Policy Integration for REDD+: Insights from Mexico

: This paper investigates how the horizontal and vertical policy integration for REDD+ has been conceptualized and unfolded in Mexico during the REDD+ readiness and early implementation phase (2008–2019). We used the document analysis and interviews with key actors to identify changes that REDD+ induced in forestry and agriculture sectors’ policy making and programs at the national level and in two REDD+ states, Campeche and Jalisco. The policy integration for REDD+ in Mexico is conceptualized as compatibility-within-a-framework, i.e., promotes coordination and compromises among the objectives of land-use policies endorsing sustainable rural development. The state environmental agencies play an active role in involving the agricultural authorities in formal and informal interactions. This resulted in the design (Campeche) and even implementation (Jalisco) of REDD+-aligned programs and instruments. The progress at the subnational level is hampered by the national agriculture and forestry agencies’ policy inertia and lack of resources. To support the subnational REDD+ policy integration advancement, more resources and powers should be made available to the state environmental authorities which would help them directly reach more international funding and include other important REDD+ actors and establish mutually accountable relationships with them.


Introduction
Public policies are a series of decisions and activities designed by the government, alone or together with other actors in the society, i.e., governance, to respond to a problem of social interest, such as deforestation and forest degradation [1]. Deforestation and forest degradation are complex environmental problems, as it is difficult to disentangle their multiple-scale biophysical and societal causes, predict their consequences and reversibility, and know with certainty whether and when a forest will be clear-cut or disturbed. The drivers of deforestation and forest degradation are both internal and external to the forestry sector and operate at different scales, such as local migrations, national trends in urban development, and global markets for agricultural commodities [2]. Consequently, policy making to combat deforestation and forest degradation is a 'complex enterprise', and the governments were largely unable to individually provide effective responses to them [3]. The policies of different land-use sectors are poorly coordinated and enter antagonistic policy interplays [4]. Usually, instead of improving deforestation and forest degradation, policies shift them over time, space, and environment, or even generate new environmental and social problems [5].
Between 2009 and 2014, 800 million USD were channeled for REDD+ in Mexico, predominantly from international funds (59%) mainly in form of loans from WB, but also from national resources (41%) [23]. There has also been international funding to civil society organizations such as the Alliance MREDD+, a consortium of international and Mexican conservation organizations and research institutes led by The Nature Conservancy (TNC). The Mexico REDD+ project (MREDD+) (2010-2016) was financed with 30 million USD by the United States Agency for International Development (USAID) and had the objectives of institutional strengthening and capacity development through multi-stakeholder events and local pilot projects [24,25]. Besides the mentioned REDD+ funding sources, others include carbon markets, both voluntary and regulated [26]. It is still unclear whether REDD+ credits will be included in the final agreement on a global carbon market.
In order to explore different institutional arrangements, the REDD+ early actions (ATREDD+) were implemented in five Mexican states, including Campeche in the Yucatán Peninsula (PY) and Jalisco. These federal states are also implementing the World Bank's Forest Carbon Partnership Facility (WB FCPF) "Emission Reduction Initiative" (IRE), aimed at strengthening the alignment of land-use public policies at the subnational level and financing resulting emission reductions through the Carbon Fund which was put in place for this purpose [27].
The objective of this study is to investigate the policy integration for REDD+ in Mexico, both as a process and as a degree of vertical and horizontal integration of the forestry and agriculture sectors, the latter identified as causing most of the deforestation and forest degradation in Campeche and Jalisco, e.g., [28,29]. The following questions guided the investigation: How is the policy integration for REDD+ conceptualized in Mexico? Which changes have been induced by REDD+ in forestry and agriculture policy making at the federal and state level, in Campeche and Jalisco? To which degree have the federal and the states of Campeche and Jalisco agriculture programs been canceled or reformed to respond to the objectives of REDD+?
In the studied period of 2008-2019, which encompasses Mexico's REDD+ readiness and early implementation phase, the progress of policy integration at the subnational level was observed. However, this progress was hampered by insufficient integration between the agriculture and forestry sectors at the national level.

Materials and Methods
To understand how policy integration for REDD+ has been conceptualized in Mexico, we analyzed the five key national and state REDD+ policy documents published up to July 2019: the ENAREDD+; the state REDD+ strategies of Campeche (EEREDD+-Cam) and Jalisco (EEREDD+-Jal); and Campeche and Jalisco Investment Programs (PI-Cam and PI-Jal), a five-year REDD+ activity planning documents published under IRE.
