Considerations on Further Research

Further research needs to be situated within the context of emerging trends in international politics and national forest policy developments (Humphreys 1996, FAO 1999, Michaelsen 2000, Schmithüsen 2000, United Nations Forum on Forests 2001). During the last 10 years the conditions for policy-making have changed fundamentally as reflected in trends towards globalisation, multilevel policy-networks, privatisation, and increased democratic participation. New international treaties and regulations add new dimensions to the existing patterns of governance at national, regional and local levels. The distinction between private enterprise and public administration becomes increasingly permeable and sometimes blurred. The private sector has to deal with the incorporation of external effects in management and public authorities start to work with models from business administration.

The civil society’s expectations are high and extent to new issues and in particular to environment protection and sustainable development. People want politics to be transparent and responsive to their needs and they also want the right to participate in policy decisions if they want to do so. The civil society expects that measures taken by policy-makers and public administrations are effective and efficient and, moreover, that the number of regulations and overall public expenditure will be reduced. In such an environment individual policy-actors have less power and influence to alter the existing situation by their own. Policy-makers and civil servants need to take into account the wider context both in the international as well as the local dimensions. They have to be able to interact flexibly with a variety of actors, such as citizen’s action groups, international organisations, private companies or public administrations from other domains.

In the forest sector society’s demands for goods and services are changing and growing. Forests are expected to provide timber, to protect watersheds and soils, and to provide shelter from natural disasters. They are expected to host a wide range of rare species and offer a scenic backdrop for sportsmen and city dwellers seeking tranquillity and recreation. Newly created demands, such as the use of forests as carbon sinks, illustrate that the social meaning of forests is a dynamic one. It can be subject to rapid and unpredicted expansions. Forests are only one option among several land-use alternatives which means that there is competition between maintaining the forest cover and land clearing, and between the forest sector and other sectors of the economy. In fact in countries with a rapidly growing population the conversion of forests to other land uses is often a necessity and may, if properly managed, contribute to the sustainable development of a society. However, the growing pressure on forest lands may also lead to conflicts among the users, to the disadvantage of the poorer part of the population, and to the degradation of the state of forests.

In such a context, aims, strategies and instruments are sought to overcome complex problems and to find effective and efficient solutions. Taking cross-sector linkages into account and approaching problems in a more integrative manner are key concepts for improving the result of public policies, governmental action and administrative decisions and activities. The call for these concepts can be traced in many international and national debates in search for more adequate policy development and implementation.

At the national level cross-sector linkages as an issue emanate from at least two broader discussions. First, the tasks of governments and administrations were and still are expanding. In industrialised countries political interventions have expanded from traditional fields to new areas such as trade, economic and social affairs, environmental protection and sustainable development. It also holds for developing countries and countries in transition to market economies where the state has diminished the tasks in some fields but faces new challenges and engagements in others. Public interventions are, however, not always well co-ordinated and may lead to contradictory policies. Decisions made in one public domain have usually positive and negative effects and often unforeseen repercussions in other areas. Most public policies show strong effects from decisions taken outside of their own field of competencies. The success or failure of any public policy depends to a considerable extent on the effects from external policies and from cross-sector linkages.

The types of policy instruments used are about to change, putting more emphasis on financial incentives, persuasion and procedures than on regulation. Labelling, for example, aims to influence the behaviour of timber customers by making the external costs of products more transparent. Another trend is to seek the voluntary agreements of forest owners for the establishment of nature protection zones by compensating them for income losses on a contractual basis. As measures can be more effectively if implemented by many stakeholders which understand and agree on them, procedural and persuasive instruments are more widely used. Regional planning, the Local Agenda 21, and other participatory and co-ordination mechanisms are important policy steering instruments in this context.

National forest programmes (NFP), which are actually promoted as planning instruments at the national and sub-national level in order to reach the goals of sustainable forest development use a holistic approach which is much different from previous sector planning procedures. Within NFP, inter-sector approaches are seen as a necessary core element (CSD 1997: § 10). This reflects lessons learned from formerly developed policy and planning instruments, and in particular from the Tropical Forestry Action Plan (TFAP). Experiences with the preparation and implementation of the TFAP at country level have shown that many actions failed to halt deforestation because the objectives and instruments concentrated too narrowly on the forestry sector (Humphreys 1996: 44).

