The initiative known as Making REDD Work for Communities and Forest Conservation in Tanzania is implemented by the Tanzania Forest Conservation Group (TFCG) and the Tanzania Community Forest Conservation Network (MJUMITA). Its aim is to demonstrate how emissions from deforestation and forest degradation can be reduced through CFM (TFCG and MJUMITA 2009). While the initiative includes a suite of interventions at the community scale, it makes direct payments to individuals for the environmental services provided through reduced deforestation and forest degradation.

14.1 Basic facts: Where, who, why and when

14.1.1 Geography

The initiative includes community forests in two districts: Kilosa in the Morogoro Region and Lindi Rural in the Lindi Region (Figures 14.1 and 14.2). Kilosa District (population 488,191 in 2002) is in the Eastern Arc Mountains and has both mountain forests and miombo woodlands. Lindi Rural (population 214,882 in 2002) has miombo woodlands, coastal forest, regenerating forest and coastal scrub.

The intervention area covers 1850 km2; it has 12 villages in Kilosa, 2 villages in the Mpwapwa District and 10 villages in Lindi Rural. The participating villages in Kilosa are at higher elevations (678 to 1209 masl) than those in Lindi Rural (125 to 295 masl). The proponent estimated a 0.7% annual deforestation rate in Kilosa during 2000–2006. Lindi Rural has a higher rate of forest loss per year at 1.9% (MJUMITA 2014).

Of the 24 villages participating in this REDD+ initiative, CIFOR selected three1 in Kilosa and four in Lindi Rural for the CIFOR-GCS sample (Figures 14.1 and 14.2). We held meetings to implement the village survey in both districts, but only surveyed households in Kilosa. There were 993 households (3686 people) in the three selected villages in Kilosa, from which we drew a random sample of 90 households.

Agriculture, including livestock, employs about 85% of the labor force in Kilosa District (URT 2007a). The most commonly held livestock in Kilosa are chickens, followed by cattle, goats, sheep and pigs. Other important economic sectors are small businesses and natural resource extraction, such as forestry and fishing. Agriculture is also the most important economic sector in Lindi Rural, followed by forestry and tourism. Major cash crops grown are cashew nuts, sesame and coconut, while cassava, rice, sorghum, maize and yams are important subsistence crops. The most common livestock kept in Lindi Rural are poultry, ducks, goats, sheep and cattle (URT 2007b).

14.1.2 Stakeholders and funding

The two lead proponents, TFCG and MJUMITA, are both Tanzanian NGOs. MJUMITA is a network of community groups involved in PFM. Other collaborators include government agencies (e.g. Department of Environment, Forest and Beekeeping Division of the Ministry of Natural Resources and Tourism, and Agricultural Research Institute in Kilosa), universities (e.g. University of Dar es Salaam Institute of Resource Assessment, Sokoine University of Agriculture) and other NGOs and grassroots organizations (Tanzania National Resource Forum, Regional Community Forestry Training Center, Katoomba Group). The most significant stakeholders are the district administrations and the participating villages in Kilosa and Lindi Rural, including an independent MJUMITA network in each village. The primary sources of start-up funding were the RNE and the African Rainforest Conservancy. The initiative plans to sell carbon offset credits in the voluntary market.

14.1.3 Motivation

Kilosa and Lindi Rural were selected for the initiative because they are biodiversity hotspots and face multiple sources of deforestation pressure. The primary deforestation pressures are related to local livelihoods, including small-scale traditional agriculture, small-scale illegal timber harvest, brick manufacture for local construction, and both subsistence and commercial fuelwood and charcoal collection. Weak internal institutional arrangements and collective action challenges are prevalent in both sites, and lack of compliance and enforcement are key underlying drivers of deforestation and forest degradation (Dokken et al. 2014). In addition, fires represent a threat to the forest. The proponents expect that smallholders within the intervention area will continue to be the primary agents of deforestation and degradation in these districts for the foreseeable future.

14.1.4 Timeline

The proponents were awarded funding in September 2009 but the launch of REDD+ at site level took place in 2010. In 2011, the initiative laid the groundwork for conditional incentives by engaging the villages in discussion about land-use plans and regulation of forest use, and inventorying carbon stocks. In 2012, the initiative made the first trial payments to individuals in Kilosa. Figure 14.3 summarizes key events over the course of the initiative.

14.2 Strategy for the initiative

The objective of the initiative is to reduce deforestation and forest degradation, but biodiversity co-benefits are also a high priority for the proponents. Community development is another important goal. Both biodiversity and the potential for poverty reduction were considered in the selection of villages. These can also be important attributes of carbon offsets sold in the voluntary market, and the proponents plan to seek CCBA certification. They are in the process of finalizing the VCS and CCBA validation and verification for Lindi Rural (October 2014), while Kilosa is planned to start in late 2014. For the VCS, the proponents are the villages represented by their respective village chairs.