The selected documents contain official decisions on REDD+ design and implementation at the national and state scale. In each document we looked for: 1. Explicit references to policy integration using the term 'policy' in combination with the term 'integration' and its synonyms, such as 'efficiency', 'co-ordination', 'compatibility', 'alignment', etc.; and 2. Implicit references to policy integration when the basic policy elements (object, goals, instruments, and target population) were described and commented in the text.
To comprehend how the quest for policy integration for REDD+ impacted agriculture and forestry policy making we performed 15 semi-structured interviews with state and municipal REDD+ actors from forestry and agriculture governmental sectors, civil society organizations (CSOs), and academia, and participated in five REDD+ bodies' sessions in both states. We analyzed the interviews and session transcripts containing information on REDD+ actors' knowledge and perceptions on inter-and intra-sectoral interactions, and about the ability of forestry and agriculture programs to complement each other. Although we primarily relied on using the data collected over the period 2017-2019, for the national level analysis, we also consulted the interviews and previous versions of REDD+ policy documents obtained from 2010 to 2014.
To learn how policy integration impacted the design of agriculture programs in the country, we examined if REDD+ objectives were included in their operating rules using the following criteria: general objective, specific objectives, instruments or technique, amount and periodicity of benefits, temporality, target population and typology of beneficiaries, target ecosystem, expected benefits and costs (economic, social, and environmental), and corresponding safeguards.

Conceptualization of Policy Integration for REDD+ in Mexico
From the beginning of Mexico's readiness processes, inter-and intra-sectoral coordination appeared as a normative prerequisite for REDD+ effectiveness, deriving from national environmental laws. Namely, a first intermediate product of this process, the Mexico's REDD+ Vision document (2010) calls for both: (a) promotion of alignment of forestry with agriculture, infrastructure, energy, and tourism sectors, and (b) coordination of activities at subnational and national levels [30]. In the ENAREDD+, public policy "alignment" and "coordination" are most commonly and interchangeably used terms and mean that: "The objectives, strategies, and lines of action of the ENAREDD+ should be incorporated in policies and programs of multiple sectors" [17] (p. 58). The specific objective is to "achieve the transversality, coordination, coherence, and integrated operation of public programs and policies that are favorable for REDD+" [17] (p. 60) and in particular of forestry and agriculture sectors, as the main deforestation driver in the country. This should be achieved through existing and new institutional mechanisms and spaces that include federal, state, municipal, and local actors [17].
The ENAREDD+ does not explore the possibility of switching to other types of policy instruments beyond economic subsidies. PI-Cam (2016) explains that, currently, it is easier to influence local producers' extrinsic (e.g., tangible reward, avoidance of sanction) than intrinsic motivations (e.g., satisfaction, responsibility) to adopt sustainable practices [31]. PI-Jal (2015) adds, that besides the lack of coordination among the public programs, their targeting is not optimal [32]. However, the ENAREDD+ does not state the necessity of calibration of policy programs and instruments, such as adjustments of their spatial and temporal scales to help the long-term planning. Instead, it explains that "the lack of coherence of sectoral policies at the level of instruments or actors is less important in causing land-use change than the antagonistic interplays among their objectives, which are defined in the laws" [17] (p. 58).
The state REDD+ strategies should be aligned with the ENAREDD+ (August 2017). The EEREDD+-Cam (2015) [33] is based on the regional strategy (2012) [34] and the previous versions of national strategy, and it will be continuously updated to incorporate new REDD+ agreements. The EEREDD+-Jal was published later in 2017, and contemplates the biannual evaluation of the REDD+ state policy [35].

REDD+ Policy Integration as a Process: Actors and Bodies
In 2009, the Inter-ministerial Commission on Climate Change (CICC), founded and led by the Ministry of Environment and Natural Resources (SEMARNAT), created the working group (GT-REDD+) to discuss national REDD+ development with the rest of governmental agencies. However, the National Forestry Commission (CONAFOR), a SEMARNAT's decentralized agency, is the main responsible for national forest governance and REDD+ development in Mexico. Namely, CONAFOR is the REDD+ focal point under the UNFCCC, and it coordinates the development of the WB's IRE, and signs the ENAREDD+.
In 2010, CONAFOR founded the Technical Advisory Committee (CTC-REDD+), a multi-stakeholder advisory forum of the GT-REDD+ that includes other sectors in the society. The working group on ENAREDD+ of the National Forest Council (CONAF) is another multi-stakeholder forum established in 2013 on demand of CSOs.