So far, knowledge and experiences on how to deal with interference and cross-sector linkages from external policies is limited, also under the frame of National Forest Programmes. Some concrete recommendations have been made by FAO. They stipulate that such programmes need a multidisciplinary effort to become successful, that professionals in different fields and with different land use experiences should be associated in their preparation, and that cross-sector linkages have to be taken thoroughly into account (FAO 1996: 18, 34). FAO proposes some examples for national co-ordination mechanisms such as a high-level inter-ministerial co-ordinating bodies (FAO 1996: 31-2). For co-ordination, the importance of smooth administrative procedures, the elaboration of a well structured, transparent and accessible information and data base, and procedural arrangements for a continuing dialogue among stakeholders are stressed (FAO 1996: 47).

Ideas that linkages between public policies are important and need to be considered for improving the state of forests and for developing a sustainable forestry sector have been widely diffused and accepted in the 1990ies within the system of the United Nations and other international organisations. The UNCED sees inter-sector approaches to be a necessary prerequisite to achieve sustainable development. Enhancing policy and programme co-ordination and fostering cross-sector co-operation has become a core elements of the mandate for the newly established United Nations Forum on Forests (ECOSOC 2000, United Nations Forum on Forests 2001). The World Bank is currently reformulating its forest policy and stresses the need to contribute in the dialogue with governments to a more effective co-ordination of public policies and projects. A new and more comprehensive cross-sector approach is considered as of particular importance (World Bank's Forests Team 2000). The European Union provides an example for the diffusion of these concepts on a continental scale. A Council Resolution of December 15, 1998 on a Forest Strategy for the European Union identifies, inter alia, as a substantial strategic element "the need to improve co-ordination, communication and co-operation in all policy areas with relevance to the forest sector within the Commission, between the Commission and the Member States, as well as between the Member States".

Probably the most important trigger for policy makers to focus on cross-sector linkages was the UNCED Conference 1992 in Rio de Janeiro. In Agenda 21 cross-sector approaches are seen as a prerequisite for the sustainable development of society in general and for the sustainable use of forests in particular. The prevailing separation of different land management issues and of public policies which refer to them is considered as a reason for the lack of a development which balances economic advancement, sustainable ecosystem management and environmental protection. A more comprehensive approach is advocated which integrates economic, social and environmental requirements. National governments and the international community are advised to pay more attention to cross-sector linkages and to develop more consistent public policy frameworks for sustainable development. Section 8.4 of the Agenda 21 calls specifically for a progressive integration of economic, social and environmental issues in policy formulation and implementation. The need for more co-ordination between sectors relevant to forestry development and forest resources conservation is underlined, for instance, in Chapter 11, Sec. 31(e) of Agenda 21 on combating deforestation. It is also advocated in Section 9 of the Non-Binding Instrument on Forests as adopted during the Rio Conference 1992.

In the following-on process of the International Panel on Forests (IPF) and later the International Forum on Forests (IFF), the need for considering linkages was further stressed. Discussing steering elements for promoting sustainable forest management at a country level, inter-sector approaches and co-ordination mechanisms with any program and project affecting forests in one way or another were seen as necessary. This also reflected past experiences with Tropical Forest Action Plans, an instrument to promote the sustainable use of forests but that often failed due to lacking harmonization with other projects and policies (Liss 1999: 34). Particularly issues affecting land use, poverty, food security, energy needs and environmental protection are seen as important (Commission on Sustainable Development 1996, Intergovernmental Forum on Forests 1999). Taking the continuity of the discussion into account, it is not surprising that fostering inter-sector approaches has become a core element of the mandate of the UNFF.

The World Bank is another international actor trying to integrate forest related goals with external projects and programmes. For its new forest policy, the World Bank wants to centre on poverty alleviation, economic development, and protection of environmental values. Therein, it proposes to focus on action where the Bank can apply a comparative advantage, for example on analysing and co-ordinating policies and projects to ensure a cross-sector approach to sustainable forest management or on assisting governments to ensure that indirect and cross-sector impacts of policy and investments on high conservation and protection areas are minimised (World Bank's Forests Team 2000: 61). The need for such measures is found in the observation that the Bank Group invests far more in external sectors frequently inferring in forests than it does in direct forest projects; and that "credits schemes favouring grazing, agricultural research and technology focuses on capital intensification in frontier areas, directed agricultural settlement, and the perpetuation of uneven land distribution (...) often have damaging effects on forests." (World Bank's Forests Team 2000: 42). Addressing forest use within a multisector context has already been an issue for the actual forest policy of the World Bank. Already in its 1991 policy paper, the World Bank stressed the need for adjusting to embrace goals of the forest policy in projects such as reforms of concession policies, forest revenue systems, fiscal, tax, and agricultural policies, infrastructure and land use planning procedures, and land tenure systems (World Bank 1991: 62).

