in Parliament soon heated up and turned into a political bidding war over climate policy integrity [3].Then, just a couple of months after the letter was sent, Prime Minister Stoltenberg announced at the UN Framework Convention on Climate Change (UNFCCC) Conference of the Parties (COP) 13 in Bali that Norway would grant 3 billion NOK (500 million USD) annually for rainforest protection.The decision came as a surprise to the general Norwegian public and most of the public administration, and was strikingly similar to the proposal from the two ENGOs apart from the sum, which was cut by half from the original proposal-but still extremely large in relative terms [3].The ENGOs had argued that the funding should not be allocated from the aid budget, but as a political compromise, NICFI was funded by an increase in the Norwegian aid budget, thus simultaneously achieving Norway's political target of allocating 1% of Gross National Income (GNI) to Official Development Assistance (ODA) [38].Approximately one month after the Bali meeting NICFI was established as a part of the cross-party climate settlement.In 2012, a new climate settlement was agreed, prolonging NICFI until 2020.At COP 21 in Paris in December 2015, Norway announced that NICFI would be extended to 2030.
When NICFI was approved by Parliament in January 2008, Norway did not have the governmental apparatus in place for implementing it.The Ministry of Foreign Affairs (hereafter MFA) and Norad (the Norwegian Agency for Development Cooperation, a Directorate under the Norwegian Ministry of Foreign Affairs) were both considered as possible hosts for the initiative, as NICFI was funded from the ODA-budget.However, NICFI was primarily thought of as a climate initiative, not a development initiative, despite being funded as aid.Erik Solhem (Erik Solheim in May 2016 became Executive Director of the United Nations Environment Programme, UNEP), who at that point had the double minister-post as Minister of the Environment and International Development (international development is a branch under MFA), agreed with Prime Minister Stoltenberg that the Ministry of the Environment should host NICFI.Part of the reason for this decision was that Norad was considered too risk-averse and bureaucratic by Solheim.Here is Solheim's own account from our interview: "They had to have the activist attitude, they had to have daily direct contact with the political leadership.If not it would just be bogged down in bureaucracy, it would be killed by all the bureaucratic objections that the system produces every day and hour.The point is that this is a political project to drive the climate fight.To locate it within Norad was never an alternative as I see it, because the leaders of the program needed daily access to me as the Minister.This was a signature Norwegian effort to change global behaviour and could only succeed at the highest level of politics, meaning at the centre of the Ministry."(our translation) Instead, a small fast-working task force called "the climate and forest secretariat" (in Norwegian 'klima-og skogsekretariatet'-hereafter KOS) consisting of just a few persons was established in the Ministry of the Environment to manage NICFI.An experienced diplomat, Hans Brattskar, was handpicked by Solheim to head the secretariat.Other staffers had background from NGOs, bureaucracy, and management consulting.However, there were very few people with development experience, at least when first established.Norad was given an advisory role and tasked with managing the vast pot of civil society funds.(Although under the Ministry of Foreign Affairs, in matters regarding NICFI Norad reports to the Ministry of Climate and Environment).KOS is still the primary policy and implementing body and has now grown to 17 staff in Norway [40].In, addition, KOS has a network of eight staff located in six different Norwegian embassies.But the number of staff in KOS is still relatively small compared to the scale of the funds they manage.
The competence on tropical forests in the Norwegian state bureaucracy was virtually non-existent at the time when NICFI was established.Although Norad and the MFA had experience in large international development projects, and advised KOS on these issues from a general perspective, they too had very little competence on tropical forests.The first director of NICFI and manager of KOS, Hans Brattskar, in fact had no experience with or competence on rainforests at all.RFN, on the other hand, has a very substantial knowledge base due to almost three decades of work on rainforest projects.Consequently, competence from RFN played a decisive role in the initial setup of NICFI.RFN even educated the KOS staff through courses on rainforest issues [41].In the early phase, KOS also commissioned the high-profile professional not-for-profit organisation "Meridian Institute" to facilitate the production of a number of influential reports [42] drawing on input from a range of different actors.Over the years KOS has somewhat expanded its own competence base, while also making substantial use of international consultants.Nevertheless, RFN has remained an important partner for KOS and NICFI more generally (for instance through a close co-operation with Norwegian embassies) regarding policy design and portfolio development, both in Norway and abroad.
In 2014, 125 million NOK (21 million USD) of RFN's total revenue of 151 million NOK (25 million USD) stemmed from Norad and the Ministry of Foreign Affairs [43].RFN received sizeable funding from MFA and Norad before NICFI was established, most notably through MFA's Amazon programme, but their total budget increased substantially thanks to REDD+.The foundation has been given a central role in the implementation phase of NICFI and greatly increased its staffing.RFN is by now the largest rainforest conservation organisation in Europe.The clearest indicator of RFNs central role in advising KOS-and NICFI more broadly-is the organisation's share of the NICFI funds for civil society administered by Norad: RFN is the second largest beneficiary, and in the period 2009-2012 received 7.5% of the total funds, next after CIFOR (Center for International Forestry Research), which received 12.3% [44] (p.113).Third on the list was WWF International, with 6.8%.The funding period 2013-2015 shows a similar pattern: in the total period 2009-2015 CIFOR is on top of the list, RFN second and WWF International third.Both RFN and WWF Norway, and particularly the former, have frequent meetings with KOS and are in continuous dialogue with them.
Thus, RFN has certainly played a significant role in advising on policy design and implementation, i.e., co-operation.However, what about confrontation?There are a few instances where RFN has voiced its concerns in public.In the early phase, RFN engaged in the question of what organisations Norway should co-operate with (e.g., favouring the UN over the World Bank), and criticised inadequate rainforest funds in the state budget.RFN and FOEN have also on a few occasions voiced public criticism towards the official Norwegian position on REDD+ in the international climate negotiations under the UNFCCC.In 2010 [45] and 2012, letters were sent signed by Lars Løvold (RFN) and Lars Haltbrekken (FOEN) to the Ministers in charge of NICFI (The 2010 letter was addressed to Minister of the Environment and International Development (Solheim) 2010), while the 2012 letter was addressed to the Minister of the Environment (Solhjell) and the Minister of International Development, Holmås).Both letters addressed a recurring point of disagreement between RFN and the Norwegian government regarding the official Norwegian position on REDD+: whether Norway should pursue a narrow, carbon-focused approach to REDD+, or take a broader view on the issue.RFN and FOEN have repeatedly criticised the Norwegian official position for being too narrow and carbon-focused, and instead argued for a broader approach where safeguards such as human rights, biodiversity, good governance, and inclusive processes are given a more central position.In addition, in both letters Løvold and Haltbrekken argue for their original position, that REDD+ should not be a carbon offsetting mechanism similar to the Clean Development Mechanism (CDM), but come in addition to other mitigation efforts.RFN and FOEN have also made an effort to generate media attention to their position on these issues, but have found it quite difficult to catch journalists' attention, as the issues may seem rather technical and beyond the scope of public interest.Norwegian public authorities have on their side largely avoided debating the broad/narrow REDD+ issue in public.