In the PDD, the reference level for carbon emissions will be 122,965 tCO2e/year for Kilosa. The reference level for Lindi Rural is based on a fixed deforestation rate applied to a decreasing forest area. Thus, the reference level for year one (2012 to 2013) is a loss of 126,560 tCO2e from the project area, but by year ten, the reference level declines to 95,765 tCO2e. The proponents are monitoring changes in forest cover using remote sensing with ground truthing. The proponents have already involved local people in the MRV work. Each village has a committee for measuring carbon stocks, and MJUMITA will serve as their communications channel to VCS.

To reduce deforestation and forest degradation, the proponents are undertaking a suite of interventions intended to restrict forest access, raise awareness about the importance of conservation and enhance livelihoods (Table 14.1). This initiative encompasses many of the different types of interventions that have been proposed for REDD+.

Most of these interventions are targeted at the village level. The five main criteria for selection of villages to participate in the initiative were: high potential for biodiversity co-benefits, high level of forest carbon, size of the forest, low profitability of deforestation (low opportunity costs) and clear land tenure. Additional factors considered included: significant threat of future deforestation/degradation, good governance and rule of law, potential for scaling up to similar areas, high poverty rates, high potential for community or poverty reduction co-benefits, strong partner organizations, and villages where the leader is willing to cooperate and carry forward the initiative.

While forest-use restrictions, education and training have all been organized at the village level, the initiative also offers payments at the individual level. In the pilot phase, the proponent allocated part of the funding from RNE to trial incentive payments to individuals. The funds for trial payments have been distributed as PES to individuals in the villages based on their actual reduced deforestation rate. Every household member is entitled to the payments, including up to three children – their payment is made to their mother. In village meetings, the community decides whether some of the money they have received should be paid to the community to support the construction or improvement of schools and health clinics. In 2014, these payments were paid in advance and financed by donor money. While the national framework for carbon credits is not yet clear, the proponents argue that communities should receive all of the revenues from the sale of carbon credits, except for any government tax and 5% retained by MJUMITA to cover their monitoring costs.

In addition to the proponents’ support for construction of village offices and for the interventions listed in Table 14.1, the study villages included in the CIFOR-GCS sample have received external support from the government and NGOs for improvements in agriculture, fisheries, education, public health, water supply and roads. There is no indication that any external support has been diverted from these villages because they are benefiting from the REDD+ initiative.

14.3 Smallholders in the initiative

The administrative level that coordinates and implements REDD+ in Kilosa and Lindi Rural is the village (kijiji). The seven villages studied by CIFOR-GCS are characterized in Table 14.2. In all of these villages, key informants identified the village government as the most important village decision-making body and indicated that the leader of the village government is elected. In some villages, all residents are invited to attend village government meetings, while only religious leaders are invited in other villages. Other important decision-making bodies mentioned by key informants are security committees, sub village committees and village land committees. We asked the women’s focus groups about their perception of women’s participation in community decision making and participation in forest-use decisions, rule making and monitoring at the village level. In both Kilosa and Lindi Rural districts, women participate in village decision making, but are less involved in forest-use decision making at the village level, even though they actively participate in forest activities on a weekly or even daily basis (Larson et al. in press).

Fuelwood collection and charcoal production undertaken by villagers is an important cause of deforestation and forest degradation in the area (Kajembe et al. 2013). Five of the villages report that they experienced a decrease in the forested area within village boundaries in the two years prior to our visit in 2010. All villages reported that the villagers were the cause of this change. Other causes listed are neighbors, forest fires and drought. Underlying drivers related to the decrease are lack of rules or lack of enforcement of existing rules, and the need for agricultural land. Three of the villages experienced a decrease in the forest quality over the same period, caused by villagers and neighbors. Similar to deforestation, lack of rules and/or enforcement is listed as a key driving force.

In each of the Kilosa study villages, we interviewed a random sample of 30 households (9% of the total number of households) in July 2010. The socioeconomic status of the sampled households is described in Table 14.3.

Agricultural crops are the dominant income source in Kilosa and Lindi Rural. From the household-level data collected in Kilosa, we see that income from agriculture and livestock dominates the average household income portfolio in all three study villages (Figure 14.4). Maize is the main staple crop and the main agricultural product (ranked by gross value of annual production) in KILO1 and KILO2, while it is the second most important crop in KILO3. Both KILO1 and KILO3 reported that villagers produced noticeably more maize over the past two years prior to our visit in 2010 because of increased maize prices, while KILO2 reported that they produced noticeably less due to drought. KILO1 reported that they were producing less beans due to pests, while KILO3 produced less groundnuts due to drought. In the village meetings in Lindi Rural, it is reported that production of maize has increased while groundnut and sesame have decreased in the two years prior to the interview.