Under the title "Improving decision-making processes" the Agenda 21 states:

8.4 The primary need is to integrate environmental and developmental decision-making processes. To do this, Governments should conduct a national review and, where appropriate, improve the processes of decision-making so as to achieve the progressive integration of economic, social and environmental issues in the pursuit of development that is economically efficient, socially equitable and responsible and environmentally sound. Countries will develop their own priorities in accordance with their national plans, policies and programmes for the following activities:

Ensuring the integration of economic, social and environmental considerations in decision-making at all levels and in all ministries;

Adopting a domestically formulated policy framework that reflects a long-term perspective and cross-sectoral approach as the basis for decisions, taking account of the linkages between and within the various political, economic, social and environmental issues involved in the development process;

Establishing domestically determined ways and means to ensure the coherence of sectoral, economic, social and environmental policies, plans and policy instruments, including fiscal measures and the budget; these mechanisms should apply at various levels and bring together those interested in the development process;

(Agenda 21, Chapter 8 (Integrating Environment And Development In Decision-Making), emphasis through the authors).

Principal Functions of the United Nations Forum on Forests

2. [The Economic and Social Council] decides also that to achieve the objective, [the UNFF] will perform the following principal functions:

(a) Facilitate and promote the implementation of the Intergovernmental Panel on Forests/Intergovernmental Forum on Forests proposals for action as well as other actions which may be agreed upon, including through national forest programs and other integrated programmes relevant to forests; catalyze, mobilize and generate financial resources; and mobilize and channel technical and scientific resources to this end, including by taking steps towards the broadening and development of mechanisms and/or further initiatives to enhance international cooperation;

(b) Provide a forum for continued policy development and dialogue among Governments, which would involve international organizations and other interested parties, including major groups, as identified in Agenda 21, to foster a common understanding on sustainable forest management and to address forest issues and emerging areas of priority concern in a holistic, comprehensive and integrated manner;

(c) Enhance cooperation as well as policy and programme coordination on forest-related issues among relevant international and regional organizations, institutions and instruments, as well as contribute to synergies among them, including coordination among donors;

(d) Foster international cooperation, including North-South and public-private partnerships, as well as cross-sectoral cooperation at the national, regional and global levels;

(e) Monitor and assess progress at the national, regional and global levels through reporting by Governments, as well as by regional and international organizations, institutions and instruments, and on this basis consider future actions needed; and

(f) Strengthen political commitment to the management, conservation and sustainable development of all types of forests through: ministerial engagement; developing ways to liaise with the governing bodies of international and regional organizations, institutions and instruments; and the promotion of action-oriented dialogue and policy formulation related to forests;

ECOSOC 2000: 2, emphasis through the authors

Empirical social research displays a wide array of alternative designs in survey methods and analysis techniques. The aim is to find a custom-tailored solution that serves to answer the research questions as precisely as possible, that shows a good cost-benefit ratio, and that can be conducted with the available resources and accessible data. There is some guidance available with regard to appropriate design and methods that can be chosen for a certain kind of question. Research designs are, for example, different whether the issues to be investigated are descriptive ones or focus on a causal relationships. They are different, too, if the findings relate to single cases only or are to be generalised (GAO 1991: 68-9). In this context it is important to ask how many variables have to be looked at. Holistic approaches start from the assumption that many issues and factors are connected in some way and should be studied in a comprehensive design. Reductionism on the other hand argues that problems and issues can be divided in different parts and analysed separately. What matters is that these approaches usually work with different numbers of variables. They lead to different choices on qualitative and quantitative research approaches, Studies with a holistic approach tend to be qualitative ones (but not always) whereas reductionism in research usually leads to quantitative research methods. In the following, some possible designs are discussed and compared. They refer mainly to a holistic approach in research and examine as well to what extent quantitative methods are feasible.

