The primary sources of start-up funding for the initiative were the Vale Fund, Bank of America and the Amazon Fund. Other sources of funding include USAID, the Norwegian International Climate and Forest Initiative, the British Embassy, the Anne Ray Charitable Trust and the Moore Foundation. Because of the perceived risks of relying on voluntary carbon markets for financing, the proponents decided to only sell credits in a regulated market under the UNFCCC. For this reason, they do not plan to certify the initiative under any voluntary market scheme such as VCS. Instead, they aim to obtain additional financing through economic instruments adopted by the Brazilian Government under its REDD+ strategy (still under development).

6.1.3 Motivation

Federal, state, and local governments have historically incentivized deforestation in the state of Pará by promoting activities that are the driving forces behind forest loss (Barreto et al. 2005; Ferreira et al. 2005). However, since 2004, the federal government has also made reducing deforestation a policy priority. As a result, Pará and the municipality of SFX have been under considerable pressure to reduce deforestation. The policy mix to achieve this goal includes measures such as: (i) the Programa Municípios Verdes (Green Municipalities Program) and (ii) the Municipal Embargo.

Municípios Verdes was launched in 2011 and, according to the Governor of Pará, it helped reduce deforestation levels from 6000 km2 to 3000 km2 in its first year (Pará 2012). One requirement for municipalities to participate in the program is to create a pact for achieving zero illegal deforestation by 2020. SFX’s local government signed the municipal pact with diverse local representatives and producer associations in 2011. It also established a commission under the pact to serve as a forum for inclusive environmental governance. The involvement of smallholders in the commission has been a crucial step, given their exclusion from previous political processes (Schneider et al. 2013). The commission’s goal is to define a long-term agenda to sustain anti-deforestation efforts in the municipality. However, it lacks a clear decision-making process and financial resources, given that it currently relies on funding provided by TNC.

The most effective action for reducing deforestation in SFX has been the federal blacklist for embargoed municipalities. Since 2001, SFX has topped the list of municipalities in the Brazilian Amazon for rates of deforestation, and in 2008 it was included in the Ministry of Environment’s blacklist of embargoed municipalities. Since then, all farmers in the municipality have faced restricted access to credit and other sanctions. According to the former municipal Secretary of the Environment, these sanctions include constant monitoring and inspection of land-use operations, increased enforcement, and restrictions on issuing licenses for activities with environmental impacts. The consequences of being blacklisted were quickly felt in the municipality as expansion of cattle pastures nearly came to a halt and deforestation fell significantly.

To be removed from the blacklist, municipalities are required to reduce their rates of deforestation and register 80% of private properties in the CAR system, which includes geo-referenced property information, serves as a first step towards clarifying tenure and allows landholders to access subsidized credit. The embargo led to implementation of the CAR system in SFX, with support from TNC. More than 80% of private properties have been registered. Another outcome of the embargo was that the national government was able to share responsibility for reducing deforestation with the municipal government and private actors, which was crucial to improving governance and reducing deforestation. In 2012 and 2013, deforestation rates started to rise again (PRODES/INPE 2014) suggesting that other measures may be needed to reach a zero deforestation goal over the long term.

TNC was the first organization to involve local actors in SFX in actions related to REDD+. Beginning in 2009, the pilot program focused on financial and technical support for CAR implementation. Due to the lack of state resources and institutional capacity for CAR implementation, TNC supported the process with a USD 19 million grant from the Amazon Fund (distributed among 12 Amazonian municipalities), and financial resources from USAID and the Vale Fund.

Analysis by TNC (Balieiro 2013) showed that deforestation in 2011 and 2012 increased in areas registered in CAR and in land reform settlements. Just 9% (2011) and 0.6% (2012) of deforestation occurred on indigenous lands and no deforestation happened inside conservation units. This reflects a recent pattern in the state of Pará where deforestation has been particularly high in land reform settlements (Brandão et al. 2013). Many of these settlements were established by the government in the 1970s and 1980s and are now abandoned; smallholders live there largely without government support. Laws against deforestation are difficult to enforce in these areas, because INCRA (a federal government agency) owns the land, rather than individual households. Moreover, local actors do not have any sort of incentive to change their land use. They feel helpless and blame the national government and INCRA for not helping them develop land-use alternatives and land management plans.