Another example of public criticism emerging from the media analysis relates to RFN's criticism of the Norges Bank Investment Management (NBIM) investment practices.NBIM is the Norwegian state's sovereign wealth fund; the world's largest of its kind.On several occasions, RFN has both publicly and privately criticised NBIM's investments for having negative consequences for rainforests, through NBIM's investment in petroleum extraction, mining, palm oil, and pulp.RFN argues that it is paradoxical that Norway, as the world leader in REDD+ finance, also (through NBIM) invests up to 40 times more in commercial activities that destroy rainforests.After RFN-and others-had pushed this case for several years, NBIM finally took action in 2013 and included "severe environmental damage" among the criteria for excluding companies from their investment portfolio.In 2014, "climate damage" was included in the exclusion criteria, which also include rainforest destruction.RFN advised NBIM in the process of developing the exclusion criteria.Although NBIM have several times pulled out of companies associated with rainforest destruction, RFN is still not satisfied.The last time they expressed strong public criticism on the NBIM issue was 22 May 2015 [46].Approximately one year later, on 9 March 2016, RFN publicly acknowledged that NBIM was on the right track, having reduced the portfolio of rainforest destroying companies by 20 per cent in one year [47].However, RFN also stated that NBIM still had a way to go.It is highly likely that RFN will continue to pursue this case in the Norwegian public sphere.In summary, RFN publicly criticises Norwegian authorities, such as the Ministry of Environment, and NBIM, and indirectly the Ministry of Finance and Norges Bank (the Norwegian national bank)-but not to any similar extent NICFI, KOS, Norad, or the Ministry of Foreign Affairs, i.e., RFN's biggest funders.
In fact, KOS has unofficially backed RFN efforts on the NBIM issue, as they saw that RFN could (and actually did) achieve things that KOS was not able to achieve themselves.RFN, on their side, allied with MFA for pushing KOS towards including development issues into the REDD+ strategy, rather than taking a narrow carbon-focused approach.These points also remind us that we should be careful seeing the state as representing a single position articulating one coherent voice.Rather, it has been documented that the Norwegian state operates with polyphonic voices on environmental issues [48].That certainly applies to REDD+ and NICFI as well, as there have been tensions between KOS/NICFI/The Ministry of Environment, MFA, and Norad from the very start regarding NICFI policy and practice.RFN has been aware of these tensions and, when appropriate, attempted to take advantage of them to pursue their agenda.Finally, whenever Norwegian politicians have discussed possible cutbacks in NICFI funding, for instance in the fall of 2013, as well as 2015 and 2016, RFN and FOEN have mobilised heavily in the media against politicians (mainly from the Conservative Party and the Progress Party, the latter not being part of the cross-party parliamentary climate settlement) to prevent budget cuts.At one point RFN even called for a broader political debate about NICFI by publishing a story on their website titled "Let the forest into the Parliament" [49] (our translation).
RFN (together with FOEN) have also been critical of NICFI's engagement in Indonesia in Norwegian media [50].While Solheim praised the 2011 moratorium on new forest concessions as an "important step forward" [39] (p.21) the two ENGOs criticised the moratorium for having loopholes and limitations, and suggested that financial transfers to Indonesia should be reduced [39,51].On the one-year anniversary of the logging moratorium, the two ENGOs together with Greenpeace Norway arranged a public event where they handed Solhjell (newly appointed Minister of the Environment) a letter from several Indonesian NGOs, expressing concern that deforestation in Indonesia was not decreasing, because a strong lobby had been able to weaken the moratorium [39,52].The Norwegian ENGOs argued that Norwegian authorities should push Indonesia to prolong the moratorium, and stand up against the pressure from the lobby.At the same time, Løvold from RFN recognised that the moratorium had led to more openness around forest issues, which opened up opportunities for Norway to exert pressure on the Indonesian government [39,52].
WWF International is the third largest recipient of NICFIs/Norad's civil society funds.WWF Norway, by contrast, only received 0.2% of the NICFI civil society funds in 2009 [44] (p.114).WWF Norway has several projects funded by Norad and the Ministry of Foreign Affairs, but a significant part of WWF Norway's project portfolio is not directly related to REDD+, and many projects were agreed before NICFI came into play.In 2014, 85 million NOK (14 million USD) of their total budget of 117 million NOK (20 million USD) came from Norad, the Ministry of Foreign Affairs and the Ministry of Climate and Environment [53].Despite being a significant player, WWF Norway has not been involved to the same degree as RFN in NICFI, but like RFN, WWF Norway also has meetings with KOS and co-operates with them.WWF Norway also has an advisory role in WWF International's REDD+ projects, which are more focused on implementation "on the ground" compared to RFN's engagement.WWF Norway does not have the same public voice on rainforest issues in Norway as RFN, although it is generally rather vocal in criticising the Norwegian government, primarily on domestic policies, including climate policy.
Ironically, WWF has itself been subject to public criticism in Norway regarding implementation of a forest project in Tanzania funded by Norway [39] (p.18).In November 2011, three Norwegian researchers wrote an opinion piece where they accused WWF of being too close to Tanzanian authorities and thus indirectly involved in activities depriving local population of their livelihoods [54].The chairman of WWF Norway responded and rejected the allegations [55].In January 2012, many of the same claims were repeated in a news article based on a published study [56,57].In the wake of these incidents, a fierce debate broke out between the researchers and WWF in Norway, culminating in a public seminar about the issue.The controversial project was, it later turned out, not connected to REDD+ as such [58].A few months later, two other Tanzanian REDD+ pilot projects were stopped [59].What is interesting for this paper is that these events did not attract a lot of media attention [39] (p.18).NICFI as such was not questioned.
FOEN, although instrumental in initiating and advocating for what became NICFI [3,38], has since then taken a rather limited role.After succeeding with their proposal back in 2007/2008, RFN and FOEN have arrived at a sort of division of labour: RFN co-operates closely with government authorities and is actively involved in advising and assisting KOS, occasionally voicing rainforest issues in the public arena.FOEN-which primarily focuses on environment rather than development issues-steps in when there are discussions about more overarching issues, such as cuts in NICFI funding, or whether REDD+ should be part of an international carbon trading mechanism (of which they are critical).Put differently, FOEN mainly concentrates on the domestic domain, and does not co-operate directly with or advise KOS.Neither does FOEN confront KOS or criticise NICFI in public to any large extent.FOEN has received only a small share of NICFI's funding for civil society, a grant amounting to 0.7% of that budget-which was shared with RFN [44] (p.114).That Norway's REDD+ funding is classified as aid is particularly anomalous with regard to Brazil, as this middle-income "BRIC" country has consequently become the largest recipient of Norwegian ODA funds [39] (p.16).The decision to finance NICFI with aid money (which did not please Brazil) was taken for political reasons; it enabled Norway in achieving the aid target of 1% of gross national income (GNI) [38] (p.42).There are good arguments against treating global public goods such as REDD+ as development aid [60] (p.119).And some NGOs working with poverty and development, as well as the Christian Democratic Party and the Socialist Left Party, have from the very start been critical of the possible negative effects on poverty reduction by redirecting aid money to rainforest projects in relatively rich countries like Brazil [39] (p.16) [61,62].But ENGOs have been comparatively silent on this point, seeking to sidestep the issue by emphasising the possible poverty amelioration effects of REDD+ [39,61].