Natural resource accounts (NRA) and environmental resource accounts use quantitative and qualitative information on natural resources primarily in physical terms. In the be ginning of the 1990ies, the OECD gained experiences with forest accounts in some countries (OECD 1994). In input-output tables the production, transformation and utilisation of forest resources throughout the economy was traced in quantitative physical terms. These NRA did not incorporate quality aspects of natural resources utilisation and could not analyse complicated causal relationships. With regard to cross-sector linkages NRA show to which sectors flow which quantities of wood and processed forest products (and sometimes also part of the forest cover). So far, specific national forest accounts do not put much emphasis on a more detailed analysis of flows and linkages between the forest and other sectors. They are mainly concerned with the linkages between forestry production and wood processing. This restrains the usefulness of NRA, at least at present, for policy analysis on cross-sector linkages. Meaningful national forest accounts require quantitative information not only on wood production but on the full range of forestry services. This needs a fairly detailed and highly institutionalised data base. In many countries it may be quite difficult to provide sufficient and verifiable information if one considers that already reliable data for forest related activities of farm households are hardly surveyed.

Some studies model cause-and-effect relationships in complex econometric analysis. This approach has been used, for example, in order to assess the impact of trade liberalisation agreements on environmental policy (OECD 2000). Relevant models are, for instance, the computable general equilibrium (CGE) and the comprehensive model for policy assessment (COMPASS). Without having own experiences we value such models as a useful approach for research on multi-sector policy networks and see the main application of econometric models for studying linkages with macro-economic policy. The collection of adequate data will remain difficult and a limiting point to such research.

Another feasible research design are indicator frameworks. Following clearly specified criteria, a set of indicators can be defined which are used as a reference base in order to monitor important linkages and trends. This approach has been developed, for example, in order to assess the degree of integration of environmental concerns into transport policy (OECD 1999). In the case of forestry linkages one could construct a set of indicators in order to assess the main factors of deforestation e.g. those due to agriculture, mining, forest fires etc. One could also evaluate positive effects from cross-sector linkages with indicators that assess the main factors of success in sustainable forest management, increase of biodiversity in improved forestry practices, or in protecting reserved areas and national parks. An advantage in using such a research design is that it can be custom-tailored to the research questions of a particular empirical study. The quality of the data base that is available or that can be assembled decides on the quality and the usefulness of the research findings. If these are satisfactory the qualitative and quantitative information collected for such an investigation from different sources has considerable potential for uses in other contexts.

The outcome of external policy domains as affecting the forestry sector can be evaluated by using impact study designs. Relevant research questions are, for instance, the following ones: To what extent does an agricultural policy programme encourage respectively reduce the effects of deforestation? Do wildlife management programmes significantly impede or promote forest regeneration? What are the effects of policies on the privatisation of forest lands and forest management with regard to sustainable forest uses and forestry practices. Possible alternatives of this kind are quasi-experimental, cross sectional or longitudinal section designs (Rossi et al. 1988). If data availability is sufficient quantitative studies can be carried out. The start for impact studies is usually based on a hypothesis on what the important linkages of a policy network are. It has to be confirmed or falsified during the investigation.

Case studies are defined as "an empirical inquiry that investigates a contemporary phenomenon within its real-life context (1); when the boundaries between phenomenon and context are not clearly evident (2); and in which multiple sources of evidence are used (3)" (Yin 1989: 23). Case study designs have been chosen in most of the research that has been undertaken so far. They will remain of importance in further studies considering the complexity involved in analysing interference among different public policies as well as the difficulties to assemble reliable data that allow meaningful findings with new information. Case studies could be undertaken in order to document and interpret policy processes that relate, for instance, to changes of forest area, changes in outputs from forestry goods and services, changes of the production potential in relation to defined use systems, changes in protective functions, variations in biodiverstiy in different forest management systems, and changes in nature protection and park management. It is important in future case studies not only to consider the negative effects, as has been done largely in the past, but to identify and analyse systematically the positive effects of cross-sector policy linkages that contribute to a more efficient and sustainable use of land and forest resources.

Whatever the concrete research question is, the use of case studies provides a first step to provide more empirical knowledge on the actually prevailing conditions by using multiple sources of evidence such as documents, legislative texts, available quantitative information as well as the opinions and judgements of the stakeholders involved. The use of a range of information sources will contribute to a better understanding of structures and issues that result from complex policy linkages. However, a major aspect one has to be aware of in using such a research design is the problem of how to compare and put into a common frame of interpretation the findings from different information sources. The difficulty of developing weight factors for the various observations and findings is probably the limiting point in using the results from case studies. This is particularly true if one wants to compare and evaluate different studies in order to draw more general conclusions in a more comprehensive context.

