OECD Workshop on the

Evaluation of Agri-environmental Policies 20-22 June 2011 The Johann Heinrich von Thünen Institute, Bundesallee 50, 38116 Braunschweig, Germany

Tuesday, 21 June Session 5 : 11h00-13h00

Agri-environmental Policies in Brazil and Perspectives for Evaluation Matheus A. Zanella1 and Lea V. Cardoso2 1. Humboldt-Universität, Germany 2. Socio-environmental Institute (ISA), Brasilia, Brazil

1

Abstract Brazilian policy-makers are not used to refer as agri-environmental policies to a number of different policies marked by the interaction of agriculture and environment. We describe and classify eight Brazilian policies that suit a practical definition of agri-environmental policies commonly used by OECD country members. Four of these policies, selected due to their importance to agricultural and environmental agendas, are described in details, emphasizing policy monitoring and evaluation mechanisms and methodologies: i) the Low Carbon Agriculture Plan, ii) the Amazon Fund, iii) Payment for Environmental Services projects, and iv) the Forestry Code. Given the relatively freshness of these policies, full and comprehensive ex-post evaluations are exceptions, even though most of these consider monitoring mechanism and ex-ante evaluations in their intervention plans. Besides, relatively low concern is given to apply methodologies that could disentangle the impact of policy measures from external effects. We conclude that these policy innovations represent a turning point into more use of economic instruments in Brazilian agri-environmental policy-making and that, albeit with important exceptions, better interagency coordination is required to design more efficient policies.

Introduction The purpose of this paper is to briefly describe some important agricultural and environmental policies in Brazil using current OECD definition and classification of agri-environmental policies. Focus is given to the newest policy evolvements, which include recent Federal legislation discussions and implementation of national programmes and initiatives aimed at reducing environmental impacts of agriculture. Given the recentness of these policies – the majority is less than 5 years under force and some are still at design stage – complete evaluations are still rare. Thus, the paper presents what the perspectives are and which methodologies are intended to conduct future policy evaluations. Section one introduces the term agri-environmental policies (AEPs) in the Brazilian context. It is explained that this term does not find explicit correspondence in agricultural or environmental policymaking debate, neither in Brazilian academic literature. Other terms or conceptualizations are usually found to describe policies that interact both with agriculture and environment. Nevertheless, if we define agri-environmental policies according to its current use by OECD country members, one could classify a number of important agricultural and environmental policies as AEPs, both within national and subnational administrations; and this is one of the intentions of this article. The most important Brazilian policies that would suit the definition of agri-environmental policies are presented in section two. A preliminary and non-exhaustive list of these policies is provided and these are classified using the categories of information instruments, economic instruments and regulations. These policies are briefly described following similar categories existing in the matrix classification used by OECD country members to construct their inventories of policy measures addressing environmental issues in agriculture. Four of these agri-environmental policies, selected due to their importance to agricultural and environmental agendas, are explored in more details: a) the Low Carbon Agriculture Plan (Plano ABC), an inter-agency plan constituted of climate change mitigation and adaptation policies, b) the Amazon Fund (Fundo Amazônia), an initiative coordinated by the Brazilian Development Bank to support conservation projects in the Amazon forest, c) Payment for Environmental Services projects, a series of decentralized local and regional interventions which provide payments for farmers that support the provision of environmental services, d) the Forestry Code (Código Florestal), the most important private land-use

2

regulation and currently under intense discussions at the National Congress in order to reform current legislation. Section three continues to explore these four policies into details emphasising policy monitoring and evaluation mechanisms and methodologies. Again, given the relatively freshness of these policies, full and comprehensive evaluations are exception. Nevertheless, most of these consider monitoring mechanism and ex-ante evaluations in their intervention plans. The situation with ex-post evaluations or impact evaluation studies is still unclear for the majority of these policies. Further, relatively low concern is given to apply methodologies that could disentangle the impact of policy measures from external effects. Finally, the paper concludes re-affirming the challenges that arise during the design and implementation of relatively new agri-environmental policies. With some exceptions, the inexistence of coordination between environmental objectives and different government agencies poses an additional problem of policy coherence that should be cautiously considered. The variety of plans, programmes, actors and instruments are common characteristics of agri-environmental policies in many countries, and Brazil is not an exception at this. We argue that some of these policies do not follow a strategic framework of intervention or a national strategy that coordinates different actions into unified objectives. There is a mix of policy instruments and environmental targets, widespread throughout the Federal, State and local administrations, but without significant linkages or coordinated plans that usually categorize national strategies of government intervention. Some important exceptions must be stressed, for instance, a Federal programme to reduce carbon emissions from the agricultural sector (Plano ABC), encompassed in the Climate Change National Policy, which presents significant collaboration of different government agencies and civil society representations. 1. Agri-environmental policies in the Brazilian context At first sight, one might be surprised when conducting a bibliography review about agrienvironmental policies in Brazil. By looking at academic journals, policy reviews and even non formal publications, such as policy-making presentations and working papers, hardly any reference can be found, neither in Portuguese nor English. The few papers published use the concept agri-environmental policies in different meanings than usually understood in current OECD literature (Vojtech, 2010). The term rather proximate to agro-ecology and other alternative environmentally-friendly farming systems. However, to simply conclude agri-environmental policies in Brazil being inexistent or unimportant to agricultural and environmental agenda is misleading. Even though this concept does not find correspondence in Brazilian context, other terms and conceptualizations are generally used to describe policies that in other context would fall into agri-environmental policies categorization. Those can be identified in several academic domains, from general agricultural economics, to environmental and ecological economics, and literature is vast in policy analysis of these agriculture and environment interactions. This article does not intend to provide an explanation why we observe this difference in terminology. Our intention is to apply a current practical definition of agri-environmental policies to a number of policies that are usually categorized as of agricultural policy or environmental policy domains, allowing us to explore some elements of these into details. We believe that this exercise would allow fruitful comparisons with policies in OECD country members, especially in the evaluation domain. Therefore, while referring to agri-environmental policies in the Brazilian context, similar to what was proposed by Jones (2005), we indicate policies that: i) select farmers as important agents and targets of intervention; and ii) address environmental objectives as primary objectives.