The Ministry of Agriculture and Rural Development (SADER, formerly SAGARPA; we use SAGARPA when we refer to the events and activities that happened until 2018, and SADER for those occurring after) is responsible for agricultural policies. SAGARPA is a secretary of CONAF and participates in GT-REDD+ and CTC-REDD+. In 2010 the Working Group for Territorial Projects to monitor REDD+ early actions was established under the SAGARPA-led Inter-ministerial Commission for Sustainable Rural Development to coordinate land-use sectors' policies [36].
The work of CICC's GT-REDD+ was interrupted by the federal election in 2018, and its reactivation-through meetings with other REDD+ 'co-responsible' environmental and agricultural agencies-was identified as a priority by new CONAFOR staff (CTC-Jal sessions, 20 November 2019 and 29 June 2020).

Forestry sector
The early REDD+ activities in Campeche and Jalisco were implemented through the Special Forestry Programs for the Yucatán Peninsula (PEPY) (2012-2014), and the Jalisco Coastal Watersheds (PECCJ) (2011-2014) financed from the WB's Forest Investment Program and Climate Investment Funds. The 'payment for ecosystem services' (PES) received the most resources in comparison to the totality of other PEPY's and PECCJ's components, including: 'forest restoration and reforestation', 'forest studies, assessments, and plans', 'timber and non-timber forest products management', 'fire prevention', 'agricultural/livestock/forestry productive activities' (agroforestry, silvopastoral systems, and management of secondary forests), and 'community forestry governance'. The last component included support for the elaboration of the P-Predial, a local five year planning document meant to align local implementation of forestry programs [37,38].
The Special Programs' menu was adopted from ProÁrbol (2008-2012), an overarching CONAFOR's program, except for the last three components. The PECCJ was implemented for five years, as planned, while PEPY finalized after four years due to lack of budget. Both Special Programs experienced yearly changes in their operating rules. In parallel to PEPY and PECCJ, CONAFOR implemented two other programs targeting the same population and promoting same activities: The Community-based Forestry Development Project in the Southern States (2012-2015, only Campeche) and the National Forest Program (PRONAFOR) (2014-2018), which provoked certain confusions at the local level (Initiative DICOS, 9 July 2018). PEPY's budget was minimal compared to these other two programs, while PECCJ received the same amount as PRONAFOR.

Agriculture sector
The most important federal agricultural programs in Campeche and Jalisco, based on the number of beneficiaries, land area, amount invested, and impact on forests, e.g., [39][40][41][42] are: Direct Field Support Program (PROAGRO, formerly PROCAMPO), Program for Sustainable Livestock and Beekeeping Production (PROGAN), Program for Corn and Bean Producers Support (PIMAF), and Program for Food Security (PESA). The operating rules (2008-2019) ( Table 1) of all programs focus on productivity and do not contemplate negative social or environmental impacts. The rationale behind this is the belief that, by achieving the programs' objectives, the pressure on the forests will decrease, and that it is the local producers' responsibility to use the subsidies in a sustainable manner (SAGARPA, 27 July 2011). In 2018, SAGARPA, for the first time, explicitly included environmental restrictions to the forest land-use changes in the Agricultural Promotion Program's operating rules. This was a result of an agreement signed between SEMARNAT and SAGARPA in 2016 in the framework of the 13th Conference of the Parties (COP13) of the Convention on Biological Diversity (CBD) [43].
In 2019, the new federal government implemented two rural development programs: Trust-Based Livestock Credit (Crédito Ganadero a la Palabra) of SADER, and Sowing Life (Sembrando Vida) of the Ministry of Welfare (only Campeche). The Livestock Credit program has the objective of increasing the productivity of small farmers by providing them with technical assistance and up to 35 cows depending on the size of their pasture area [44].
The Sowing Life program has the objective to contribute to the social well-being of small producers who should established agroforestry systems on 2.5 hectares. The support is monetary, 5000 MXN pesos (approximately 220 USD) per month, and in-kind (plants, supplies, tools, and technical support) [45]. The program is criticized for its negative impact on both secondary and primary forests, due to flaws in its design and lack of implementation surveillance, respectively, e.g., [46,47]. Additionally, with ten times more funds than the entire CONAFOR, the Sowing Life dominates over forestry programs that should be focused on stopping forest loss and promoting community forest management [48]. As such, it is perceived as unlinked from REDD+. However, according to the CONAFOR's managing director: "At the center of the program are the people and their wellbeing, and that is why its position under the Ministry of Welfare is adequate. Other institutions should play supportive role" [49]. Additionally, as a participant in the Sowing Life technical advisory committee, he tries to integrate it with the national REDD+ framework (CTC-Jal sessions, 29 June 2020).