Expert panels offer an opportunity to policy makers, professionals and stakeholders to present their views and use their experiences on what they consider as relevant policy linkages and the likely effects which they may generate. Such panels can be organised in face to face sessions with various participants or indirectly through written communication or e-mail exchanges. Usually several rounds of exchange of opinions and statements are needed to have sufficient time for summarising and comparing the different views and for letting the participants find a consensus on what they think to be the important issues and possible conclusions (e.g. Hopple and Kuhlmann 1981, Linstone and Turoff 1975). Quite a comprehensive study based on expert opinion has been conducted recently in order to assess the sustainability of Swiss forest policy (Limacher et al. 1999).

An argument in favour of this kind of research is, that the danger of subjectively biased results is reduced through communicative validation within the panel. Forthcoming results have a chance to be more widely accepted in view of a number of different opinions from various groups of stakeholders as compared to studies elaborated by one single researcher. This is an important aspect in circumstances where very different and controversial problem, perceptions and concrete interests prevail. Expert panels as a research design require comparatively small inputs which is an advantage if only limited resources and data are available. The critical point, on the other hand, is to provide a meaningful composition of a panel and to find competent staff for organising the sessions and interpreting the findings.

In comparing the usefulness of various research designs one can refer to criteria such as the necessary requirements to carry out a study, the significance and informative value of the findings, and the resources respectively the costs of the undertaking (Table 5). For the requirements of different research designs one can state that quantitative studies need previous knowledge of the type of relevant policy linkages and their likely effects as well as sufficient and confirmed data. Findings are restricted to quantifiable policy linkages. Qualitative research designs do not necessarily require an intensive previous knowledge on the nature and structure of the policy context. Quantifiable information and the data are one element of the research. Expert panels require less detailed knowledge of the empirical background of the investigation since a considerable part of it comes forth during the research process itself. This is different with case studies which need to be conducted by researchers that are familiar or can familiarise themselves with the subject or field of investigation.

With regard to the significance and informative value case studies, expert panels and econometric modelling are the explicit ones. They may provide valuable information on direct and indirect policy linkages, a fairly detailed picture of all of different factors at work, and of the various actors involved. Indicator frameworks and quantitative impact assessment are useful in order to illustrate the influence of certain variables in the policy network and to monitor their performance. Due to a widely spread faith in the significance of numbers studies that quantify linkages are likely to have more attention from the public and within the circles of policy-makers and analysts than qualitative research. However, one has to keep in mind that quantitative investigations may risk to fail in presenting the overall context and the empirical complexity of the whole web that usually results from cross-sector linkages. Considering the significance of National Resource Accounts it appears that they are at least until present of less informative value since their information tend to be limited to wood flows in the forest and wood processing sectors.

Considering the required resources respectively the cost of various research designs one may state that all of them need different kinds of expertise and that the costs will depend on many factors and vary considerably. As a general rule the costs of national resources accounts, econometric modelling, and indicator frameworks will be rather on the high side whereas impact and case studies or expert panels will be in the medium or lower range.

Comparing the different strengths and weaknesses of alternative designs one should try to combine some of their relative advantages and informative value. Further research could start with case studies and lead to quantitative impact analysis or to econometric modelling at a later stage. Qualitative studies can provide recommendations for the quantitative monitoring over longer periods of critical effects that result from particularly relevant policy linkages.

Table 5: Comparison of Research Designs on Cross-Sector Linkages

It is proposed to continue further research on the importance of cross-sector linkages and public policy networks mainly through qualitative case studies. This allows to draw a more comprehensive and empirically based picture that includes the most relevant factors and the direct and indirect cause-and-effect relationships. The aim is to provide an actual, insightful and thorough understanding of the problems but also of the opportunities for a limited number of countries in their specific regional context.

The six typological criteria discussed in section 1.3 can be used in order to formulate relevant research questions. The starting point will be in any rate the identification of the most relevant public policy domains and the type of linkages which interact with forest policy programmes. This allows to obtain a more solid knowledge of the forest sector’s institutional framework which results from macro-economic, privatisation and land tenure policies. Public policies related to economic sectors as agriculture, mining, energy and transport, as well as those addressing rural and urban development, and environmental and nature protection will have to be considered. In order to obtain an accurate assessment of the important cross-sector links the analysis of chosen cases should not be restricted to a single regional context, direction and valence. Research issues should not be limited to negative effects from external policies. They should be based on a more comprehensive view on the actual and potential positive impacts on forestry programmes that can be associated with external public policy domains. Different levels of decision-making and different types of instruments will have to be examined.