Perhaps this points to a bigger issue: that the ENGOs have been vocal with public criticism of overarching NICFI policy design issues (e.g., Norway's position on REDD+ in the UNFCCC negotiations), but relatively silent on implementation issues, at least in the public sphere.This observation also supports the point made above, that RFN publicly criticises NBIM, and indirectly the Ministry of Finance and Norges Bank (the Norwegian national bank)-but not NICFI/KOS, Norad, the Ministry of Climate and Environment or Ministry of Foreign Affairs, i.e., RFN's biggest funding sources.Development NGOs, however, have publicly criticised these agencies, directly and/or indirectly [61,62].Part of the reason why ENGOs have been silent on implementation issues may be that they have had good internal channels to KOS for critique, feedback and advice.
Summing up, we observe that ENGOs have co-operated extensively with the state and been instrumental in policy design of the Norwegian rainforest initiative, as well as giving policy implementation advice regarding, for example, which agreements to enter into and on what conditions.Including NGOs and drawing on their competence has been Norwegian government policy from the start of NICFI [44] (p. 7).Regarding confrontation, public criticism of NICFI as such, or KOS and related agencies, has from NGOs been limited, but not entirely absent.Development NGOs have arguably been more vocal on implementation issues "on the ground", while ENGOs have been more vocal on overarching policy design issues.Although some researchers have publicly voiced criticism of various kinds, this criticism has not resonated loudly in the Norwegian public sphere.Overall, there is surprisingly little debate about NICFI in the Norwegian public domain [39,63], which is interesting given that a core aim of the Norad civil society funding is to "enhance the engagement of civil society at national level to generate an open, inclusive, and comprehensive debate on REDD+" [44] (p. 7).

Brazil
The development of REDD+ in Brazil is marked by an internal conflict between different sectors of the government and a strong participation of NGOs.When the concept of REDD+ was proposed by a group of Brazilian and North-American scientists and activists in the mid-2000s [1,2], it was well received by the Ministry of Environment under Marina Silva (2003)(2004)(2005)(2006)(2007)(2008) and Carlos Minc (2008-2010).In addition to the favourable political context, Brazilian NGOs have developed over the years the capacity to exert national and international pressure to reduce deforestation in the Amazon and in this way influence environmental policies in the country [64].One of the advocacy efforts of civil society during this period was a zero-deforestation pact carried forward by a group of Brazilian researchers from the Federal University of Minas Gerais and a group of NGOs led by IPAM that included cost calculations for ending Amazon deforestation by 2015 through a mechanism similar to REDD+ [3,65].At the same time, more conservative sectors of the government such as the Ministry of Foreign Affairs openly opposed carbon markets with credits from REDD+ because it would allow developed countries to maintain their high emission levels.In private conversations, some diplomats and scientists representing Brazil in the climate negotiations also recognised that REDD+ may be problematic, since the creation of legal obligations for Brazil regarding conservation of the Amazon would represent a threat to its sovereignty [66,67].These concerns were also made public during the initial stages of the creation of international fund for the Amazon that later became the Amazon Fund.When asked by reporters about what he thought about the creation of the fund, President Luis Inácio Lula da Silva said that he was worried about the possibility of foreign intervention in the Amazon: "I will talk with the Minister [of Environment . . .] it is a concern that if someone donates US$ 10 he will soon think that he owns the Amazon".(our translation) Brazilian and Norwegian NGOs played a key role in circumventing the political obstacles for the establishment of the Amazon Fund.On the Norwegian side Lars Løvold from the RFN was able to communicate to NICFI/KOS the scepticism voiced by some Brazilians against channelling funding through the World Bank [39] (p.15), a solution adopted in a previous large-scale scheme funded by the G7 countries (known as PPG7), that would restrict Brazilian control over the fund.As an alternative it was proposed that BNDES (the Brazilian Development Bank) would receive and distribute the resources to projects in the Amazon.This solution not only safeguarded Brazilian sovereignty, but was also well received by NICFI/KOS, which recognised that it would be difficult to manage such a fund from a distance.On the Brazilian side, key NGOs, such as ISA (Instituto Socioambiental, Socio-Environmental Institute) and IPAM (Instituto de Pesquisa Ambiental da Amazônia, Amazon Environmental Research Institute), followed closely the negotiation process.This enabled the emergence of a governance structure for the Amazon Fund in which representatives from the Federal government, Amazonian states, and NGOs would have equal voice in the fund's steering committee (COFA).Many of the Brazilian officials involved in the negotiations on behalf of the Ministry of Environment also had strong ties to the third sector having worked in NGOs for many years prior to joining the government.Therefore, people with NGO-background were involved in governmental decision-making in both Brazil and Norway, and were important for the outcome of the negotiations [68].The Minister Carlos Minc would later recognise in an interview that this innovative arrangement was crucial for the approval of the Amazon Fund: "Lula almost did not sign [ . . .] but I explained that the Fund is much more autonomous and sovereign than the PPG7.The resources will be managed by the Brazilian National Development Bank and the people from the federal government, scientific community, NGOs and the state governments of the Amazon."[69,70] (our translation) The Memorandum of Understanding between Brazil and Norway was signed in Brasília on 16 September 2008.According to this agreement, Norway would until 2015 donate up to 1 billion USD to support actions to further reduce deforestation in the Amazon.In order to obtain this donation Brazil would need to continue reducing its emissions from deforestation in relation to their historical levels.The amount receivable was calculated by considering a reference level corresponding to the average emissions taking place due to deforestation in the Amazon between 1996 and 2005.By measuring the difference between this reference level and the actual emissions taking place between 2006 and 2010, the Amazon Fund would be able to receive a donation of 5 USD for every ton of CO 2 (tCO 2 ) of avoided emissions from deforestation.
While the fund's steering committee was responsible for establishing the main directives and priorities of the Amazon Fund, it was up to BNDES technical team to decide which projects were eligible for funding.These approvals, however, took place very slowly, mostly due to the complex bureaucratic requirements from BNDES.Many NGOs were already used to dealing with large bureaucracies, such as Norad and USAID, to obtain funding.However, according to members of NGOs that submitted projects to the fund, BNDES was more stringent as it adopted the same procedure and criteria used to evaluate large-scale investments such as the building of dams and airports.With these procedures it is expected that the funds disbursed by BNDES become more traceable, and the likelihood of corruption reduced (The Brazilian government is currently facing major corruption scandals involving the country's main political parties and large public and private companies but to our knowledge BNDES staff have not been the target of corruption accusations).Thus, many NGOs had to make substantial investments in order to engage in lengthy application processes and to adapt their organisational structure and procedures.Consequently, by the end of 2012 the fund disbursed only 71 million Reais (approximately 23 million USD), threatening Norway's' 1% ODA target since the funds remained in Norwegian bank accounts [39,71].Thus, the Amazon Fund's slow implementation became a point of criticism both in Brazil and in Norway even in a context where deforestation rates continued to decline [72].The situation started to change in 2013 when the maximum value allowed to individual projects was raised, and state governments in the Amazon applied for large projects.At the same time, the BNDES staff learned rapidly and were able to start evaluating the applications to the fund more effectively, while NGOs adapted to the requirements set by the fund.