3

From this practical definition, we eliminate some policies that could be considered agrienvironmental policies, but fail to comply with both elements of our definition. One examples of these would be Ecological-Economic Zoning policies (Sombroek et al., 2000; Nogueira et al., 2008), since those are land use policies aimed at ordering economic behaviour of a number of agents but not only and mostly farmers. And another illustration would be biofuels policies, these latter eliminated due to the fact that we consider that in these policies environmental objectives are most of times complementary to other rural and economic development objectives. Other examples that could not comply with the two main characteristics of the definition here proposed are the Action Plan to Prevent and Control Deforestation in the Amazon (PPCDAM) and the Action Plan to Prevent and Control Deforestation and Fires in Savannahs (PPCerrado) (Brasil, 2003, 2009, 2010a). Respectively launched in 2004 and 2010, these two policies have in their programme of actions considerable interventions that are targeted to impact farmers towards environmental objectives, mostly by using regulations, monitoring, and other command and control policy instruments to curb illegal deforestation in these two regions. As with ecological-economic zoning policies, these two wide action plants involve a number of other local agents besides farmers, resulting difficult to affirm that agriculture concentrates mostly of the intended intervention. The main axes of these plans consist in the creation of natural protected areas in endangered zones and more rigourous and control of land use regulations. To achieve these objectives, considerable effort has been given to develop effective governance structures and institutions, which are usually not the main objective of agri-environmental policies. Some actions could indeed suit perfectly our definition – mainly those related to the “support of sustainable production activities” – however, we consider that these specific actions are not at the core of PPCDAM and PPCerrado, since they consist of implementing other already existing agri-environmental policies which are described in this paper, such as the Forestry Code. The last clarification of our agri-environmental policy definition is its indifference to the execution or implementation agency, coordination ministry or political origin of the policy. As indicated in the following section, agri-environmental policies in Brazil have heterogeneous governing bodies. At the Federal level, some are more proximate to Ministry of Agriculture, some to the Ministry of Environment, while others are completely decentralized or governed by financing agencies. This is similar to other approaches to the categorization of agri-environmental policies found in international literature.

2. Main Agri-Environmental Policies Although some policies were not selected in the previous section, some other important Brazilian policies suit the definition of agri-environmental policies. These have various origins and are coordinated by different agencies, but they rely mostly in one or two bodies of the Federal government level. The majority of these have less than 5 years of implementation, some are still being developed and/or in constant reform, even though other policies can have their origins tracked back decades ago. One important feature of most of these policies is the usage of more than one policy instrument, sometime combining both information instruments with regulations, or economic instruments with information instruments. Another characteristic is either the identification of more than one environmental target or even the non-identification of a singular explicit environmental objective, the later being categorized as agri-environmental policies with general or broad spectrum. Table 1 presents a preliminary and non-exhaustive list of these policies, classifying the policy instruments and environmental objectives according to similar categories found in the matrix classification used by OECD country member in their agri-environmental policy inventories. These policies are here briefly described and special attention is given to the Low Carbon Agriculture Programme, the Amazon Fund, Payments for Environmental Services projects and the Forestry Code.

4

Micro watersheds programmes Since 1980´s, Micro Watersheds (MW) Programmes have been carried on by some sub-national governments from States from Southern and South-Eastern regions of Brazil, in many cases supported by external financing from the World Bank and Inter-American Development Bank. The main goals of these programmes consisted of broad objectives of agricultural productivity and farm income increase, and improving rural population living conditions (Zafaroni, 1998). These programmes can be categorized as agri-environmental policies because of their early innovative approaches of focusing in micro watersheds as target areas and explicit concerns with sustainable natural resources use, water erosion control, land use planning and general environmental conservation. The MW-programmes have become largely dependent on participatory planning methodologies, involving rural communities into policy design and execution. Thus, besides community-based measures, other implemented policy instruments consists mostly of information instruments, such as technical assistance, information dissemination and rural extension activities. In the following decade, some contextual factors contributed to reinforce some objectives of MWprojects. According to Navarro (2009) there are many emerging aspects within Brazilian rural development context which contributed to the relevance of MW-projects. Falling commodities prices, high levels of inflation, a context of increased indebtedness and other economic disincentives in the early 90s forced many farmers into rural migration, and demonstrated the need for new broad and integrated proposals of public intervention. Later, the understanding of “poverty” as a public issue in mainstream debates, after the launching of UN-inspired indexes of social development, influenced the social shape of MC-projects. Another fact was the institutionalization of the concept “family farm”, not only as a new category to rural social group classification, but specially to implement targeted policies for this type of farmers. Another relevant leverage for modelling new MW-projects resulted from the consolidation of environmental legal concepts into legislation and a growing capacity of States to enforce environmental regulations. For the author, the convergence of all these factors forced MW-projects to change the initial rationale from being merely a natural resources management perspective to incorporate a broader focus capable of addressing these processes of social and economical change in course at that time, reinforcing their participatory nature in design and implementation. In some States, these programmes were successfully implemented and consecutively renovated for another intervention period, while others were already on its third project cycle. Nevertheless, these programmes remain concentrated in Southern and South-Easter States. The Federal government do have a national Micro Watershed Programme, however these consists of local projects with low impact on communities and few integration among the different projects. Main activities under national MW programme consist of providing information about good agricultural practices for water and soil conservation (Brasil, 2011a).

5

Table 1. Main Agri-Environmental Policies in Brazil Name

Execution / Implementation Agency

Main Supportive Bodies

Policy Instruments1

SEAs3

Extension Agen.