REDD+ Actors and Bodies, Campeche
The Ministry of Environment and Natural Resources, Campeche (SEMARNATCAM; formerly SMAAS, now SEMABICC) is in charge of the state forest fire and reforestation programs, oversees the implementation of the federal environmental programs, and leads the REDD+ development in the state. The Ministry of Rural Development (SDR) is responsible for the design and implementation of Campeche agricultural policy. The central SDR program implemented in coordination with municipal authorities (2010-2018) was agricultural mechanization to increase production, reduce the use of agrochemicals and fertilizers, and decrease the deforestation and soil erosion (municipal agriculture authority, 9 July 2018).
Since 2011, the Campeche CTC-REDD+ (CTC-Cam) is a main REDD+ advisory body in the state. Representatives of the government, CSOs, academia, and local communities participate in the CTC-Cam, but the faculty to make decisions is reserved to the representatives of five main productive sectors (agriculture, livestock, forestry, beekeeping, and hunting and fishing) from eight municipalities hosting ATREDD+. The CTC-Cam is presided by a CSO (Initiative DICOS), but in practice, the meetings are convened by its technical secretary SEMARNATCAM. The CTC-Cam had an important role in elaborating the EEREDD+-Cam, but it was only involved in the validation of the PI-Cam (2016) (Initiative DICOS, 9 July 2018).
PI-Cam was elaborated by federal and state environmental authorities in workshops with small local producers in three IRE municipalities. Although this process was financed and facilitated through the Alliance MREDD+, receiving a status of a REDD+ development and implementation partner, the government had the final decision on the REDD+ design under the national approach investigated here. Likewise, the final version of PI-Cam was authorized by the REDD+ working group (GTREDD+) under the Inter-ministerial Commission on Climate Change of Campeche (CICC-Cam), which was reinstated for this purpose in 2017 (SEMARNATCAM, 5 July 2018).
However, REDD+ is principally discussed through the REDD+ working group under the State of Campeche Forestry Councils (GTREDD+COEF-Cam). It is a multi-stakeholder forum chaired by SEMARNATCAM, in which the right to vote is reserved to the accredited CSOs representatives and SDR participates as a guest. The GTREDD+COEF-Cam was suggested for the role of Public Agent of Territorial Development (APDT) for Campeche. However, given that APDT should be a decentralized public entity in charge of assisting integrated rural development in IRE regions [50], CONAFOR rejected this proposal.
The municipal environmental authorities participate in both the CTC-Cam and COEF-Cam. However, their responsibilities within REDD+ are limited to supporting implementation of the forestry programs, and elaboration of the Ecological Territorial Planning Programs. The municipal agricultural authorities were largely absent from the REDD+ process. Nevertheless, through productive sectors' representatives in CTC-Cam, REDD+ was brought to discussion in the Municipal Councils for Sustainable Rural Development (Initiative DICOS, 9 July 2018).
The state branch of national forestry authorities finds the fact that the SEMARNAT state branch is absent from the REDD+ process in Campeche particularly worrying, given that this agency is in charge of issuing the permits for large industrial agriculture development in this state and as: "[SEMARNAT] is the mother of the environmental sector, it should have coordinated us [ . . . ] but instead, we had to involve them. We told them: 'REDD+ is your responsibility too'" (CONAFOR Campeche, 10 July 2018).
The EEREDD+-Cam (2015) is still not fully implemented as a result of the slowdown in the national REDD+ process and the fact that: "We are a little disconnected from those negotiations with WB, which is seen directly by CONAFOR; actually, we requested through national CTC-REDD+ to be informed about IRE" (SEMARNATCAM, 5 July 2018). In addition, only three municipalities with the highest deforestation rates were elected as IRE polygons, while only one was part of the ATREDD+ areas (CONAFOR Campeche, 10 July 2018).