With the exception of the criticism in 2012 concerning the high level of bureaucracy and slow pace of project evaluation, the Amazon Fund has been largely shielded from public criticism since its inception.In closed meetings, however, NGOs have voiced their concerns about different aspects of the fund's management to representatives from the Norwegian and Brazilian governments.In particular, concerns were raised on various occasions that the projects supported by the fund may not be leading to additional reductions in deforestation as initially expected.It should be noted that the disconnect between many projects of the Amazon Fund and the reduction in deforestation can be traced to the very design of the fund.As mentioned above, the fund receives donations in proportion to results already achieved; thus in principle the projects it supports are not required to show evidence of further reductions in order to be justified.Furthermore, it was decided in 2013 that the fund was to support mainly "structuring" projects that aim at improving environmental governance in the long-term rather than produce direct short-term results.
Nevertheless, there has been mounting private criticism by NGOs and researchers seeking to increase the effectiveness of the fund in different ways.First, the fund was criticised for taking a passive stance whereby it waits for project proposals rather than actively looking for situations that may require resources to slow deforestation.Second, some NGOs and researchers pointed out that the fund has invested a sizable share of its resources into the implementation of the rural environmental registry (CAR), despite evidence that so far this policy tool did not contribute to reduce deforestation [73,74].Third, contradictions have been pointed out between projects supported by the fund.In one case, a representative of an influential NGO criticised a large grant to equip the National Force.While this elite police force plays an important role in escorting forest rangers from IBAMA (the federal environmental agency), the National Force has also used violence against native Indians opposing the construction of Belo Monte Dam in the state of Pará-a project at odds with the socioenvironmental aims of the fund [75].Even though these are important matters, and members from NGOs were very concerned, these issues did not find a place in the traditional spaces of public activism such as the press, reports, protests, and social media.
The contradiction between closed-door criticism and public silence (and thus apparent approval) of the Amazon Fund by large NGOs can be explained by several factors.First, the Amazon Fund has become an important financial source for the third sector as a whole and especially for NGOs, as illustrated in Figure 1 below.By the end of 2015, 31 of the 69 projects accepted by the Amazon Fund were proposed by third sector organizations.This corresponds to 30% of the R$ 1 billion (approximately US$ 300 million) approved by the fund.It should be noted that FUNBIO and FBB, the two largest recipients from the third sector, are foundations managed indirectly by the Brazilian government.Furthermore, NGOs supporting indigenous groups and small farmers such as the Centro de Trabalho Indigenista, have also received substantial resources.Nevertheless, almost all large environmental NGOs with a long history of engagement in the region have projects approved by the fund, corresponding to nearly half of the grants to the third sector (see Figure 1).The increasing reliance on government funds implies that NGOs could lose part of their ability to promote the environmental agenda independently.In this regard, a director of a large environmental NGO reported in a private conversation that his organisation strives to keep the contribution from the Amazon Fund below one third of its total budget in order to avoid being co-opted by the government.This implies that NGOs perceive that the Amazon Fund provides not only an opportunity for the expansion of conservation projects on the ground but also poses a risk for environmental policy advocacy.
Forests 2017, 8, 64 15 of 27 structure of the Amazon Fund the board of REDD+'s national committee is composed of eight representatives from the Federal Government and two representatives from the states, one from the municipalities and two from civil society.Seeing this decision as unfair, a coalition that includes most large NGOs in Brazil requested, in an open letter, that the Ministry of Environment adopt for the REDD+ national commission the same governance structure used in the Amazon Fund [76].
Even though the governance structure of the fund was perceived by NGOs as being overwhelmingly influenced by the Ministry of Environment, many saw the Amazon Fund as a very important first step in the direction of the establishment of larger and better mechanisms in the future since the REDD+ strategy opened the possibility of allowing states to receive resources directly from donors, thus bypassing BNDES.Fourth, NGOs fear that criticism of the Amazon Fund may disturb the delicate political balance that made possible the collaboration with Norway.If the fund is publicly criticised, Norwegian society may call into question the investment of public money in the Amazon Fund and request that Second, the central role of NGOs in the governance of the Amazon Fund implies sharing responsibility for its performance.As in the case of WWF in Norway, NGOs end up having their reputation tied to the fund by implementing projects.For instance, some of the same NGOs that privately criticised the excessive emphasis of the fund on the implementation of CAR were also members of the steering committee that approved in 2013 the change to the fund that gave priority to structuring projects.Furthermore, the Amazon Fund granted to environmental NGOs substantial resources for the implementation of large projects, such as "Sustainable Settlements in Amazonia" from IPAM and the "Green Turn" by TNC with total investments of approximately US$ 8 and 5 million, respectively.Therefore, by criticising the fund, they would indirectly become more vulnerable as well.
Third, NGOs fear that by criticising the Amazon Fund they might hinder an initiative that, despite its shortcomings, has qualities that other governmental initiatives lack.The virtues of the Amazon Fund become particularly clear when it is compared to the Brazilian national REDD+ strategy: the mechanism establishing the rules for the distribution of financial resources from UNFCCC's Green Fund.The of the national strategy started in 2008 in parallel with the implementation of the Amazon Fund.But in contrast to the latter, the Ministry of Environment has been very reluctant to create a governance structure that gives equal voting rights to NGOs and state governments.The Ministry of Environment finally published the national REDD+ strategy and the decree detailing its governance structure in November 2015 following a process that, according to the members of the NGOs and state governments, was largely undemocratic (This is also clear in a letter sent by the coalition of state governments of the Amazon to the Federal Government in 2014 complaining that no joint meeting to develop the national REDD+ strategy was held since 2012).As expected, BNDES was nominated as the manager of the resources, but in contrast to the balanced structure of the Amazon Fund the board of REDD+'s national committee is composed of eight representatives from the Federal Government and two representatives from the states, one from the municipalities and two from civil society.Seeing this decision as unfair, a coalition that includes most large NGOs in Brazil requested, in an open letter, that the Ministry of Environment adopt for the REDD+ national commission the same governance structure used in the Amazon Fund [76].Even though the governance structure of the fund was perceived by NGOs as being overwhelmingly influenced by the Ministry of Environment, many saw the Amazon Fund as a very important first step in the direction of the establishment of larger and better mechanisms in the future since the REDD+ strategy opened the possibility of allowing states to receive resources directly from donors, thus bypassing BNDES.