Deforestation in the settlements is a result of smallholder livelihood strategies, with higher deforestation in areas where cattle ranching is the primary land use (Pacheco 2009). Thus, improvement in tenure security alone, as provided by registration in CAR, will not assure conservation by smallholders (Gould 2006; Robinson et al. 2011). In fact, deforestation may initially increase when land rights are secured, since subsidized credit for ranching is liberated for properties in CAR. A recent study concluded that CAR alone was ineffective in reducing deforestation in Mato Grosso and Pará from 2008 to 2012 (Azevedo et al. 2014). In this sense, it is important to combine CAR registration with other incentives for landowners to meet environmental criteria. For instance, given the importance of cattle ranching in the local economy, TNC is promoting both improved pasture management and alternatives to draw people out of cattle ranching.

6.1.4 Timeline

The initiative has evolved since its conception, changing its name and intervention area more than once. According to TNC, the REDD+ nomenclature was misunderstood at the local level. For small farmers, the market-based connotation of REDD+ was of little interest; for large producers it seemed like an opportunity for profit; and for indigenous groups it tapped into anti-REDD+ sentiments. Because of these misunderstandings, TNC eliminated the term ‘REDD+’ from the name of the initiative even though actions to reduce deforestation remain at its core. Figure 6.2 summarizes the main events and interventions of the initiative to date.

6.2 Strategy for the initiative

The pilot program combines command-and-control, economic incentives and information-sharing strategies. The specific strategies are land-use zoning; improved enforcement and compliance with environmental legislation; sustainable finance and management for indigenous and protected areas; sustainable production alternatives for local actors; technical assistance and promotion of alternative livelihoods; restoration of degraded lands; enhanced participation of vulnerable groups in REDD+ related decision making; economic opportunities; and shared lessons (Table 6.2).

These strategies were designed to support local actors, especially smallholders, in the transition towards a low-carbon economy. As argued by Schneider et al. (2013), many smallholders have the will to reduce deforestation but they lack the technical and financial means to make the transition from extensive cattle production. To reduce deforestation, they need technical support to adopt agricultural practices that are more profitable on smaller areas of land, including alternative commodities. While the focus of the pilot program is to reduce deforestation and forest degradation, co-benefits of the initiative include reduction in poverty and social inequality, along with enhanced environmental governance.

One important recent achievement in June 2014 was the launch of the Terra Verde Fund, which will channel support for sustainable production and REDD+, with increased autonomy for local actors (TNC 2013b). The fund is intended to provide a less bureaucratic and more decentralized process for accessing resources, when compared to existing state and national mechanisms, such as the Amazon Fund. It was created by TNC and the Brazilian Biodiversity Fund (Funbio) through a participatory and inclusive approach in 2012–2013 and will be managed by local organizations. In the stakeholder meetings, local actors expressed the importance of meeting social, economic and structural needs through differentiated benefits in line with the already established municipal pact for reducing deforestation. The pact commits the local government to fulfilling its obligations by granting licenses, providing technical assistance, facilitating the land tenure clarification process, increasing access to credit and providing infrastructure. Simultaneously, it commits rural producers to comply with environmental regulations and to adopt sustainable models of production. The Terra Verde Fund is expected to provide financial support for achieving these goals, while the pact acts as an informal local contract.

It is important to note that both performance-based and up-front benefits are part of the pilot program. CAR, for example, is an up-front benefit that positions actors to receive performance-based benefits, such as access to the Terra Verde Fund and technical assistance for implementation of best management practices. Performance-based benefits will be based on progress in reducing emissions (Griscom and Cortez 2011). The proponents are still deciding whether to offer PES as a positive incentive, since opportunity costs are high in the area, and they want to prioritize benefits that help local actors make the transition to a low-carbon economy.

TNC is still collecting data and producing a forestry inventory to define reference levels for the area. Griscom and Kerkering (2010) estimated that 34.6 million tCO2e were emitted from deforestation and degradation each year from over 16 million ha of forests within the SFX accounting area. The majority of these emissions (~90%) were from deforestation, and the remaining 10% from forest degradation, largely due to logging. As much as 80% of annual emissions from deforestation and degradation came from the 9 million ha that constitute the northern region of SFX. In this region, deforestation is highest (3.8%) in land reform settlements and second highest on private lands (3.2%). Deforestation was relatively low in conservation units (0.18%); the lowest rates of deforestation occurred on indigenous lands (0.06%).

6.3 Smallholders in the initiative

In four selected intervention communities, CIFOR-GCS surveyed a random sample of 124 families, including at least 30 in each community, or 13%–69% of the community populations. Information on community characteristics was also collected (Tables 6.3, 6.4 and 6.5). All interviews were conducted between September and October 2010.