Although SDR did not participate in the PI-Cam elaboration, its role in policy integration for REDD+ has been more pronounced since 2017, when a new 'sustainable development area' within this agency was established at the initiative of an ex-SEMARNATCAM employee: "We have never managed to click with SEMARNATCAM, it had always been opposition. Before you talked to people [in SDR] and they did not understand what REDD+ was, they would say: 'Hey, you're blaming me'. Now with an environmental area, it is a bit easier" (SDR, 5 July 2018).
Another strategy to bring closer REDD+ to SDR's productive areas was: "I know some of the REDD+ strategic lines, but I am not relying on them. I could not come with REDD+ strategy and say: 'OK, this is what we are going to do!' It was easier to talk individually with the area directors, for example, when I mentioned the silvopastoral system, the livestock director said: 'Hey, I am interested, let's do it'" (SDR, 5 July 2018).
In 2018 SDR started convening both state and federal environmental agencies to "Round table for the construction of a shared territorial vision" to work on the implementation of the PI-Cam. The decisions made in the "Round table" are communicated with COEF-Cam. To keep the discussion manageable, other governmental sectors were deliberatively excluded from these meetings (SDR, 5 July 2018). However, SDR invited CSO PRONATURA PY which, since 2014, implements local sustainable productive projects financed by the MREDD+ (PRONATURA PY, 10 July 2018).
Since 2015, Campeche is a member of Governors' Climate and Forests Task Force (GCF). This multi-jurisdictional collaborative effort financially supports improvements in the subnational governments' capacities for REDD+ policy integration and commercialization of carbon credits through voluntary markets [51]. In 2016, as part of the IRE process, the Yucatán Peninsula Climate Fund (FCPY) was created as a partnership among TNC and the three biggest universities in each state in the PY (Campeche, Quintana Roo and Yucatán). The FCPY has an aim to receive and distribute the REDD+ benefits from the national to state level [25].

Policy Design Integration, Campeche
PI-Cam lists a broader array of REDD+ activities than PEPY, and includes both incentives and enabling measures for conservation and productive activities in forestry and agriculture sectors. In accordance with the PI-Cam, the SDR proposed how to adapt practices in its agriculture, livestock, and beekeeping production activities in order to contribute to REDD+ objectives (Table 2).   These activities will be financed from national and state concurrent funds to offset the SDR budget cuts by 25% in 2018. It could, however, be difficult to change the local producers' mentality to switch to new sustainable practices (SDR, 5 July 2018). In addition, unsustainable agricultural activities will be still supported by SAGARPA: "[At COP13 of CBD] the agreement between SEMARNAT and SAGARPA was signed, but it never trickled down. Agricultural policy is designed at the federal level, practically it gets down, and what we are doing is helping its implementation. The federal level continues with the same operating rules, which will not change" (SDR, 5 July 2018).
Harmonization of the agencies' calendars, and making their program rules and budget flexible, would allow complementation of their support for REDD+ activities (Initiative DICOS, 9 July 2018). To improve the public budget flexibility, REDD+ activities should be included in the State Development Plan (CONAFOR Campeche, 10 July 2018).
The elaboration of P-Predial under PEPY should have helped align several forestry programs implemented at the local level. However, the majority of these documents were not authorized by CONAFOR to be copy-pasted by the forest technicians (ex-SEMARNATCAM, 20 September 2019). However, the forest technician (independent, 9 July 2018) is of the opinion that P-Predial could not be implemented as components and eligibility areas of CONAFOR programs were changed or canceled. CONAFOR official considers some productive activities in P-Predial could still be implemented with the agriculture sector's resources: "If there was some activity in P-Predial that corresponded to SDR, we then informed them or SAGARPA. We did not follow up if they actually did something in respect, if they gave the money to the producers or not, we only inform them" (CONAFOR Campeche, 10 July 2018). Two additional initiatives aligned with REDD+ objectives: the Biodiversity and Sustainable Agrosilvopastoralist Livestock Landscapes (BioPaSOS) (2017-2020) and the Sustainable Productive Territories (TPS) (2018-2022) are implemented in Campeche by the National Commission for the Knowledge and Use of Biodiversity in the non-IRE municipalities. BioPaSOS is led by Tropical Agricultural Research and Higher Education Centre (CATIE) and supports the agrosilvopastoral practices [52], while TPS promotes sustainable forest management and is financed from the WB's Global Environment Facility fund [53].