Fourth, NGOs fear that criticism of the Amazon Fund may disturb the delicate political balance that made possible the collaboration with Norway.If the fund is publicly criticised, Norwegian society may call into question the investment of public money in the Amazon Fund and request that this rather be allocated to more urgent matters such as the refugee crisis in Europe.While some members of the NGOs we interviewed expressed their willingness to open a public debate about the Amazon Fund, they fear that if not well managed this process may backfire and end an influx of resources that is crucial for the continuation of many environmental policies in Brazil.Furthermore, even if in the unlikely event that Norwegian diplomacy changed its "light approach" [60] and became more demanding following an outburst of public criticism of the fund, this would be risky since more conservative sectors of the Brazilian government could accuse Norway of interfering in the country's sovereignty.This suggests that NGOs see the Amazon Fund as an initiative that could be improved from the inside, rather than attacked from the outside by adopting traditional social activism tactics.
Finally, since the creation of the Amazon Fund in 2012, Brazilian NGOs have been engaged in many important public disputes with the Brazilian government.The most important one was the revision of the Forest Code approved in 2012 that provided an amnesty for 58% of the area illegally deforested before 2008 [77].Even though many sectors of the Brazilian society, including not only NGOs but also researchers and artists, were publicly against the revision of the code, the government successfully used the strong reductions in deforestation rates occurring in that period as part of the argument for the creation of a lighter yet "enforceable" law.Similarly, NGOs have also lost the fight against the government in relation to the construction of the Belo Monte and other hydroelectric dams in the Amazon (paradoxically also funded by BNDES) that have been causing major social and environmental impacts in the region [78].At the same time, there has been an increasing distance between the Ministry of the Environment and NGOs in Brazil after Marina Silva left the Lula government in 2008, and Dilma Rousseff took office in 2011.In this process many of the more progressive senior officials working on climate have been gradually replaced by officials that work more in line with the agenda set by the conservative Ministry of Foreign Affairs.In this context, and similar to the Norwegian case, many NGOs may regard it as inappropriate to express strong public criticism of the Amazon Fund since it is seen overall as a positive mechanism, especially in comparison to other governmental policies.
The above suggests that ENGOs find it difficult to publicly criticise the Amazon Fund, given the relatively open governance structure of the fund and its history of co-operation.ENGOs are aware of, and privately point out, the many shortcomings and contradictions of the Amazon Fund.But if the fund were to cease to exist because of withdrawal by Norway, ENGOs and the environmental cause would lose more than they win-in a context in which it has been increasingly difficult for them to influence key policies such as the Forest Code.

Indonesia
As the second largest deforester worldwide, Indonesia stood out as a country which NICFI/KOS almost had to engage with, despite substantial challenges, including internal conflict between different sectors of the government, a powerful private sector, and high levels of corruption.In contrast to Brazil, Indonesia was included in NICFI's portfolio with a much more measured approach emphasising a logical sequence of phases from readiness to full implementation [79].However, some have nevertheless suggested that Norway applied an unduly 'light touch' in their relationship with the Indonesian government [60].Knowledge about the challenges of governance in the Indonesian forest sector is widespread in Norway, and such concerns were often aired in public meetings about REDD+.Adding to this cautiousness, Norwegian authorities had some negative experience with an environmental agreement with Indonesia initiated by the Brundtland government (1990)(1991)(1992)(1993)(1994)(1995)(1996) as part of its strategy to strengthen economic ties with Indonesia.This agreement also included rainforest conservation projects in Sumatra.However, these were harshly criticised for insufficient implementation and inadequate concerns for local populations by a Norwegian consultancy charged with evaluating the agreements [80].
Angelsen [79] (p.17) notes that NICFI's agreement with Indonesia was delayed for several reasons, among them a low willingness to reform on the Indonesian side.After lengthy discussions, Norway and Indonesia signed a Letter of Intent (LoI) in late May 2010, at the same time as the large "Oslo Climate and Forest Conference" was arranged in Norway's capital.The bilateral agreement pledges up to USD 1 billion in performance-based payments before 2020, in a sequential approach starting with preparations, capacity-building and pilot project identification [81].President Yudhoyono was very supportive of REDD+, and shortly after the LoI was signed he created a REDD+ Task Force to implement it.Indonesia presented a REDD+ strategy in June 2012 (one of the conditions of the agreement), and established a REDD+ agency in 2013.A moratorium on new forest concessions was established in 2011, extended to 2017.
According to a 2014 report [82], Indonesian stakeholders diverged in their views on REDD+, and the institutional situation "on the ground" in Indonesia was perceived as cluttered [39] (p.20).This observation led the authors to ask: "[A] major question for NICFI is how to support the necessary changes, and whether and how to do so within the context of the pay-for-performance approach."[82] (p.10).Implementation of REDD+ has indeed proved difficult, but it is widely agreed that one of its effects has been to enhance the role of civil society.Based in part on their earlier experience in Indonesia, and, more generally, on the need for local participation for REDD+ to be successful, Norway included as one of seven 'principles' in the LoI the following requirement: "b.Give all relevant stakeholders, including indigenous peoples, local communities and civil society, subject to national legislation, and, where applicable, international instruments, the opportunity of full and effective participation in REDD+ planning and implementation."(While NGOs have often been critical of conditionality, they do sometimes in effect recommend it-to promote what they regard as valid ends.) Promoting 'participation' has been sought mainly by involving NGOs, to try and ensure that the people affected have the opportunity to influence the decisions taken.RFN has been actively involved in Indonesia for many years, initially mainly through their collaboration with KKI WARSI (Komunitas Konservasi Indonesia Warung Informasi Konservasi, Indonesian Conservation Community, an alliance of twenty NGOs in Southern Sumatra), starting in 1996 with the project mentioned above.Their strategy has been to partner with local groups, such as KKI WARSI, WALHI (Wahana Lingkungan Hidup Indonesia, The Indonesian Forum for Environment) which is part of the Friends of the Earth network) and AMAN (Aliansi Masyarakat Adat Nusantara, Indigenous Peoples Alliance of the Archipelago), building capacity and seeking to influence government policy.In Indonesia, a large number of NGOs-both national and local-have been heavily involved in REDD+, often with indirect financial support from NICFI.In the most recent allocation WALHI has been granted NOK 13 million for the period 2016-2018, directly-not via RFN [83].Norway has also provided funding to Indonesian NGOs via UN-REDD, which has been a major recipient of Norwegian REDD+ money.The NGOs have been engaged to undertake a range of different tasks: to assist in selecting pilot sites; provide socio-economic background information; activate communities by communicating about REDD+; to assist in seeking 'free, prior, informed consent' (FPIC); and to prepare action plans.In future, when further progress has been made, they may also, for example, be engaged to oversee implementation, and monitor progress.It should nevertheless be noted that the total funding involved is far less than the amount allocated to Brazilian NGOs.