11, 12, 13

1, 2, 3

1980/90’s

Decentralized

ANA

2, 3, 4, 12, 13

2, 5, 9

2000’s

Organics

MAPA

MDA

7, 11, 12, 13

2, 7, 11

2003

Environmental cross compliance Amazon Fund

BC BNDES

MF, MAPA COFA4

8, 9, 10 11, 12, 13

1 1, 5, 9

2008 2009

Minimum prices for socio-biodiversity products

CONAB

MDA, MMA

3, 13

1, 12

2010

1, 4, 11, 12

3, 6, 9, 12

2010

7, 11, 12, 13

2, 7, 11

1965/2011

Micro Watersheds programmes Payments for Environmental Services projects

Low Carbon Agriculture Plan Forestry Code Notes: 1 Policy Instruments: Economic Instruments (1) Payments based on fixed farm assets (2) Payments based on land retirement (3) Payments based on farming practices (4) Payments based on investments (5) Environmental taxes /charges (6) Tradable rights / quotas (7) Labelling standards / certification

MAPA, MDA

Several

IBAMA

MMA

5

Main First Year of Environmental Implementation Objectives2

2

Environmental Objectives (1) Generic / broad spectrum (2) Water resources / pollution (3) Soil quality / protection (4) Landscape (5) Biodiversity (6) Addressing adverse events

Regulations (8) Regulatory measures (9) Cross-compliance (10) Land-use regulations Information Instruments (11) Research / education (12) Technical assistance / extension (13) Community based measures

3

(7) Organic (8) Reducing air pollution (9) Reduce carbon emissions (10) Renewable Energy (11) Risks to health (12) Other sustainable resources

Sub-national Environmental Agencies. Amazon Fund Guidance Committee (COFA), responsible for setting guidelines and monitor fund results, is composed of 7 Federal agencies, the 9 State administrations of the Amazon region and 6 civil society representations. 5 14 Government agencies and 18 civil society representations (farmers’ and workers’ unions, environmental NGOs, research institutes, etc) participated during the elaboration of the plan. Sources: National Water Agency (ANA), Central Bank (BC), Brazilian Development Bank (BNDES), National Food Supply Company (CONAB), Brazilian Agricultural Research Company (EMBRAPA), Brazilian Institute of Environment (IBAMA), Ministry of Agriculture (MAPA), Ministry of Agrarian Development (MDA), Ministry of Economy (MF), Ministry of Environment (MMA), Navarro (2009), Zanella (2011). Elaborated by authors. 4

6

Organics Brazilian policy for organic agriculture was initially implemented in 2003, when the first national legislation was approved (Law 10.831/2003), defining concepts related to the production and commercialization of organics. However, it was only in 2007, with the publication of an Executive Decree that the processes of certification, accreditation and other procedures necessary for the commercialization of organics were established. Nevertheless, the organic market evolved in despite of insufficient legislation, due to producers’ association efforts and driven by consumer demand. The Ministry of Agriculture is responsible for coordinating this policy in cooperation with the Ministry of Agrarian Development. The policy consists of defining labelling standards and certification procedures, activities of research and education – such as publications, training materials, consumer campaigns – and technical assistance and extension (Brasil, 2011b). Given the well known complexity of certification and accreditation processes, recently some policy innovations directed to collective certification and other mechanism have been developed. These community based measures are called Participatory Mechanism of Guarantee (Sistemas Participativos de Garantia), in which credibility of certification is built not upon rigid controlled processes from a third part certifier, but on collective responsibility of theirs members, allowing some accreditation flexibility according to different social, cultural, territorial, and institutional contexts (Brasil, 2011b). Associations or cooperatives that are able to demonstrate to a conformity body attached to the Ministry of Agriculture that their processes follow organic agriculture rules, might also use the term organic in their commercialization strategies. As a result, with the exception of standards definition, Brazilian organic policy does not employ economic instruments, such as premiums or payments based on organic farming practices. One can conclude that the incentives farmers have to switch to organic agriculture is purely defined by market forces, farmers own personal values and technical, information and knowledge restrictions. Environmental cross compliance Environmental cross compliance policies have been inaugurated in 2008, due to the ongoing effort to refrain Amazon deforestation. The most notorious measure relates to the access of subsidized credit to agriculture production in Amazon biome. In order to access this preferential credit line, agricultural producers and ranchers are requested to apply for a georreferenced database and provide information about environmental law compliance, particularly regarding the Forestry Code (Brasil, 2008a). Also, the pressure for implementing environmental cross compliance within the government-based rural credit system for other biomes such as Cerrado and Mata Atlântica is rising. The Ministry of Environment has recently launched a reference term to evaluate the efficiency of such measure for Amazon biome (Brasil, 2011c). Early analysis suggests strong evidence of leakages. Credit disbursement in that region shows that the total amount is increasing while the number of contracts is declining. It might be true that the mechanism is leading to more concentration on public credit, since the cost of compliance for some farmers is higher than the subsidy offered, make the incentive itself incapable to change practices or to help environmental regulation enforcement (ISA, 2011). Besides, credit from the government-based credit system is estimated to fulfil around one quarter of all annual financial needs required by the agricultural sector (OECD, 2005; Kumar, 2004). The other three quarters have different sources, from own private capital to direct financing from trading companies and other origins, and these are not subject to referred environmental cross compliance regulation.

7

Minimum guaranteed prices for socio-biodiversity products (PGPM-Bio) Under its traditional agricultural policy, the National Company of Food Supply (CONAB) operates a minimum-price policy for most of agricultural commodities. Being the core of its agricultural policy during the 70s and 80s, these policies were substantially reformed during the 90s towards more market-friendly interventions, reducing its operational coverage to manage crisis situations, even though it is still an important pillar of Brazilian agricultural policy (OECD, 2005). In 2008, a number of “socio-biodiversity products” were included in the list of products eligible for minimum guaranteed prices, however limiting the maximum amount of payments per person to values between US$ 200 and US$ 950 per product/year (Brazil, 2008b). This limitation was necessary to target the policy towards rural populations and cooperatives that sustainably extract forestry resources, such as Brazilian-nuts, rubber, and other 9 typical natural products that have their image associated with Brazilian socio-biodiversity. Whether to consider this minimum-price policy as an agri-environmental policy could depend on the definition of farmer, its budget allocation and coordinating agency are the same responsible for operating the traditional agricultural price-policy, demonstrating the PGPM-Bio represents an extension of this policy to different targets. Hence, policy instruments consists of premium prices based on farming practices and community-based measures, such as support to the establishment of rural cooperatives and associations. So far, no impact evaluation study has been conducted for these measures. The monitoring system relies on number of beneficiaries, value and quantity of products supported by the policy.