REDD+ Actors and Bodies, Jalisco
The Ministry of Environment and Territorial Development (SEMADET) is responsible for the state forest program for fire prevention and management and supports the implementation of CONAFOR programs. In turn, the Ministry of Agriculture and Rural Development (SEDER) designs Jalisco agriculture programs and coordinates the implementation of SADER programs. The technical and financial assistance for rural development is also provided by the Trust Fund for the Administration of the State Forest Development Program (FIPRODEFO). Since 2016, SEMADET led the EEREDD+-Jal elaboration workshops with forestry and agricultural agencies, academia, CSOs, and the WB. The EEREDD+-Jal drafts were approved by the GT-REDD+-Jal, the main REDD+ inter-ministerial body established under the Inter-ministerial Commission on Climate Change of Jalisco (CICC-Jal) in 2016 [32]. Since 2018, the GT-REDD+-Jal's operational structure includes a "Taskforce on silvopastoral and agroforestry systems" led by state agricultural authority.
Besides forestry and agriculture authorities, the CICC-Jal include the Ministry of . In practice, the municipalities' ecology directors participate in the councils, but hold no decision-making power (JIRCO, 18 July 2019). In 2015, each junta elaborated its PI in workshops with local actors. Therefore, the EEREDD+-Jal (2017) included the thus-far experience of IRE implementation, and it was publicly consulted in 2018.
The CTC-REDD+ in Jalisco (CTC-Jal) was established only in 2017 and includes as members CSOs, the private sector, academia, forest technicians, and local communities, while forestry and agriculture agencies are permanent guests. The CTC-Jal is presided by the civil sector (currently an academia representative), while SEMADET is its technical secretary. This multisectoral body operates though the subcommittees that emit recommendations to the GT-REDD+-Jal. Many CTC-Jal participants have been actively involved in the readiness process from its beginning as members of several different government sectors' agencies and CSOs.
In 2018, SEMADET, together with state and federal agriculture, forestry, and social development authorities, and with guidance by the Subministry of Planning and Evaluation, Jalisco (SUBSEPLAN), designed the Predial Program of Integral Development of Medium Term (P-Predial-Jal) of both forest and agriculture activities, that was pilot tested in 2018 by juntas (JIRCO, 18 July 2019). Jalisco is also a member of and receives financial support from GCF initiative. In addition to the Fund for the Administration of the Forest Development Program (1998), since 2016, Jalisco counts with the State Environmental Protection Fund created by the State Government to receive and manage public and private resources for environmental projects and actions at the state level, including REDD+ [25].

Policy Design Integration, Jalisco
The REDD+ activities listed in PI-Jal of the JIRCO (2016) include a wider array of incentives and enabling measures than PECCJ, such as conservation and productive activities in both forestry and agriculture sectors ( Table 3).   Since 2014, the activities of traditional agriculture improvements were supported with the Sustainable Production Alternatives Initiatives (IAPS), including zero logging, fire and tillage use, pest control, and live fences and agroforestry. The IAPS were implemented by JIMAs with financial support from Institutional Trusts in Relation to Agriculture (FIRA) and National Development Finance (FND). In JIRCO, this initiative also included forest recovery of areas cleared for avocado plantations. However, this reforestation was not successful, given that, instead of native timber species, fruit trees were provided by the municipality authorities. The number of participants in IAPS decreased in the subsequent years due to a lack of resources (JIRCO, 18 July 2019).
Following the suggestions in the Jalisco PIs, in 2016 SEDER included establishment of intensive silvopastoral and agroforestry systems (SSPI) in its program's operating rules. SSPI was financed from governmental concurrent funds. Despite the cuts in the SEDER budget, the SSPI was promoted in a separate program in 2020. The targeted population are experienced livestock producers with extensive farming, who have the capacity to finance half of the project's costs, and have the part of their property "sustainably managed" with some of the forest conservation or management programs. This should guarantee that once silvopastoral activity becomes profitable, it will not be expanded to the adjacent forests [32]. The SSPI initiative is described by SEDER as an example of alignment of agriculture and forestry instruments: " . . . [this is] the first time that two agencies with different approaches harmonize their operating rules to generate tools for territorial development through an approach that is both productive and environmentally friendly [ . . . ] a program that considers the environment and does not affect agricultural production" [54].
JIMAs were in charge of distribution the call, receiving the applications and providing technical assistance to the SSPI participants. Additionally, JIMAs were assigned with the authority to endorse the property as "sustainably managed" (JIRCO, 18 July 2019). The Jalisco state agriculture and environmental authority think that it will be hard to change the local livestock practices, in particular as they are supported through new federal program. Since 2018, the silvopastoral systems in Jalisco are also supported by BioPaSOS initiative.