The attitudes-and responses-of Indonesian NGOs towards REDD+ vary widely.At one extreme, WALHI (Friends of the Earth, Indonesia) have a slogan: "REDD Wrong Path-Pathetic Ecobusiness" which sums up their attitude, echoing the views of some international NGOs which have been sceptical to REDD+ for a number of reasons, for example: rejection of the market and commodification of nature; scepticism towards private property rights; objection to the poor solving the problems of the rich; and concern about diversion of attention from the real villains, who are not the poor but large-scale commercial operators [84].More supportive of REDD+ is KKI WARSI, in South Sumatra, with their primary focus on the forest rather than the indigenous people [85].An intermediate position is taken by AMAN (Indonesia's Indigenous Peoples' Organisation).Their slogan "No Rights, No REDD" indicates conditional support, and reflects their primary focus on the people, and an emphasis on land tenure as the key issue [84].WALHI's open, public criticism of REDD was manifested very clearly by their criticism of the huge Australia-supported Kalimantan Forests and Climate Partnership (KFCP) project whose aim was to repair the damage done to the forest by the 'mega-rice' project.So great was the criticism that the project was abruptly terminated.
A study of Indonesian media coverage by CIFOR also reveals "a broad range of stances by NGOs" [86] (p.17), from advocates to adversaries; with WALHI identified as a particularly strong critic.Such a varied response-as evidenced by WALHI, AMAN and KKI WARSI-is not an uncommon situation where NGOs are faced with public policies that may be controversial.Some work 'inside'-co-operating with (some would say co-opted by) the government-while others work outside, as critics.A sort of 'division of labour' is established.What is novel in Indonesia is the sheer extent of public participation-often critical-which would not have been possible under the Suharto regime.Whatever else, Norway's support to REDD+ in Indonesia has strengthened the hand of NGOs.While it might be claimed that some have moderated criticism of REDD+ because of the benefits of Norwegian funding, others certainly have not.NICFI/KOS did not have (and chose not to establish) the capacity to become involved in detail in the planning and implementation of REDD+ in Indonesia [60].The Norwegian Embassy in Jakarta had until recently only 1-2 staff members employed to deal with REDD+; one of whom, at a crucial period, was a former RFN staff member.NICFI has thus been very dependent on RFN, both for advice and also by virtue of their contacts with Indonesian NGOs.RFN have, on their side, had to walk a difficult middle path: supportive of REDD+ while aware of, and when appropriate pointing out, its potential shortcomings; collaborating with, and supporting, local NGOs some of which are very critical.Their policy has been to pursue a "dual goal of protecting the rainforest and securing the rights of its inhabitants" [87] (p. 12, italics in original).Strengthening the rights of local people over the forest may be seen as both an end in itself and the most effective way of limiting deforestation.Largely for this reason, certainly in Indonesia, there has been a notable shift of focus in REDD+, from the trees to the people [88].Additionally, the organisations which have been involved in REDD+ have been not only environmental NGOs but also those concerned with human rights, such as AMAN and HuMa (Perkumpulan untuk Permbaharuan Hukum Berbasis Masyarakat dan Ekologis-Association for Community Based and Ecological Law Reform) (which played a central role in bringing about a crucial ruling of the Indonesia Constitutional Court regarding the Revised Forestry Law).
RFN, together with AMAN, has defended NICFI against criticism in REDD-Monitor, the very influential website based in Indonesia.In a joint opinion piece in Development Today (26 March 2014), reposted in REDD-Monitor (27 March 2014) two staff members from RFN and AMAN write: "We would argue that more has been achieved in forest protection and the rights of indigenous peoples in Indonesia since the signing of the Indonesia-Norway agreement than over the course of the previous 15 years".
In summary, Indonesian NGOs have been involved with NICFI as both collaborators and critics, and Norwegian NGOs, and most particularly RFN, have played a number of different roles: as policy design advisers, implementers (if only indirectly through their collaboration with local NGOs) and critics.NICFI has been extremely reliant on their expertise, and has provided very substantial funding.The challenging task of combining these very different roles has been alleviated by two main factors: one, a close relationship of mutual trust between RFN and NICFI; second the policy summarised as 'no rights, no REDD' which has made it possible to support REDD+ on a provisional basis-the same slogan, and conditional support, adopted by AMAN, the major Indonesian NGO.

Tanzania
The agreement with Tanzania was Norway's first REDD+ bilateral agreement.A Letter of Intent (LoI) was signed by PM Stoltenberg and Solheim on April 21 2008, during their visit to Tanzania.By comparison with the agreements with Brazil and Indonesia, the (LoI) on a Climate Change Partnership on April 21 2008, places more emphasis on "REDD-Readiness", though also aiming at the further purpose to "implement programmes on adaptation and mitigation of climate change" [89] (p.xiii) (By comparison, Brazil was the most 'ready to go' with Indonesia as an intermediate case).The Norwegian government promised 500 million NOK (83 million USD) over five years and the first contract, for the development of a national REDD+-strategy, was signed in 2009.
The project idea was actually developed at the Norwegian Embassy in Tanzania in early 2008 before NICFI was established, following discussions between one of the members of staff and a local NGO.It continued to be mainly managed through the Norwegian Embassy in Dar es Salaam, the main driver of the project.REDD+ activities in Tanzania have been hugely dependent on Norwegian funds.As noted in the mid-term review [89] REDD+ policy development was entirely financed by NICFI, as well as all activities of the REDD+ task force and Tanzania's REDD+ secretariat.This funding was used mainly for three purposes: establishment of pilot projects, by NGOs; support to policy and planning, mainly through the local Institute of Resource Assessment, based at the University of Dar es Salaam; and a large number of research projects, mainly based at Sokoine University of Agriculture (SUA) in Morogoro (NICFI is also the biggest financier of UN-REDD in Tanzania with a budget allocation shared between UNDP, UNEP and FAO).However, as the subsequent review found [90], there has been a lack of 'national ownership' which has hindered progress; whether this is because of, or despite, Norway's substantial funding is open to debate.The evaluation also states: "NICFI support has been the driving force behind the national readiness process, but there is no evidence yet that the Tanzanian government is able to commit to building sustainable and permanent institutions without continued external support, for example from Norway" [91] (p.318).
The decision to prioritise pilot projects is largely explained by the pressures on NICFI to demonstrate early results [91] (p.313).And the main reason for giving the task of implementation to NGOs, effectively by-passing government to a very large extent, was the issue of corruption.It was particularly problematic for the Norwegian Embassy that at the time of the REDD+ negotiations they were involved in a dispute concerning the relevant government body, the Ministry of Natural Resources and Tourism, over a major environmental programme that the Embassy had supported over many years (After a long and thorough audit, the Ministry repaid Tsh 2,802,480,600 (2 million USD) [92].
In contrast to the Brazilian and Indonesian projects, neither RFN nor other Norwegian NGOs were involved in the Tanzanian REDD+ programme.But NGOs with a base in Tanzania have been very important in its implementation.NGOs in Tanzania were invited to submit proposals, and the nine pilot projects were selected from 46 applications, evaluated by the National REDD+ Task Force and the Embassy.The NGOs were primarily Tanzanian, often with international links: involved in conservation, for example Tanzania Forest Conservation Group (TFCG), and Mpingo Conservation and Development Initiative (MCDI); several were wildlife organisations, but also included were, for example, Tanzania Traditional Energy Development and Environment Organisation (TATEDO), and CARE.