Low Carbon Agriculture Plan (Plano ABC) The Low Carbon Agriculture Plan is an inter-agency plan constituted of climate change mitigation and adaptation policies. The plan is one of the twelve sector plans due to the National Climate Change Policy (Law 12.187/2009) launched after the UN Climate Change Conference COP-15, held during late 2009 in Copenhagen. Brazil has voluntarily agreed to reduce greenhouse gases emissions (GEE) from a 36.1 to 38.9 percent by 2020 compared with business as usual scenario considering 2005 level (Brasil, 2011d). The ABC Plan is coordinated by the Ministries of Agriculture, Agrarian Development and Casa Civil and it is to be implemented by an Executive Group responsible for implementing Climate Change National Policy. Besides the explicit environmental objective to reduce carbon emissions by 133 to 166 million ton, other secondary environmental objectives such as soil quality and protection, addressing adverse events (climate change adaptation) and other sustainable resources practices are involved. The objectives are stated as follow: (a) To contribute to the achievement of GEE reduction as internationally compromised; (b) To guarantee the continuous and steady improvement of good agricultural practices that reduce GEE emissions and additionally increase carbon storage in vegetation and soil; (c) To incentivize the adoption of Sustainable Production Systems that reduce GEE emissions while improving revenue, principally with: (i) degraded pasture renovation; (ii) Integrated crop-livestock-forestry systems (ILPF) and Agroforestry systems (SAFs); (iii) No-tillage system (SPD); (iv) Biological nitrogen fixation (FBN); (v) Planted forests; and (vi) Animal waste treatment;

8

(d) To incentivize adaptation strategies for plants, productive systems and rural communities, in particularly vulnerable ones, to global warming scenarios; (e) To engage efforts to reduce deforestation led by livestock and agricultural production in Amazon and Cerrado biomes (Brasil, 2011d) The delivery mechanism detailed on an operational plan relies on information and economic instruments. Payments based on fixed farm assets and on investments are to be made through subsidized credit. However, there is a general understanding that credit itself does not represent a sufficient incentive for the landowners to adopt good practices that reduce carbon emissions. Government organisations has planned a number of actions related to research, education and technical assistance and extension to promote these innovations on agricultural fields and rural areas. ABC Plan environmental targets are set to two intervention periods, 2011 to 2015 and 2016 to 2020. These targets refer to the area under implementation, with the exception of the sub-programmes of animal waste treatment which refers to volume treated, and these are illustrated on the table below:

Table 2. ABC Plan Targets (area in million hectares, volume in million m3, GEE reduction in million ton CO2 eq.) Sub-programmes

Targets 2011/2015

Targets 2016/2020

Estimated GEE reduction (by 2020)

Degraded pastures renovation (area) Integrated crop-livestock-forestry systems (area)

6.0 1.5

9.0 2.5

83-104 18-22

No-tillage systems (area)

2.8

5.2

16-20

Biological nitrogen fixation (area)

1.0

4.5

10

Planted forests (area)

1.0

2.0

8-10

4.4

6.9

Animal waste treatment (volume) Source: Brasil, 2011d.

Even though ABC Plan explicit enumerates a number of environmental objectives and set targets for achievement, farmer eligibility criteria are still unclear. The Plan has the intention to be applied nationwide, but it states a priority for those areas where investments are most needed, demonstrating a concern with economic efficiency that so far has yet to be better defined. Till this date, farmer selection has been defined by financial agents who operate the subsidized credit programs without additional guidelines related to prioritization. Further, some activities related to technical and extension assistance are to be integrated within subnational governments. This division of implementation and responsibilities has been charactering the Plan since its first ideas. The ABC has been discussed in a multi-stakeholder group, involving substantial collaborations between public, private and third sectors organisations. The government expects that an amount of BRL 197 billion (approximately USD 130 billion) will be applied by the ABC Plan until 2020, in which BRL 35 billion should be provided by Federal government itself. This numbers places this Plan on the top expenditures list of Brazilian agricultural and environmental policies. There are huge challenges to the ABC Plan success but it is undoubtedly the most ambitious initiative related to agri-environmental policy held by Brazilian government so far.

9

Amazon Fund (Fundo Amazônia) Inspired by the climate change international negotiations leading to the establishment of sustainable financing mechanism anchored in avoided deforestation, in 2008 the Brazilian Development Bank (BNDES), in cooperation with the Ministry of Environment, launched an initiative to support and finance conservation projects in the Amazon forest. It consists of a fund hold by the BNDES responsible for receiving national and foreign development cooperation donations and coordinate project financing under certain pre-requisites. The Amazon Fund is strongly influenced by the Reducing Emissions and Forest Degradation mechanism, more knows for its acronym REDD+, even though primary it does not pursue to become a financing mechanism attached to the climate change negotiations, neither to generate carbon credits or other forms of compensation rights for its donors. So far, both the Governments of Norway and Germany have (Brasil, 2010b) compromised resources to the Amazon Fund, even though the majority of resources come directly from BNDES budget. The Fund can be categorized as a policy, rather than just as a development fund, because it is connected to the National Policy for Climate Change and it counts with significant interactions with other Federal agencies and States representations, and civil society organisations. The Amazon Fund Guidance Committee (COFA), constituted of these representatives from these bodies, is the forum for guideline settings and project approval, being also the forum for discussions of promotion strategies and project orientation. The environmental objectives targeted by this fund are of generic and broad nature. The legislation that inaugurates this initiative (Brasil, 2008c) states “prevention, monitoring and reduction of deforestation and the promotion of conservation and sustainable use of Amazon biome”, even though by analysing the project profiles, it is possible to notice a concentration into biodiversity protection and carbon emissions reduction. Since the Fund operates by financing and supporting a number of projects into different areas, to define which policy instruments are in use is more complex compared the other agri-environmental policies. As of July 2011, nine projects were already approved and are in execution (Brasil, 2010b). In those that have farmers are important targets of interventions, the three categories of information instruments dominates, the majority being applied to support farmers to comply with land use and other regulations. A common action is the development of georeferenced farms database with environmental information, allowing for more efficient land use monitoring. It is still possible to identify some economic instruments in those projects, such as payments based on farming practices, but these are clearly exceptions. Several organisations can apply to receive funding from this mechanism, both from public, private and third sectors. The Fund is open to Federal, State-level and local public agencies, research institutes, non-governmental organisations, private companies, cooperatives, research and technological centres, environmental associations, among others. So far, the profile of project final beneficiaries is vast, even though farmers are important actors in most of already approved plans, which allows the classification of the Amazon Fund as an agri-environmental policy. Payments for Environmental Services projects A series of decentralized and independent local projects that remunerate farmers for land retirement and for performing certain environmentally-friendly agricultural practices that secure or support the provision of environmental services are categorized as Payment for Environmental Services projects. Some of projects that are related to water services are stimulated and technically supported by the national policy