Since 2016, the state Ministry of Development and Social Integration (SEDIS) implements productive projects for women and young people, such as tree nurseries for foraging and timber trees for SSPI. Similar projects are supported by JIMAs in partnerships with private sector (JIRCO, 18 July 2019). Forestry activities were supported with local model of PES (MLPSA) (2017) financed from governmental concurrent funds. In 2019, FIPRODEFO implemented reforestation and forest pest control initiatives in line with the federal rules, in addition to the elaboration of P-Predial-Jal and strengthening of local governance. Forest management activities in Jalisco are also supported by the TPS initiative [53].

Discussion
The conceptualization of policy integration in REDD+ documents in Mexico approximates to compatibility-within-a-framework, i.e., promotes accordant relationships among land-use policies and allows for compromises among their goals and objectives while endorsing sustainable rural development as an overarching principle [10]. Although it is clear that REDD+ should not only concern forest-based activities, this process in Mexico is led by the forestry authority (see [55,56]). The institutional inertia, i.e., the choice of the lead institution being influenced by past circumstances that are no longer relevant [57], combined with the historically antagonistic interplays between the objects and goals of forestry and agriculture policies, triggered certain tensions (see [58]). In addition, National Forestry Commission's capacity to coordinate policy reform for REDD+ is challenged by the persistent decrease in its financial and human resources [48]. This is common to forestry agencies developing REDD+ in various countries [59][60][61][62]. The ministry of environment might have more authority than forestry commission to pursue the policy integration for REDD+, such as to sign the Emission Reductions Payment Agreement (ERPA) with FCPF. However, SEMARNAT did not develop strong ownership and interest in REDD+, while it suffers from low efficiency and underfunding [63].
The REDD+ working groups established under several inter-ministerial formal bodies with decision-making power, as well as informal spaces of discussion, tried to overcome these institutional shortcomings and motivational difficulties. The establishment of the REDD+ working group under the Commission for Sustainable Rural Development was an intention to develop a more sense of ownership and interest in REDD+ within the agriculture sector. However, this group has never reached the importance within the national REDD+ discussion, and the Ministry of Agriculture continues to deliberately play only a secondary role in the REDD+. This could be explained by the fact that the opinion of agricultural representative is that the Borlaug hypothesis will be proved in Mexico; that is, the agricultural intensification achieved with subsidies will result in saving more forests [64]. However, the opposite situation, known as the Jevon's paradox [65], can occur, and more profit from the agricultural intensification might prompt farmers to expand agriculture to forest land (see [42,66]). In addition, and with the aim of achieving shared responsibilities, the ENAREDD+ calls for the coordination of the public policies favorable for REDD+, i.e., those in synergistic interplays with forestry policies, while for REDD+ to be effective those in antagonistic interactions should be reformed or even canceled.
Differently, in Peru and Vietnam, the agricultural federal agencies share authority over national REDD+ with the environmental and forestry agencies, respectively [67,68]. However, this did not automatically result in more policy integration; in fact, lack of it is common to many REDD+ countries [62,69]. Therefore, not surprisingly, the FCPF's Carbon Fund lifetime was extended by an additional five years (2020-2025) due to globally low achievements in the alignment of the land-use national policies [14], which was also noticeable for Mexico.
The adoption of a jurisdictional approach to REDD+ under the IRE process led to more REDD+ responsibilities delegated to the state environmental authorities. This process can be characterized as a 'deconcentration' of administrative and/or technical functions [70], given that the resources and power stayed within the national forestry authorities, to whom the state actors should be accountable in a procedure that is largely unclear to them. The federal governments in Mexico and other REDD+ countries still portray themselves as more capable and reliable than state authorities, particularly in managing the REDD+ financial resources [71,72]. According to the UNFCCC, REDD+ countries must follow the national approach and centralized accounting in order to avoid double counting of emission reductions. However, the sub-national approach is recognized as an interim measure towards national implementation [15]. The sub-national approach makes visible conflicts between policies at the local implementation stage that might not be foreseen by national governments who formally design them with the knowledge obtained in international negotiations [4,13].