A case study design that fits the best these objectives at country level can be based on already available qualitative information and data from quantitative studies, and be supplemented through more intensive document analysis and expert interviews. Researchers with field experiences in a certain region can be mandated to examine previously documents and findings and to undertake supplementary research within a defined set of policies and linkages to be analysed subsequently. It is important is to show the actors and stakeholder involved, and the mechanisms through which influences are exerted. Causal relationships between forest and other policy domains and the links and effects acting in both directions need to be explained.

The proposed case studies provide findings that are of use for concerned stakeholder groups, policy and decision makers, government officials in their respective sphere of competencies and the public. They will increase the understanding of cross-sector linkages as a decisive factor in forest conservation and forestry development and demonstrate the need to look beyond sector limits and boundaries of administrative bodies acting as if they could move in an interdependent policy environment. They will also be suited to provide new findings on complex cause-and-effect relationships. A phased approach in carrying out the case studies allows in a first stage to select polici es and types of linkages that are particularly important in the empirical research context and to focus subsequently on those that merit further analysis. Based on the findings monitoring systems of policy development and interference can then be established which are of great importance for the implementation of more co-ordinated public measures related to sustainable land and forest uses.

The proposed research framework is illustrated in Figure 4. Focussing on an actor-centred approach the behaviour of forest owners and forest users is considered as the key determinant for evaluating the impact of public policies on the state, use and management of forests. As the particular qualities and characteristics of the prevailing ecosystems determine to a large extent the scope of action of forest owners and users, the relationship between forests and forestry actors is reciprocal. In addition, the state of the forest is largely influenced by natural processes and hazards. The socio-economic factors that influence the actions of forest owners and users have been identified on the left side of the figure They are largely determined by social and cultural values, forest and other relevant public policies, and the national and local economies. All of these factors are interwoven with eachother.

The central content of what is important in studying forest related public policies is thus embedded within the general framework. Cross-sector linkages refer to the effects of all relevant public policies on forest owners and forest users and through their actions and decisions on the extension and state of forests. The principal research question is what public policies mean exactly in a given context and to what extent they exercise positive and negative influences on forest owners, forest users, and on forests. The following considerations are relevant in this context:

The impact of a policy on the behaviour of forest owners and users forms the primary cause-effect relation (Knoepfel et al. 1997: 103). Policy impact means of course not only an influence on identifiable actions, it may also imply lack of influence in orienting or changing behaviours of the target addressees. In this context, some behaviours will be particularly important to observe like forest clearing, timber harvesting, uses of non-wood products, utility of forest services, management practices, and measures of biodiversity conservation and nature protection.

The outcome of a policy on the state and development of forests is in to fact be considered as the ultimate policy goal and a kind of derived effect. How are the forests influenced through the behaviour of forest owners and users which are addressed by the objectives and instruments of the prevailing forest policy as well as by those which result from cross-sector linkages? The answer to this question characterises in fact the actual net effects of a policy network related to forests and forestry (Rossi and Freeman 1993: 214). Interesting effects that can be observed are, for instance, the size and distribution of forest land, the stand volumes maintained, the richness of different species, and the sustainability of forestry practices.

Figure 4: Framework Explaining Behaviour of Forest Owners and Users

The focus of the case studies is on public policy networks since sectors are always components of a more comprehensive economic, social and political pattern. A sector view only would not allow to analyse and understand all of its relevant effects and impacts. Arguments in favour of such an approach are the following ones:

The users of the findings from the country studies will probably be mainly policy makers and government officials from ministries and administrations concerned with forestry development and conservation as well as from other ministries and public authorities engaged in the promotion of productive land uses and environmental protection. The empirical findings on the combined impacts of various policies will contribute to and increase the understanding for more coherent implementation activities, decisions and plans of the public administration as a whole.

Public policies are the domain where the state and the competent ministries have the possibility to influence the actual behaviour of forest owners and users.

It is also necessary to reduce the complexity of a model to a level where meaningful new insights can be gained with a reasonable amount of resources for the research to be undertaken. Focussing on public policies appears to offer a reasonable balance and a feasible research framework. The major components are the analysis of the policy goals and instrument, of the policy impact on target groups, and of the net effects of cross-sector linkages on forests and forestry.