Most people living in the sample communities2 came from other regions of Brazil, especially the central and southern parts of the country. Rising land prices in southern Brazil (relative to the north) were one catalyst for this migration. Because of this occupation process, land conflicts are an intrinsic characteristic of the area. Land tenure was considered secure in SFX2 and SFX3, as they are located in land reform settlement areas. Residents of the other two communities reported feelings of insecurity as they were informally settled on public lands without any land documentation (Duchelle et al. 2014).

The main reported causes of deforestation and degradation in the communities were: large-scale ranching (SFX1), small-scale traditional agriculture (all communities), small- and medium-scale ranching (SFX2, SFX3 and SFX4), small-scale illegal logging (SFX2, SFX3 and SFX4) and subsistence fuelwood collection/charcoal production (SFX1). Local land-use and livelihood dynamics reflected a high reliance on livestock (especially cattle ranching) and agricultural production (Figures 6.3 and 6.4), which motivates promotion of better cattle ranching practices as an initiative strategy. Cattle ranching generates employment, and wage labor is a significant source of household income. Another important source of income is the Bolsa Família (family stipend) program, which has been credited with alleviating absolute poverty in Brazil; it is also blamed for the spread of clientelism and patronage and the growing dependence of Brazil’s poor on income transfers rather than on productive employment (Hall 2012).

Average annual household income in the communities was between USD 10,184 and USD 11,835 (Table 6.4), and most households considered their income to be sufficient for their needs (Table 6.4). In the two years before the survey, the number of households practicing permanent agriculture increased, while the number practicing swidden agriculture and the area used for swidden agriculture decreased in three out of four communities, due to stricter environmental regulations.

In all of the study communities, respondents in community meetings indicated that production of cocoa had increased over the two years prior to the survey. Reasons cited by respondents included farmers’ need to diversify agricultural production and the high income earned from cocoa. Local associations and NGOs have promoted this expansion of cocoa production, because it is produced in an agroforestry system and because it has the potential to be more profitable than cattle. Indeed, cocoa production is now the main alternative to cattle ranching in the municipality, with a recent price increase and TNC support for implementation of best management practices (O Globo 2014).

Households did not rely heavily on the forest and environment as an income source in any of the communities (Figures 6.3 and 6.4; Table 6.5). Households reported using forests mainly for collection of wild fruits (SFX1 and SFX4), poles and thatch (SFX2), logging (SFX2) and hunting (SFX2, SFX3 and SFX4), with the first two sets of products collected mostly by women and the last two by men. None of the households received forest-related PES.

Community members are normally organized in a local association and choose leaders by consensus to represent them in important meetings outside the communities. Some of them are also part of other local organizations, including religious organizations, cooperatives, and labor and rancher unions. The participation of smallholders in these networks and unions encourages them to have environmental awareness (Schneider et al. 2013). The role of women is still to be strengthened. Most women in the sample communities felt they were not well represented in decision-making processes concerning the community because of women’ s lack of interest and time spent on household activities. TNC aims to develop strategies for women’s empowerment, such as facilitation, promotion and articulation in the municipal commission and in local decision-making.

Smallholders are key to reducing deforestation rates in SFX. Those living in land reform settlements (SFX2 and SFX3) are responsible for high rates of deforestation and lack incentives and enforcement measures to change their land-use practices. Together with large-scale farmers, they are the group that is expected to make the greatest contribution to reduced carbon emissions by changing their land or forest use. The mix of strategies designed by the pilot program reflects the diversity of smallholder needs that must be met in order for them to change their land- and forest-use practices. Examples include the dissemination of better practices for cattle ranching, promoting more permanent agriculture (i.e. cocoa production) and facilitating environmental compliance through CAR implementation and restoration activities. Because these benefits focus on ‘asset building’ (Pirard and Treyer 2010), their performance is more challenging to measure than that of ‘use restricting’ benefits.

6.4 Challenges facing the initiative

According to the proponent, a mix of measures to reduce deforestation – beyond command and control – is important in SFX given the diversity of actors involved. Yet this creates the challenge of coordinating these different measures, tailored to motivate different actors to change their BAU land-use practices. The Terra Verde Fund is designed to address this challenge by supporting a multidimensional approach to benefit-sharing, instead of a one-size-fits-all approach (see Gebara 2013).