10

Water Producers, coordinated by the National Water Agency (ANA). But the influence of this policy is rather complementary, since design, management and financing are either local or regionally financed. The first experiences with PES schemes in Brazil dates from early 2000s, firstly targeting environmental services related to carbon mitigation and later to water secure and provision. These two main environmental objectives still dominates the majority of PES projects in the country, even though some projects related to biodiversity conservation can be identified, while others focus more in supporting fragile rural populations and promoting rural development. It is extremely challenging to categorize PES projects in Brazil, given it wide heterogeneity, diversity of environmental and social objectives, governance structures and financing sources. May (2011) identified 15 carbon-related PES project already implemented in the Atlantic Forest region, and another 15 which are in design and development stages. Also considering only the Atlantic Forest region, Gavaldão and Veiga (2011) identified 8 PES-water projects already in implementation and 20 in design and development stages. And Seehusen et al. (2011) identified 2 PES-biodiversity projects implemented and 3 being designed in the mentioned region. Besides, two States, Espírito Santo and Amazonas, have already implemented State-level PES programmes and policies (Pagiola 2011), while others – São Paulo, Minas Gerais and Paraná – are designing their schemes (Zanella 2011). Given the booming of these projects during the last few years, it is possible to affirm that PES has become a trendy policy instrument among Brazilian environmental managers. Once more, a complete assessment of policy instruments used in these initiatives would have to be conducted on project-by-project basis, but what really differentiate PES projects from other agrienvironmental projects is the usage of economic instruments as a conceptual pillar in their intervention strategy. Therefore, the PES focus relies on payments based on fixed farm assets, land retirement and farming practices. The vast majority also conduct considerable amount of technical assistance and extension activities, given its local profile and the common assumption that farmers’ behaviour are result of not only economic incentives. Other PES projects also complement their intervention with communitybased measures, such support to associations. Some of these policies are directed to support and assist farmers to comply with other agri-environmental regulations, such as the Forestry Code, even though PESprojects usually do not aim to establish new regulations per se (Pagiola, 2011, Zanella 2011).

Forestry Code (Código Florestal) The Brazilian Forestry Code is a half-century nation-wide legislation that, among other things, introduced land-use restrictions on private farms. It is considered as the most important land-use regulation, given its national applicability and the definition of limits to private property rights by considering forest and other forms of vegetation as public interests goods. It is also one of the most controversial environmental legislation. Decades of low or inexistent enforcement did not inhibit the agricultural frontier to expand into environmentally sensible areas defined by the Code, while other agricultural activities were already being conducted in areas that the Code considered should be restored and preserved. This chronic problem of enforcement is back again in the agricultural and environmental agendas, given wide and heat discussion within the National Congress to reform substantial parts of the legislation. The Code currently in force refers to the Law 4.771, from 1965. Its origins can be tracked to an Executive Decree from 1934, though, demonstrating an early concern with forest protection within private properties. The Law 4.771 was significantly changed through a series of Laws, Executive Decrees and emends during the 80s, 90, and early 2000’s, resulting in a complex set of legislation pieces negotiated by political parties in different historical contexts.

11

Two main land use restrictions compose the core of the Forestry Code, although many others can be considered of importance depending on local characteristics. These two are the Permanent Protection Areas and the Legal Reserve. Permanente Protection Areas (APPs) are defined as margins of rivers, lakes and other water bodies, ranging from 30m to 500m on each side as protected area, depending on the length of the water body, and 50m of protected area around water springs. Top of hills and mountains, areas higher than 1.800m of altitude, areas with inclination higher than 45°, dunes, mangroves and cliffs are also considered APPs and therefore must be protected. According to the law, the natural vegetation of these areas should be safeguarded, resulting that these protection areas cannot be used for farming, grazing or any other agricultural activity, in order to “ preserve water resources, landscape, geological stability, biodiversity, gene flow of fauna and flora, soil protection and to assure the well-being of human populations” (Brasil, 1965). Legal Reserve (RL) is defined as a share of rural private properties, additional to those defined as APPs, necessary to “the sustainable use of natural resources, conservation and rehabilitation of ecological processes, biodiversity conservation and shelter for native flora and fauna” (Brasil, 1965). This share varies from 20% of the farm in South, central and north-eastern Brazil, to 80% in forest areas of the Amazon region. Farmers are still required to provide public agents maps of the location of APPs and RL areas within their rural properties and to register these at official register offices. Even though environmentally progressive in its conceptions, enforcement of the Forestry Code has consistently failed. There are no official figures of legislation compliance, but event government officials assume that full compliance is rare (IPEA, 2011). While environmental groups reaffirm the importance of the Forestry Code and state that successive administrations had not invested sufficiently in monitoring and controlling legislation compliance (SOS Florestas, 2011), farmer groups advocate that legislation is too restrictive, far from real and practical application and too generalist, resulting economic unviable for many farm to operate full legislation requirements (ICONE, 2011). However, they seem to have a consensus on the necessity to develop more supportive information and economic policies necessary to assist farmers to adapt their agricultural practices and land uses to legislation restrictions. Recently, controversies about the Forestry Code reached top of agricultural and environmental agendas, leading the National Congress to discuss a wide and substantial legislation reform. The Deputy Chamber has approved a Law Proposal (Brasil, 2011e) that revokes all previous relevant decrees and emends into a new Law, reforming several requirements, especially those related to land use restrictions. Farmer groups dominated negotiations and are generally more satisfied with preliminary reform results, while environmental groups were claiming that undergoing negotiations represent a retrocession of the legislation (CNA, 2011; WWF, 2011). By early June, this proposed legislation was being discussed in the Senate, but expectations are high that the new Law could be approved and entry into force still in 2011.