The role and capacities of municipal environmental authorities for REDD+ in Campeche and Jalisco did not substantially change with the jurisdictional approach, nor did it manage to involve municipal agricultural authorities. In addition, in at least one municipality in Campeche, which was not selected for IRE implementation, REDD+ disappeared from the agenda of the Municipal Councils for Sustainable Rural Development led by the agricultural authorities (Initiative DICOS, 9 July 2018). Although in Jalisco, the juntas play an important role in reaching the target population, they still experience difficulties in work with municipal authorities (see also [73]).
In both states, REDD+ is discussed in the multi-stakeholder working groups that should assure that the new land-use programs are REDD+-aligned, i.e., they integrate REDD+ objectives. In Campeche, this group is part of the State Forestry Councils, in which the agriculture sector holds less power than the environmental, while in Jalisco it is hosted by the Inter-ministerial Commission on Climate Change, in which a larger number of governmental sectors participate on an equal basis. In addition, a circulation of staff between environmental and other agencies, although unplanned [3], contributed to better understanding among REDD+ sectors in both states. The positive effects of permanence of human capital in REDD+ discussion was particularly observable in Jalisco, which resulted not only in design, but also in implementation of a variety of programs proved to contribute to REDD+ objectives. However, subnational programs are executed in parallel with federal agricultural programs, whose rules were not modified. Although altering the policy written content is not a panacea, it could provide a stronger political signal that REDD+ is being taken seriously by all land-use sectors at all governance levels, and at all stages of the policy process [4]. In the meantime, a local planning instrument designed in Jalisco should help implementation of agricultural and forestry incentives from all levels in a complementary manner.
Policy integration is needed to "manage the numerous policy interconnections so that policy supply meets demand" [3] (p. 4). REDD+ was criticized for principally targeting small agriculture producers, instead of attacking the main forest loss drivers, such as industrial agriculture [74]. For example, in Campeche, the REDD+ target population does not include large industrial farmers, despite the fact that their agricultural activity is recognized as one of the most important deforestation drivers in the state [28,31]. Although the criticized REDD+ targeting trend largely repeats in our results, proposed REDD+ policy includes sustainable certification as an instrument to tackle the production of main commercial crops, avocado and agave in Jalisco, and oil palm in Campeche.

Conclusions
The jurisdictional approach to REDD+ under the IRE process in Mexico positively influenced the alignment of public policies for REDD+ at the subnational level. Namely, this process promoted the active role of environmental agencies, which in turn encouraged the participation of agricultural authorities through the reinstallation and/or establishment of formal and informal REDD+ institutional structures. This brought some concrete programs and planning instruments that were designed in Campeche and even implemented in Jalisco, in parallel to several other initiatives aligned with IRE objectives and led by international actors.
However, the progress at the subnational level is hampered by insufficient and slow vertical and horizontal integration at the national level, caused by the inertia and lack of resources in the national agriculture and forestry agencies, respectively. Moreover, with the change of federal government in 2018, and despite the allegations in the official documents, the forestry sector and REDD+ has moved down the agenda. This change can be evidenced by the design and implementation of the two main agriculture and rural development programs. At the same time, the addressing of the vertical and horizontal fragmentation of national REDD+ policy in Mexico is just beginning with the resumption of inter-sectoral interactions. Additionally, the further progress of the integration process at the national level might be challenged by an uncertain future of international REDD+ funding.
Although ambitiously conceptualized in documents as a compatibility-within-aframework, our results suggest that the policy integration for REDD+ in Mexico materialized as efficiency, i.e., coordinated actions of state agriculture and environmental sectors to maximize the use of available subnational resources and achieve objectives of both policies, rather than REDD+ objectives.
The solution for the lack of policy integration at the national level could be to organize REDD+ discussions on a neutral ground, such as through an overarching institution for enabling interaction management. We believe this would push for more symmetric integration, i.e., forestry policy incorporates concerns of agricultural sector, as well as generate more ownership and interest in REDD+ by land-use sectors. To support current progress of the policy integration for REDD+ at the state level, we suggest 'devolution' of resources and powers to the subnational environment authorities [70]. This would help them directly obtain more international financial resources to include-through the most appropriate multi-stakeholder structures-all important REDD+ actors from both governments, such as municipal agricultural authorities, and non-governmental sectors, and establish mutually accountable relationships with them.

Informed Consent Statement:
The verbal consent was sought and gained from all participants when conducting interviews and when participating in events, as well as for note-taking and voice recording. The interviews are anonymized in the text as per ethical guarantees.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to ethical guarantees of anonymity and respect of consent of research participants. All analyzed documents are duly cited in the text and can be found in the reference list.