Figure 14.4 shows that forest and environmental resources play an important role in the household economy in Kilosa. The average household in KILO2 and KILO3 derived more income from forest and environmental products (counting the value of both subsistence and cash sales) than from wage labor, household business or other income sources. Of the total income reported by respondents in Kilosa, 11% is derived from the forest and 6% from the environment (Figure 14.5), which are substantial shares, although still dwarfed by the 62% of income generated by agricultural crops.

The households collected a range of forest products. While women mainly collected fuelwood and NTFPs, men harvested poles and trees, made charcoal and hunted animals. From the household level data in Kilosa (Table 14.4), we see that all households relied on biomass as their primary cooking fuel. All but two households in the sample use the ‘three-stone open fire’ technology for cooking and fuelwood was by far the most important forest product, on average accounting for more than half the forest and environmental income for households in all of the villages. Fuelwood was also reported as the most important NTFP in Lindi Rural. Most of the forest income in Kilosa was subsistence income, but a few households also reported cash income from forest products (Table 14.4).

The smallholders living in the villages are considered the primary agents of forest carbon emissions in both Kilosa and Lindi Rural, and thus, are expected to make changes in order to reduce emissions. For example, once the study villages agreed on the areas to be designated as village forest reserves (VFRs), the village governments asked households living in those areas to move to areas within the village that were designated for agriculture. Depending on how this was carried out, including where and how much new agricultural land they were allocated, the relocated households may or may not consider the process to be fair. In all of the study villages, it was also reported that people in neighboring villages harvest resources from the new VFRs, and thus they will also be affected by restrictions on forest use.

14.4 Challenges facing the initiative

National REDD+ policy in Tanzania is a challenge for this initiative. To date (October 2014) the national legal framework and benefit-sharing mechanism for REDD+ has not yet been clarified. Further, community rights to forests and lands are weak in the 2012 draft of the National Strategy for REDD+ Policy, and the area within village boundaries is not recognized as village land (URT 2012). The proponents have made efforts to secure local tenure rights by addressing the issue at the national level (Sunderlin et al. 2014b), but the process of securing village land certificates in the intervention area has been both challenging and more time-consuming than planned. This might continue to have an impact on this initiative, particularly with respect to the potential for scaling up.

Among the other major challenges recognized by the proponents are national agricultural and investment policies, and general economic conditions, such as an economic recession and a weak forest carbon market. It is not clear whether REDD+ will be competitive with alternative land uses. The initiative’s model of REDD+ is built around performance-based payments to individuals in the communities to incentivize them to avoid deforestation. In the event that the price of carbon credits is significantly lower than the opportunity and transaction costs, communities are likely to choose other land uses. Within the communities, individuals and households use the forest differently and to different extents, but all individuals within the same village receive a similar amount of PES. Thus, for some, the payment is likely to fall short of the opportunity cost.

In addition, weak governance at the village and district levels poses a threat to the success of the initiative. Although the proponents have had a strong focus on local capacity building, they may continue to face challenges related to rule enforcement and compliance that were present in the villages prior to implementation of the initiative (Dokken et al. 2014). Finally, there are questions about the treatment of households that moved from areas that were to become VFRs. Whether or not these households were in legal settlements before the implementation of REDD+, the proponents are likely to face friction related to what kind of rights the households have, and what kind of land they are allocated to compensate them for moving from areas that have been defined as forest reserves.

14.5 Lessons from the initiative

This initiative is demonstrating a unique approach to reducing emissions from community forests by working with communities while delivering incentive payments to individuals. It pays individuals for their contributions to the environment by reducing their rates of deforestation and forest degradation. One challenge for this model is that the amount paid to individuals does not differ depending on their opportunity cost. This reduces transaction costs related to PES but means that for some, the payment may exceed the opportunity cost, while for others the payment may be too small. In sum, this pilot of an individual REDD+ payment mechanism within the framework of CFM can potentially provide important insights, not only for REDD+ but also for how CFM schemes can be designed to provide individuals with incentives to reduce deforestation and forest degradation.

14.6 Acknowledgments

We would like to express our gratitude to a number of people. Charles Meshack, Nike Doggart and Theron Morgan Brown from TFCG have provided us with invaluable information. We would also like to acknowledge the contribution of the district and ward offices in Kilosa and Lindi Rural districts. Special thanks and appreciation also goes to village leaders, village executive officers, sub village leaders and all respondents of the various interviews conducted to make this work successful. Further, we would like to thank Susan Caplow for organizing the first round of data collection in Lindi Rural in 2010; thanks also to: George Phabian Kabado, Hawa Mushi, Thadeus Kisangi, Mohamed Mmaoga Omar, Georgina Misama, Yesaya Bendera, Nanjiva Mzunda, Aklei Albert, Julius Edward and Christina Justine for their assistance in the field as enumerators; and to Johannes Dill for data entry.