3. Perspectives for Evaluation Recently, the relationship between agriculture and environment is on the top of the Brazilian political agenda, as demonstrated by the introduction of important new agri-environmental policies in the last few years. From the eight agri-environmental policies described, at least four of them have less than five years of implementation, while other were already into force, but gathered new momentum with current agriculture and environmental discussions. This timing context poses additional challenges on designing and implementing environmental effective and economic efficient agri-environmental policies. The following four boxes describe how the four previously described policies – Low Carbon Agriculture Plan, Amazon Fund, PES projects and Forestry Code – are planning to monitor and evaluate their effects and costs.

12

From analysing their intended monitoring plans and impact methodologies, it is possible to drawn some common conclusions. Firstly, full comprehensive impact evaluations – which would consider a complete set of criteria such as environmental efficacy, economic efficiency, equity and distributional impacts, political relevance, and analyse policy effects within time – were yet not conducted, neither is clear if there is willingness to conduct these analysis in the future. Since most of these are still in design and early implementation phases, concern is concentrated in implementation monitoring and policy analysis or ex-ante evaluations, and, except in some cases, there is no explicit mention to the intention of conducting ex-post evaluations after some periods of policy implementation. Secondly, scenarios and baselines adopted in those policies usually refer to “no policy situation” or “no intervention”, indicating that it is not prevented that impact assessments would consider methodologies that disentangle the impact of policy measures from external effects. With the exception of some PES projects – notably the State-level PES programme in design by São Paulo in cooperation with the World Bank – the reference concerning the use of counter factual analysis is rare. We believe that there are higher chances that these policy evaluations will be conducted independently by academia or other external analysts, rather than by programme managers or other personal in charge of policy implementation. Box 1. Monitoring and Evaluating – Low Carbon Agriculture Plan As of June 2011, the whole system of monitoring implementation of the ABC Plan was still under development. Great concern is given to the use of internationally recognized and approved methodologies of carbon mitigation accounting in the context of climate change international negotiations. Policy effects of ABC Plan in mitigating carbon emissions will be considered in the National Inventory of Emissions, nation-wide reports which are submitted every four years to the UNFCCC. Each one of the seven main components of the ABC programme have result indicators, usually area under implementation, which later are converted in total carbon emissions mitigated according to the international recognized methodologies. These analysis and reports will be concentrated in two “Virtual Laboratories”, research networks designed to coordinate the scientific effort and report official carbon accounting created by the ABC Plan and attached to the Brazilian Agricultural Research Company (Embrapa). Baseline emissions refer to the year 2005, the year of reference of the latest National Inventory of Carbon Emissions, and the comparative scenario is “no policy situation”. This choice reflects Brazilian strategy in international negotiations to propose that emerging countries should assume only voluntary compromises of carbon emission reduction referring to a normal tendency line, instead of liquid reductions compared to a given year. _______________________________________________________________________________________ Source: Brasil (2011d).

Box 2. Monitoring and Evaluating – Amazon Fund The most important instrument for monitoring the different projects financed by the Amazon Fund is an Intervention Logic matrix approved in 2010 that guide impact evaluation on both project and programme-levels. It was the result of a joint development of the main implementation agency, the Brazilian Development Bank (BNDES), in cooperation with the Norwegian Technical Cooperation Agency (NORAD), German Technical Cooperation Agency (GTZ), Brazilian Ministry of Environment (MMA) and Brazilian Statistical Office (IBGE). The Intervention Logic matrix is a common instrument used by many development cooperation agencies that establishes common methodologies to be applied by all projects in their impact evaluation assessments and connects different project goals into coordinated objectives. It considers four different levels of monitoring, from a macro and strategic perspective to micro project-level perspective. Concern is

13

given to identify and monitor valid indicators under a single analytical framework. At the micro project-level, it is relatively simple to disentangle direct policy effects from externally-driven effects, since the majority of indicators refers to number of projects conducted or number of beneficiaries touched by the policy. However, as usual with Intervention Logic matrices, at macro and strategic level this relationship is not explicit and no further methodologies are indicated. Thus, the focus still relies on the progress of implementation and soft effects evaluation, even though ex-post (2 years after first activity) impact evaluations are also requested for each one of the projects financed by the Amazon Fund. Nevertheless, under the current version of the Intervention Logic matrix, few details are given about methodologies to be applied in these ex-post impact assessments. _______________________________________________________________________________________ Source: Brasil (2010b).

Box 3. Monitoring and Evaluating – Payments for Environmental Services Given the decentralized characteristic of PES-policy in Brazil, no single framework unites the different environmental and social objectives of these projects and evaluation methodologies depend on each case. A wide range of methods have been proposed by project managers to conduct these analyses in their specific context: stylised environmental impact models, georeferenced environmental models, etc. Within the heterogeneity of PES-projects, it is possible to identify projects that have already accumulated some period of intervention and knowledge about its impacts, while others are still not able to identify impact indicators, relying only on monitoring the progress of project implementation. In fact, this is a core question to the future development of PES initiatives, since the whole conceptualization and argument behind the selection of these economic instruments is that payments are more economic efficient to deliver or secure environmental services than other policy instruments. Whether this assumption hold true for Brazilian PES projects is still an open question. Furthermore, only a few of PES-projects are dealing with the analytical problem of disentangle direct and external effects. One worth to mention is a State-level project currently at final design stage conducted by the State of São Paulo, Projeto Mina D’Água, in which there is an explicit interest of conducting ex-post impact evaluation in cooperation with the World Bank that uses control groups on its study-design. For this analysis, baseline data is being collected from municipalities that will participate and will not participate in the pilot-projects. Recently, project managers that are working with PES-projects have been declaring a high interest in establishing common methodologies or monitoring protocols to evaluate both PES environmental and socio-economic impacts. This jointly developed initiative would not only allow more efficient comparison between the different PES-projects impacts as it would also assist the expansion and replication of PES initiatives in other areas of the country. _______________________________________________________________________________________ Source: Pagiola (2011) and Zanella (2011).

Box 4. Monitoring and Evaluating – Forestry Code As already indicated on its description, during the last few decades the Forestry Code has faced significant problems with enforcement, which lately have generated a heated debate about which direction should the legislation be reformed. Both the current law that is in force and the proposed law that is in discussion within the National Congress define roles and responsibilities to different public agencies in monitoring and controlling regulation compliance. However, given this instable political environment regarding this policy, performance depends largely on institutional capacity of these controlling organisations at their local level and the different social, ecological and economic context in which farmers are inserted.