The relationship of these NGOs to Norway was, in this case, simply that of a sub-contractor: they were hired to implement a set of pilot projects [93,94].They were not engaged as advisers-as in the case of RFN in Brazil for example-although the contract with one of the NGOs, which may be regarded as typical, did include, in addition to implementing the pilot project, 'advocacy at national and international levels' [91] (p.326).Also, the NGOs were quite active in giving comments on the draft national REDD+ plan (National Strategy for REDD+: draft December 2010, final version March 2013).While not critical of the initiative, they did express strong views about certain issues, most notably advocating a project-based, 'nested' approach over a centrally controlled national fund.
As noted, Norwegian support included a major funding of applied research, primarily involving Sokoine University of Agriculture in Tanzania and the Norwegian University of Life Sciences in Norway (This Section 4.4 is based in part on this programme, some of the results of which are forthcoming and some recently published in [95]).Whereas Norwegian NGOs have not been much involved in REDD+ in Tanzania, staff from these universities-in Tanzania and Norway-have to some extent provided advice.Progress with REDD+ has proved disappointing, and now that the nine pilot projects have come to an end, the scale of Norwegian funding has been reduced.
In summary, the Tanzanian case differs considerably from those of Norway, Brazil, and Indonesia.NGOs in Tanzania, with substantial funding from Norway, have played a modest advisory role, but have been the key players in implementation of the pilot projects, which have now come to an end.They have not expressed public criticism, but they have themselves (in one case) been criticised in Norwegian media, as described in the Norway section above.

Analysis and Discussion
Overall, the evidence suggests that NGO/government relations in Norway constitute a rather special case-not least because NICFI originates with an NGO proposal, and KOS and NICFI more generally has been very dependent on policy advice.Norwegian NGOs have co-operated closely with the government authorities from the very start-including policy design-and regularly advised on overarching policy issues, such as safeguards and portfolio development.Confrontation-in the form of public criticism-has to a relatively small extent occurred in Norway, and then towards certain actors (mainly political leadership), rather than towards NICFI/KOS, Norad or the MFA, i.e., their biggest funders.Furthermore, to the degree that there actually has been public criticism from NGOs, this has mainly been about the "big issues", such as cutbacks in the NICFI budget and Norway's official REDD+ position in the UNFCCC negotiations, and not implementation issues "on the ground" in the recipient countries.But there are also examples (most recently on 19 February 2017) that RFN (lately together with Greenpeace Norway) has voiced harsh public criticism towards NICFI-funded activities in the Democratic Republic of the Congo [96].RFN has also expressed public criticism against NICFI-sponsored activities in Guyana [97].
However, regular meetings with government authorities have provided the NGOs with ample opportunities to voice points of contention, closed off from the public arena.Precisely because NGOs have had this opportunity, they have not had the same need to go public when disagreeing with NICFI/KOS.Issues of contention-for instance on how strictly "carbon-focused" NICFI should be-have, with a few exceptions, been discussed mainly in private, which has been a very conscious choice by the NGOs.In fact, the NGOs seem to perform a sort of "stage management" [98], carefully considering what issues to sort out backstage (in the form of private criticism) or frontstage (public criticism).Put differently, what is de-politicised ("cooled down") and sorted out backstage vs. what is politicised ("heated up") [30] on the public scene is by no means coincidental, rather the result of careful choices by the NGO concerned.For example, Norwegian NGOs have experienced that private criticism, including letters to one or several ministers, may be more effective than harsh public criticism in the media.The following quote summarises RFNs position: "Løvold says that RFN's main task is to protect the rainforest and indigenous peoples' rights, and therefore it is much more important for the organisation to have real influence than to, as he puts it: 'In periods of powerlessness, yell out in the press'.Løvold believes that Norway has done a good job in preventing deforestation, and therefore it would be wrong and inappropriate to come forth as a harsh critic in the media" [41] (p.37).Thus, NGOs' roles, or more precisely choices regarding what roles they take, have clear entrepreneurial aspects to them.By this we mean that the NGOs are highly reflexive and aware of their possible operating space-including gains and losses by taking on different roles.Guided by their core objectives, they carefully choose their strategy, including how they frame and draw attention to issues ("issue entrepreneurship", see [3] for a more thorough description), and proposing policy solutions (i.e., policy entrepreneurship [99]).
It is interesting to compare our account of REDD+ and how Norwegian NGOs have dealt with the issue of carbon capture and storage (CCS).In the latter case, implementation of Norway's CCS policy has been strongly criticised in public by certain ENGOs, although the same ENGOs initially were instrumental in establishing the Norwegian CCS policy [100].This fierce criticism stands in stark contrast to NGOs relationship to NICFI and REDD+.However, unlike in the NICFI policy process, ENGOs have gradually become more marginalised in the CCS implementation process, as they have become less important in supplying technical knowhow throughout the policy process.This suggests that inclusion in implementation processes may dampen NGOs' public criticism, as inclusion provides them with an effective means to influence the policy process.Conversely, less inclusion may leave public criticism the only viable route to influence and if NGOs are gradually excluded, public criticism may soar, as exemplified by the CCS case.
In view of the massive scale of the REDD+ budget, it is noteworthy how minimal the public debate in Norway has been [63].Aid is often criticised in the media-but mainly for not yielding projected results [101].Criticisms could have been levelled at REDD+, not only on these grounds, but across a wide range of issues-such as adopting market-based incentives, inadequate attention to the views and interests of indigenous people, and hasty and risky decisions without sufficient knowledge bases.One explanation for the lack of debate may indeed be because it is usually NGOs that are active in promoting such debate.
The weight of the evidence suggests that a dual strategy of co-operation/confrontation has been possible in Norway, although the NGOs have favoured co-operation more than public confrontation.Norwegian NGOs have experienced that sustained dialogue with the government-including private criticism-yields effective results.The reasons why Norwegian NGOs have comparatively large scope for combining different roles is a combination of the long-established 'corporatist' "Norwegian government/NGO model" (see Grendstad et al. [13], Steen [36]), and the special case of REDD+ where government has been so dependent on NGO knowledge and networks.
It is worth noting, however, that some development NGOs have been more critical in public than ENGOs, for instance regarding issues related to implementation "on the ground", and questioning the "crowding out" effect that the NICFI scheme has on poverty reduction [39] (p.16).ENGOs have been comparatively silent on these issues, and instead focused their public criticism on more overarching policy issues.The fact that ENGOs have received notably more NICFI funding than development and humanitarian NGOs suggests that funding may guide the choice of whether or not to adopt a public confrontation strategy on implementation issues.The same pattern can be observed in Brazil.