Studying policy goals and instruments is the initial steps of a policy programme analysis. It is based on an examination of the content of relevant legislation, official administrative documents, and budgets analysis (Knoepfel and Bussmann 1997, Knoepfel et al. 1997, Tsering 2000). The analysis should first show what the official public policy aims are with regard to forests and to the behaviour of forest owners and users. Second, it should made clear with which instruments the public policy plans to achieve these goals. However, concrete goals for the state of the forests or the behaviour of forest owners and users may not always exist. Their introduction would evoke protest from interest groups having other goals for the forests, or from forest owners and users trying to optimise their own scope of action. Knowing the goals of forest policy will be necessary to assess if an effect of external policies is positive or negative. If the forest policy has no clear goals, the cross-sector linkages are difficult to be specified. Comparing goals and instruments of external policies with those of forest policy can reveal incoherence and inconsistency. The limitation to this approach is, that one can not evaluate the importance of instruments for changing a behaviour or for solving a problems. In order to validate the findings, interviews with representatives of public administrations, land owners and land users as well as with other stake holders will be useful. This will decrease the danger of interpreting statements on goals as relevant that are in reality only of a declamatory nature.

The impacts of forest and external policies on the behaviour of forest owners and users have to be studied next. Behaviour of and influences on the actors depend on specific attributes. They cannot be studied in the same manner, for instance, for a large scale forest company, for farmers or for rural dwellers. Interviewing representatives of different actor groups reflect the perceptions and opinions of these actors on what they consider important policy influences that affect their behaviour and land use strategies. This allows to rate which policies are considered to have the most significant impacts. However, it does not provide the full picture in as much as factors which do not find attention among the actors may have a decisive influence as well. Another approach is based on case studies (e.g. Yin 1989) which use different sources of information. In the setting of rural and developing communities, ethnologic studies working with participatory observations may be appropriate (Jorgensen 1989).

Studying the net effects of policies on forests and forestry will be rather complex and costly. The choice of an appropriate research approach on net policy effects will depend on many factors and some of them may not yet be known. This refers, for instance, to the type of relevant policies, to the actors involved, and to the social and economic contexts in which these policies operate. It also applies to data availability and to the available resources to conduct the studies. It is difficult to measure net effects, i.e. those outcomes of a policy "that can be reasonably attributed to the intervention, free and clear of the effects of any other causes that may be at work" (Rossi and Freeman 1993: 221). Possible research designs are randomised experiments, quasi-experiments, simple before-and-after studies, cross-sectional studies, panel studies, and time-series analysis (Rossi and Freeman 1993). As these designs are time-consuming and expensive, one could opt for a judgmental approach and rely on an appraisal through experts who know the issues from field visits and personal experience. However, such an approach has to be rated "among the shakiest of all impact assessment techniques" (Rossi and Freeman 1993: 252).

Illustrating different kind of co-ordination mechanisms or network management: The ways of managing public intervention over the boundaries of different public policy domains need to be examined. Typical questions to be asked are, for instance: How do different agencies actually work together ? What agencies have which resources ? Are there contradictory and/or overlapping competencies at the policy setting and the implementation levels ?

Co-ordination mechanisms or network management include a wide array of possible strategies. In order to inspire readers for strategies, case studies should illustrate several approaches with different policy outputs and outcomes. Preferably, successful or rather successful strategies and cases should be compared with less successful ones. We propose to concentrate on case studies with the following institutional aspects:

Co-ordinated planning and consultation procedures

Inter-ministerial co-ordination mechanisms;

Establishment of fora for creating informal contacts (such as the Forest Forum for Decision-Makers in Finland);

Integrating issues of external domains into the activities of forest administrations (e.g. "greening" of forest administration);

Allocation of resources to forestry departments in relation to their tasks and as compared with those of other departments with important cross-sector linkages;

Research designs will vary for studying network management or inter-administrative co-ordination, the later being less standardised.

Studying policy network management can be done through a formal quantitative network analysis or through a qualitative analysis of the structures in place. The number and categories of actors and stakeholders involved in a determined policy domain can be assessed through a literature review and/or through selective interviews with experts. A formal network analysis uses a standardised interview of actors on their relation with other actors. This data is then analysed quantitatively and as a result networks can be graphically plotted (Knoke and Kuklinski 1982, Knoke and Kuklinski 1991). A qualitative analysis does not use a standardised interview methodology and relies only on selective qualitative interviews. The researcher will have to draw the network boundaries and the relations between different actors according to his judgement and interpretation of the qualitative interview statements. . Both approaches can be combined in studying the formal ways actors meet and share information.

For research on inter-administrative co-ordination, no single and commonly used methodology exists. Case studies using multiple sources of evidence (literature review, analysis of law and documents, expert interviews) are probably most appropriate. The actual mechanism to co-ordinate different agencies have to be studied, as well as the ways in which information is shared and consultation organised. Financial and human resources of different agencies have to be compared and their adequacy to be assessed in relation to the objectives of public policies for which they have governmental competencies. In order to make a judgement on the openness of public administrations for operating in more concerted policy networks , the educational background of the members, their perceptions and attitudes towards sustainable land management and conservation, as well their actual behaviour can be studied.