14

Some sub-national governments have started to place some effort in increasing regulation enforcement. Besides the PPCDAM and PPCerrado coordinated at national level, States such as Mato Grosso have also complementary state-level plans to curb deforestation and environmental impacts that relies mostly in supporting Forestry Code compliance. These programmes employ satellite monitoring combined with local interventions and georefferenced farm databanks, and some efforts of evaluating these actions are being taken by independent groups, such as environmental NGOs. Methodologies vary considerably, however more focus is given to environmental effectiveness of Forestry Code regulation, than economic efficient of making it operational. _______________________________________________________________________________________ Source: Brazil (1965, 2011) and ICV (2011).

4. Conclusions Brazil experience a dynamic moment in its policy-making debate about instruments to deal with socio and ecological interactions between agriculture and environment. New policies, programmes, projects and actions have been adding even more challenging demands to policy-makers when it comes to decide on the complex array of policy options available in agri-environmental topics. More analysis is certainly required, and there is a greater necessity of cooperating with other countries that have accumulated experience with the use of agri-environmental policies, especially those that contain economic instruments. We acknowledge two main conclusions from our succinct analysis of the eight major agrienvironmental policies identified in this paper. The first one relates to the use of economic instruments and the other to the problem of policy incoherence.

Economic instruments in agri-environmental policies In Brazil, the experiences with economic instruments are even younger than other agrienvironmental policies. Traditionally, the main strategy of environmental policy-makers has been the usage of regulations, mainly land-use policies, such as the Forestry Code and zoning policies. In many cases, these are also supported or complemented by information instruments, designed to assist farmers in complying with land-use rules. A turning-point in this tendency can be recognized in late 2000’s, as suggested by the design of PES projects, the PGPM-Bio and the ABC Plan. These three policies consider economic instruments as the core of their intervention strategies and this certainly represents an innovation in Brazilian agrienvironmental policy-making. Yet, the usage of economic instruments is still concentrated on payments, rather than other marketbased instruments, such as environmental taxes, tradable rights or quotas. A sufficient explanation of this phenomenon would require further and more complete analysis, but it well noticed that Brazilian policy, in general, has few or any accumulated experience with the introduction of environmental taxes and tradable rights in other sectors of the economy. This helps to understand why discussions to introduce environmental taxes or tradable rights in agriculture are far from reaching mainstream agricultural and environmental agendas. A great opportunity to the development of tradable rights market of forest land is being considered within the debate about the Forestry Code reform. It is well recognized that full enforcement of the Forestry Code would require that an important number of farmers would have to reforestate or retire

15

proportions of their farms to comply with the Legal Reserve requirement. An option that has been considered is to flexibilize the rules that allow farmers to acquire external forested areas from farmers that have bigger conservation areas than required by law. This is already possible under the current Forestry Code, but there are some important limitations to this practice that so far inhibited the development of a dynamic market for this particular land use right. The main challenge to the establishment of this Legal Reserve quotas market relate to the operationalisation of this scheme on a large scale that is coherent with the environmental objectives of the law. Strategic coordination and policy coherence In the eight agri-environmental policies described in this paper, at least nine different national-level organisations were identified as having direct coordinating roles. Besides those, a series of other subnational agencies also coordinate or execute part of the programmes, and most of these policies have significant participation from non-governmental bodies, both from the private and third sectors. Certainly, there are positives outcomes due to the plurality of bodies that are involved in agri-environmental policies in Brazil, but these certainly require the presence of cooperation strategies and great coordination efforts that poses additional challenges in policy making. Some of the identified agri-environmental policies do not follow a strategic framework or a national strategy that coordinates these different actions into similar objectives. In this sense, even though coherent with their own actions and programmes or even delivering results, these policies are isolated from other government strategies and might suffer from problems of lack of political support and policy incoherence, contradicting with other government policies from other domains. One example of this situation are conflicts generated by the government economic strategy of promoting beef exports by providing financing from the Brazilian Development Bank to investments in slaughterhouses and beef operations in the Amazon region. Brazilian Federal accountability office (Tribunal de Contas da União – TCU) found that these investment policies were in clear contradiction with environmental regulations, namely the Forestry Code and the Plan of Amazon Deforestation Control, PPCDAM (TCU, 2010). This example illustrates how policy incoherence not only generate environmental inefficacy, but can be a potential source for economic inefficiencies, since some of financing offered through this investment policy had later to be declined. This is certainly not the case for all agri-environmental policies. The Low Carbon Agriculture Plan is an integral part of the Climate Change National Policy, elaborated specifically as a response from the agricultural sector to the effort of mitigation and adaptation embedded in international climate change negotiations. Since its conception, the ABC Plan counted with substantial coordination from different public agencies, civil society organisations and private sector representatives. Furthermore, the cooperation between research and policy-making is one of the highlights of this plan, given the support of the Brazilian Agricultural Research Company in offering its expertise in agricultural technologies for carbon mitigation and climate change climate adaptation and the willingness of other actors to cooperate on its design. Uncertainties related to environmental achievements and high administrative and transactions costs are usually associated with implementation of agri-environmental policies, and further analysis on this issue is certainly required. OECD has built reputation on coordinating regional and international comparison, policy evaluations, development of monitoring protocols and diffusion of best practices in policy design and implementation. Further cooperation with Brazil on these topics would surely be fruitful for both sides.