Based on experience with the municipal pact, TNC believes that the greatest challenge facing the initiative is land tenure clarification, since this depends on the will of government agencies. Tenure clarification has been considered key to implementation of both regulatory and incentive-based REDD+ mechanisms (Duchelle et al. 2014), and is central to many discourses about REDD+, e.g. ‘tenure first then REDD+ second’; ‘no rights, no REDD+’ (Brockhaus and Angelsen 2012). In the case of SFX, however, land tenure clarification may inadvertently support ranching and other activities that drive deforestation by offering access to benefits that require land titles, such as public subsidies for ranching and agriculture.

The SFX initiative also faces challenges that are common to many REDD+ initiatives, such as the lack of state and municipal policies to specifically regulate REDD+, corruption and limited governance capacity.

6.5 Lessons from the initiative

This initiative provides insight on the trade-offs involved in engaging diverse local stakeholders, including smallholders. Relying on existing institutions to engage stakeholders in REDD+ initiatives can increase legitimacy. However, existing institutions often need to be adapted or strengthened in order to be effective. To overcome path dependency and ‘stickiness’ (Baumgartner et al. 2011), the creation of new institutions and the introduction of new actors, such as TNC, can also help (Brockhaus and Angelsen 2012). Local actors should be engaged early on and eventually lead the process in order to mobilize broad support. However, effective engagement of local actors is a costly and time-consuming endeavor and requires strong partnerships to be implemented successfully.

An enabling policyscape that includes a mix of regulatory, economic and informational measures greatly facilitates buy-in from local actors, including their agreement on a common objective. With this agreement, actors may work together successfully despite past or current conflicts. Diverse stakeholders with conflicting interests require diverse incentives. Therefore, there should be an early focus on what types of benefits motivate which actors (Gebara 2013). Promoting sustainable landscapes and low-carbon strategies comes with the challenge of offering sufficiently large portfolios of technological alternatives, allowing smallholders to choose from diverse livelihood strategies within the same intervention.

The ‘asset-building’ strategy and investment in sustainable land uses may be the most appropriate approaches to move a region onto a sustained low-carbon development path. However, it is more difficult to monitor and control performance in these approaches compared to a single-use restricting payment scheme. For example, the proponents must ensure that actors who benefit from different strategies are those who actually stop deforesting. Experience with CAR in SFX highlights the possibility that benefits might inadvertently result in increased forest clearing. While CAR enables improved environmental monitoring, it also allows access to subsidized credit for cattle ranching. In SFX, the proponents have responded to this concern by disseminating information about ‘low-carbon credit,’ such as through the federal program for low-carbon agriculture (ABC program).

TNC believes that despite the challenges of working at a large scale, critical on-the-ground problems are linked to a wider context that encompasses different actors, interests and jurisdictions, and thus a large-scale approach is necessary to achieve transformational change through REDD+. Given that the REDD+ arena is characterized by a multitude of actors at different levels that operate within existing institutions, interests and ideas (Brockhaus and Angelsen 2012), such an approach makes sense to generate lessons and overcome complex challenges.

Finally, the SFX initiative shows that REDD+ is losing some of the initial characteristics that made it such a novelty, as argued by Angelsen and McNeil (2012). The initiative is now targeting broader and multilevel goals, rather than focusing exclusively on climate change mitigation and results-based funding for reduced deforestation. This mitigates the risk of losing financial support if carbon funds do not materialize. While such broader strategies can be advantageous, broadening the focus too much can present a barrier to adopting the performance-based measures that were originally at the heart of REDD+ (Sunderlin et al. 2014). The enabling policyscape for REDD+ needs to be shaped by strong measures, such as fundamental policy reforms, to remove the incentives that drive deforestation. This will allow transformational change that will outlast REDD+.

6.6 Acknowledgments

I am deeply grateful to the families of SFX and surrounding areas for their patience and generosity while sharing their time and knowledge. Local organizations, especially ADAFAX and SEMMAs, also played a key role in enabling this research. Angelica Toniolo, Rane Cortez and Edenise Garcia, together with other members from TNC, were crucial and always accessible in sharing their knowledge and their experiences with the initiative. Finally, special thanks to the 2010 field research surpervisors: Gisele Aparecida Monteiro and Leonela Guimaraes, and to the members of the field team for their outstanding contribution: Rosiany Miranda, Edson Silva, Tatiana Izidoro, Hanoica Caceres and Izabel Marinho.