Moving to the other countries, the nature and extent of NGO involvement has varied in part because of the different approaches taken by NICFI/KOS in each case.In Brazil, where the state was viewed by Norway as a competent and trustworthy partner, REDD+ was designed as a national programme-for which responsibility was largely delegated to the Brazilian state; and very considerable funds have been allocated for use by NGOs.By contrast, in Tanzania, Norway was much more sceptical to the government apparatus and chose to render its support mainly in the form of pilot projects.For this reason local NGOs benefited greatly since they were given the task of implementing these pilots.Indonesia constitutes an intermediate case, where support is given to a combination of national activities and pilot projects at regional level.Although the amount of funds allocated to Indonesian NGOs was relatively less, compared with Brazil and Tanzania, the support to NGOs has played a very important part in increasing their political influence.
In Brazil, Indonesia and Tanzania non-Norwegian NGOs have not been very confrontational.However, there is some variation both between and within countries.In Brazil, some NGOs have identified important issues with the design and implementation of the Amazon Fund, especially in relation to the lack of strategic vision and the apparent low efficacy of the fund in promoting short-term deforestation reductions.Yet, Brazilian NGOs were facing much more pressing issues such as the dismantling of key aspects of the Forest Code in 2012 and the construction of Belo Monte and other large hydroelectric dams.Therefore, it is understandable why many members from NGOs perceived the Amazon Fund as an overall positive initiative that could be improved more by engagement from the inside instead of subjecting the fund to public criticism that could risk the withdrawal of donor support.In Indonesia, even some of those that have received funding have been critical.Some major NGOs-both environmental and human rights-have indeed been very critical of REDD+ as an initiative.Additionally, they have also been critical of specific projects and programmes dependent on financial support from NICFI.Criticism has been expressed both in the media and also in public meetings organised to inform about and promote REDD+.In Tanzania, NGOs have only to a minimal degree expressed criticism.
Against this background, we may briefly draw some comparisons.Co-operation in policy design: In all four countries NGOs have, to varying extents, co-operated with government in the policy design of the initiative.(In Tanzania, the Norway-supported REDD+ initiative consisted primarily of project-based activities, though one NGO gave input to the initial project idea).
Co-operation in implementation: Here, Norway is the exception, for the simple reason that REDD+ is not implemented in Norway.In all the other three countries, NGOs have certainly co-operated.This has in the cases of Brazil and Indonesia sometimes been contracted via local NGOs, sometimes in partnership/co-operation with Norwegian NGOs; and sometimes contracted by the respective government, using Norwegian funds.In Tanzania, NGOs functioned as direct sub-contractors to Norway Confrontation: Here the picture becomes more complex.In Norway, ENGOs have expressed some public criticism, but then mainly regarding overarching policy issues, and not towards their main funders.On the other hand, Norwegian NGOs have expressed substantial private criticism.The case of Brazil is similar: it seems clear that the NGOs have prioritized co-operation above confrontation, and instead expressed private criticism.The experience in Indonesia is mixed.The simplest case is Tanzania, where criticism by NGOs has been minimal.
The discussion in this section is summarised in Table 2 below.

Conclusions
Overall, the evidence across the cases suggests that NGOs have prioritised co-operation above confrontation.However, although public criticism has been limited, it has not been entirely absent and in all cases, perhaps with Tanzania as the exception, NGOs have expressed private criticism.In sum, this seems to suggest that "the Norwegian government/NGO model", as presented by Grendstad et al. [13] and Steen [36], has to some extent been exported to the other countries, particularly through the NICFI funding scheme for civil society.The point of Dryzek et al. [11] about trade-offs between inside and outside strategies seems highly relevant in all the cases.There are also good reasons to extend the argument by adding that NGOs are highly aware of and reflexive regarding their possible operating space, and perform trade-offs in an entrepreneurial manner, focused on results.
Close co-operation seems to correlate with limited public confrontation.Largely in line with the position of Dryzek et al. [11]-the evidence suggests that the relative levels of funding and co-operation do influence NGO's choices regarding levels of and channels for criticism (public vs. private).For instance, Friends of the Earth International and Climate Action Network (CAN) are less involved in REDD+ policymaking processes than their Norwegian counterparts, but voice more public criticism towards REDD+.
Limited public criticism does not, however, necessarily mean that criticism has ceased; it may simply mean that criticism instead is expressed in settings away from the public arena.Particularly the evidence from Norway and Brazil suggests that instead of publicly criticising a global initiative that they largely support and have huge stakes in-and thus put the initiative as a whole at risk-NGOs to a larger degree have used other more informal channels to voice points of disagreement.NGOs' opportunity to resist being co-opted is probably greater in this instance than is usually the case-because the interests of NGOs and NICFI/KOS are to a very large extent congruent, and NICFI/KOS is so dependent on NGO services.
Choosing private channels for criticism may be distinguished from co-optation.A clear pattern is that NGOs knowingly trade away some public visibility in exchange for having direct contact with policymakers (both bureaucrats and elected officials) and "a place at the table"; this has been the price to pay for achieving policy impact-the ultimate goal for most NGOs-which in this case has been very substantial.If, however, NICFI initiates policies conflicting with NGOs views, there are examples that NGOs go public, such as in the case of Norway's position on REDD+ in the UNFCCC, indicating they are not co-opted, at least not silenced.
Close co-operation and financial dependence does indeed increase the risk of co-optation.However, we find that the NGOs are very much aware of their possible operating space and pitfalls, and navigate strategically and pragmatically to pursue their goals in an entrepreneurial manner.This point becomes clear when we observe what issues the NGOs publicly criticise-or not.Acting in an entrepreneurial manner still implies making tough choices to navigate the fine line between co-operation and co-optation; deciding when and how to play a critical role-in public or in private.
Returning to the title of the paper-whether the relationship between NICFI and NGOs is best described as an instance of co-operation or co-optation-this analysis has revealed a much more nuanced picture.Instead-drawing on experience from several cases relating to one specific government initiative-we believe that our case is revealing of an entrepreneurial, pragmatic and largely effective NGO approach which may be summarised in the phrase result-oriented pragmatism [102].
Thus, we support Soneryd's [103] call to take an agnostic scholarly approach when analysing public involvement in policymaking, including NGOs.NGO practices and choices clearly have entrepreneurial aspects to them, which are best understood by employing the sort of empirically driven study that Yearley [12] advocates.

Figure 1 .
Figure 1.Projects funded by the Amazon Fund categorised according to the nature of the party responsible between 2008 and 2015.Data from BNDES (Brazilian Development Bank).

Figure 1 .
Figure 1.Projects funded by the Amazon Fund categorised according to the nature of the party responsible between 2008 and 2015.Data from BNDES (Brazilian Development Bank).

Table 1 .
A comparison of Norway, Brazil, Indonesia, and Tanzania across key dimensions related to Reducing Emissions from Deforestation and Forest Degradation (REDD+) and Norway's International Climate and Forest Initiative (NICFI).Forest data from Global Forest Watch.Data on land area, population and governance from Wikipedia.

Table 2 .
A schematic comparison of different roles NGOs have taken regarding NICFI in Norway, Brazil, Indonesia and Tanzania.