16

BIBLIOGRAPHY

BRASIL (1965). Lei N° 4.771, de 15 de setembro de 1965. Brasília: Presidência da República, Casa Civil, Subchefia para Assuntos Jurídicos. BRASIL (2003). Plano de Ação para a Prevenção e Controle do Desmatamento na Amazônia Legal. Brasília: Presidência da República, Casa Civil. BRASIL (2008a). Resolução CMN/BACEN 3.545/2008. Brasília: Conselho Monetário Nacional, Banco Central do Brasil. BRASIL (2008b). Política de Garantia de Preços Mínimos para Produtos da Sociobiodiversidade. Brasília: Companhia Nacional de Abastecimento (CONAB). BRASIL (2008c). Decreto N° 6.527, de 1° de agosto de 2008. Brasília: Presidência da República BRASIL (2009). Plano de Ação para a Prevenção e Controle do Desmatamento na Amazônia Legal, 2ª Fase (2009-2011). Brasília: Presidência da República, Casa Civil. BRASIL (2010a). Plano de Ação para Prevenção e Controle do Desmatamento e das Queimadas no Cerrado. Brasília: Presidência da República, Casa Civil BRASIL (2011a). Conservação de Solo e Água. Brasília: Ministério da Agricultura, Pecuária e Abasteciment (MAPA). Available at: http://www.agricultura.gov.br/desenvolvimentosustentavel/conservacao-solo-agua, accessed in June 2011. BRASIL (2010b). Quadro Lógico – Fundo Amazônia. Rio de Janeiro: Banco Nacional de Desenvolvimento Econômico e Social (BNDES). BRASIL (2011b). Orgânicos. Brasília: Ministério da Agricultura, Pecuária e Abastecimento (MAPA). Available at http://www.agricultura.gov.br/desenvolvimento-sustentavel/organicos, accessed in June 2011. BRASIL (2011c). Estudo do cumprimento, impactos e efetividade da Resolução N° 3.545/2008 do Conselho Monetário Nacional – Termo de Referência N° 01/2011. Projeto PNUD BRA 08/003. Brasília: Ministério do Meio Ambiente (MMA). BRASIL (2011d, forthcoming). Plano Setorial de Mitigação e de Adaptação às Mudanças Climáticas para a Consolidação de uma Economia de Baixa Emissão de Carbono na Agricultura. Brasília: Presidência da República. BRASIL (2011f). Projeto de Lei N° 1.876-C, de 1999. Brasília: Câmara dos Deputados. Confederação de Agricultura e Pecuária do Brasil (CNA) (2011). Nota sobre a aprovação do Código Florestal. Available at http://www.canaldoprodutor.com.br/comunicacao/noticias/nota-sobre-aprovacaodo-codigo-florestal, accessed June 2011; Gavaldão, Marina; Veiga, Fernando (2011). “Iniciativas de PSA de Conservação dos Recursos Hídricos na Mata Atlântica”. In Brasil (2011). Pagamentos por Serviços Ambientais na Mata Atlântica – Lições aprendidas e desafios: 123-182. Brasília: Ministério do Meio Ambiente – MMA. Instituto Centro de Vida (ICV) (2011). Transparência Florestal Mato Grosso – Análise do Desmatamento e da Gestão Florestal. Ano II, n° 2, 2008/2009. Cuiabá: ICV.

17

Instituto de Pesquisa Econômica Aplicada (IPEA) (2011). Código Florestal: Implicações do PL 1876/99 nas Áreas de Reserva Legal. Comunicados do IPEA n° 96. Brasília, IPEA. Instituto Sócio-Ambiental (ISA) (2011, unpublished). Relatório Estudo sobre financiamento (agro)ambiental no Brasil – subsídio para o desenvolvimento de políticas de apoio a adoção de boas práticaas agropecuárias e regularização ambiental em imóveis rurais. Jones, Darryl (2005). “Evaluating Agri-environmental Policies in the OECD.” In OECD (2005). Evaluating Agri-Environmental Policies – Design, Practice and Results: 39-54. Paris: OECD Publications. Kumar, A. (2004). “Channeling Rural Finance”. In: A. Kumar (2004). Access to Financial Services in Brazil: 277-336. Washington DC: The World Bank. May, Peter (2011). “Iniciativas de PSA de Carbono Florestal na Mata Atlântica”. In Brasil (2011). Pagamentos por Serviços Ambientais na Mata Atlântica – Lições aprendidas e desafios: 55-122. Brasília: Ministério do Meio Ambiente – MMA. Navarro, Zander (2009, unpublished). From natural resources management to rural development: learning with the experience of microcatchment-based projects in Santa Catarina and São Paulo (Brazil). Nogueira, Jorge M.; Chaves, Flávio T.; de Medeiros, Marcelino A. A; Versiani, Letícia (2008). Ecologiceconomic Zoning (EEZ) in the Brazilian Amazon: preliminary results of an economic evaluation. Methodological Discussion and Application of Ecological and Economic Zoning (EEZ) to Urban Areas. University of Brasília and Heriot-Watt University. OECD (2005). OECD Review of Agricultural Policies – Brazil. Paris: OECD Publications. Pagiola, Stefano (2011, forthcoming). Payments for Environmental Services in Brazil. Precedings of the SEMA-SP Workshop on Payment for Environmental Services in Brazil. Seehusen, Susan E.; Cunha, André A.; Oliveira Jr. Arnaldo F. (2011). “Iniciativas de PSA de Proteção da Biodiversidade na Mata Atlântica”. In Brasil (2011). Pagamentos por Serviços Ambientais na Mata Atlântica – Lições aprendidas e desafios: 183-224. Brasília: Ministério do Meio Ambiente – MMA. Sombroeck, W.; A. de Souza, Carvalho (2000). Macro- and Micro Ecological-economic Zoning in the Amazon Region – History, First Results, Lessons Learnt and Research Needs. German-Brazilian Workshop on Neotropical Ecosystems – Achievements and Prospects of Cooperative Research. Hamburg, Sept 3-8 Tribunal de Contas da União (TCU) (2011). Relatório de Fiscalização no BNDES, BASA, Banco do Brasil, Caixa Econômica Federal e outras entidades similares de desenvolvimento. Acórdão N° 266/2010 – Plenário. Brasília: TCU World Wildlife Fund Brasil (WWF-Brasil) (2011). Código Florestal: mobilização pode barrar retrocesso, available at http://www.wwf.org.br/informacoes/noticias_meio_ambiente_e_natureza/?28683/CdigoFlorestal-mobilizao-pode-barrar-retrocesso, accessed in June 2011. Vojtech, V. (2010). Policy Measures Addressing Agri-environmental Issues, OECD Food, Agriculture and Fisheries Working Papers, No. 24. Paris: OECD Publishing. Zafaroni, E. (1998). A micro-basin approach to extension and training: experiences in Latin America. FAO Training for Agriculture and Rural Development 1997-98 (Number 55). Zanella, Matheus A. (2011, forthcoming). Why do farmers join Payment for Environmental Services (PES) schemes? An assessment of PES-water projects’ participation in Brazil. Master thesis, HumboldtUniversität zu Berlin.

18