Policy and decision makers could benefit from empirical country studies of this kind at least in the following two respects:

Ways and means to effectively co-ordinate different policies are illustrated and will be better understood. Policy-makers can be sensitised with regard to the strategic points that need more attention in developing positive linkages among different policy domains and to reach better co-ordination among public administrations in charge of implementing the goals and objectives from such policies. .

The future role and potential of forest policy and forestry administrations in sustainable land and natural resources management, and their contributions in a multi-sector policy network will be reflected more clearly. In some cases the actual findings may show that the forest administrations are not yet in a position to make effective contributions to sustainable the development of renewable resources, especially if one compares their possibilities with those of administrations in other forest related public policy domains. This again will call for an institutional strengthening of the forest sector, to more co-ordinated public policy networks in sustainable land management as well as to more specialised and pro-active public forest services operating in a a constantly evolving network of public policies with important cross-sector linkages.

Important trends such as globalisation of the economy and of trade , internationalisation of environmental and nature protection, privatisation and a changing understanding of the role of the state, increased democratic participation of stakeholders concerned, a growing influence of non-governmental organisations in public decision making processes, as well as an increase and diversification of society’s demand for forest goods and services form he context for further research on forest related public policies and cross-sector linkages in the domain of forestry. Such research has not only to be concerned with sector linkages relating to wood production and sustainable forestry practices at national level. Its perspectives extend to a growing range of public and private actors interested in many other forestry goods and services as well as in maintaining ecosystems biodiversity and in protecting cultural values associated with trees and forests. Policy processes and linkages at sub-national and local levels as well as those at a regional and world-wide scale need to be considered.

Agenda 21 and the follow-up processes of Rio set the international framework for an emerging international regime on forest conservation and forestry development. They establish s a reasonable balance of economic , social and ecological factors as the primary aim of forestry development. Many public policies and cross-sector linkages have to be considered in order to improve the frame conditions for sustainable forest resources management as part of changing and evolving land use patterns. If the effects of external policies on the forestry sector have mainly found attention so far, it now time to investigate to what extend public forest policy is of importance to and has effects on other policy domains within the global context of sustainable development of societies.

Investigations on cross-sector linkages in forestry have already been undertaken and need to be continued with more elaborated research designs such as natural resource accounts, econometric analysis, indicator frameworks, and impact studies. As important as improving appropriate research methods is the necessity to launch specific case studies at national, sub-national and local levels which provide more empirical information on success as well as on draw-backs in a given social, economic and political context. Quantitative data based research is needed as much as qualitative analysis in order to provide more information on the nature, structure and functioning of different policies and cross-sector linkages.

If relevant effects from external policy linkages on the forest sector have been examined with an increasing interest, the same cannot be said with regard to the management of such linkages and of complex policy networks. There is a considerable interest to engage this kind of research more consistently and to investigate how co-ordinating mechanisms such as network management and inter-administrative co-ordination can be improved and contribute to reach national forest policy goals. This implies to reconsider the current role of forest administrations and their ability to operate with success in a given policy and administrative setting.

Cross-sector linkages and co-ordinating mechanisms are complex and their environment is dynamic and forces at work can probably not be controlled in one single study. The complexity and instability of the situation bears consequences for a feasible research approach. By now it is difficult to imagine anything else than a qualitative and open approach, concentrating on a series of case studies. It is important to organise the research work in such a manner that forthcoming empirical findings will be comparable among each other.

Further research should identify in a first stage the most significant policy domains, the kind of linkages that are most important and the scale of positive and negative effects which are likely to result from them. The criteria for selecting the major factors to be researched subsequently in more detail should be the usefulness of new knowledge gained to the main addressees of these studies, and in particular to the representatives of national public administrations and policy makers. It is advisable to start with cases studies on a national level that analyse on the linkages between forest policy options and one or two external public policies.

Researchers with country or regional experiences will be needed in order to make a meaningful selection of relevant cross-sector linkages, to carry out consistent document analysis, to organise expert interviews, and to resume and interpret their results. The focus of the research should be on the actors and stakeholders involved, the instruments and procedures that influences their behaviour, and on the causal relationship between forest and other policy domains in both directions. New approaches in co-ordination mechanisms as well as the likely limitations of co-ordination need to be examined.