Events

Decision Trees - ECTF Workshop, Edinburgh

25th September 1998

A one day workshop chaired by Marcus Sangster - ECTF (FC) & Gill Shepherd - ODI.

Seven talks and discussion sessions were followed by a panel discussion session with all the speakers which produced some lively debate. Summaries of the talks are below.

1) Farmer Responses to Incentive Payments for Forestry

Bob Crabtree , MaCaulay Land Use Research Institute, Aberdeen AB41 7DR UK

Since 1987 the UK government has operated a farm woodland policy which uses incentive payments to encourage the establishment of woodlands by farmers. This paper first examines how a utility maximising farmer would react to a given set of payment incentives, assuming that there was full information and that the tenure conditions permitted afforestation. The decision would depend on the profitability of forestry and farming, the incentive payments offered, any additional costs imposed by the policy, and the value of additional personal benefits derived from, for example, sport and amenity.

In order to assess how these various factors influence uptake of forestry schemes in practice, an assessment is made of farmer objectives when entering the Farm Woodland Premium Scheme (FWPS). A survey of entrants to the scheme in Scotland indicated that income and timber motives were only important in a minority of cases, the main reasons for participation being to obtain benefits to farming through shelter, and private benefits from wildlife, landscape and amenity. Farmers on better quality land were less interested in the benefits from shelter and more concerned with sport and amenity than those in the less-favoured areas. A logit model was applied to the data to determine which contextual factors were good predictors of uptake. Farm size and the proportion of the farm area in existing woodland were significantly associated with an increased probability of entry into the scheme. In addition certain types of farms (cropping, cereal and mixed farms) were more likely to join than dairy or pig farms. The model was able to allocate farms into entrant and non-entrants with a reasonably good degree of precision, and this provides policy makers with an understanding of the factors associated with uptake of the scheme at the Scottish level.

The paper finishes with an examination of the importance of farmers’ behaviour for efficient scheme design. It assesses the success of the scheme in attracting farmers and the reasons why planting levels have been maintained over time. It also examines two aspects of efficiency in the design of voluntary schemes. One is the additionality of the FWPS, that is the extent to which the scheme encourages planting that would not otherwise have taken place. The second is the extent to which payments are unnecessarily high, such that a lower payment would have attracted a farmer to enter the scheme. The paper finishes by assessing the potential for improved efficiency by increasing additionality and by the re-structuring of payment rates.

2) Community Decisions on Forestry in Scotland

Bill Slee, University of Aberdeen

Forests have widely acknowledged public good characteristics, which affect both geographically adjacent communities and those further afield. A number of mechanisms exist to incorporate these public good and other concerns of the public into the appraisal of forestry proposals. These mechanisms include formal statutory procedures, informal mechanisms and, increasingly, a range of quasi-formal non-statutory procedures.

Decisions on forestry are regulated formally in a variety of ways, including through Environmental Assessment, local authority Indicative Forestry Strategies and through the mediating actions of Regional Advisory Committees. These statutory procedures usually provide formal channels for participation by interested citizens.

Community involvement in forest-related decision making has also been promoted directly and indirectly by activities surrounding the development of community forestry, although such activities have usually been geographically constrained to urban fringe locations. Beyond the formal statutory channels, community involvement in forestry proposals and practice in more deeply rural areas has generally been restricted to modest informal consultation.

However, over the last decade, there has been growing evidence of pressure group activity associated with the promotion of greater community involvement in forest-related decision making in Scotland. There have been a number of instances where communities have engaged in actions aiming to influence forest management decisions, including engaging in formal partnerships and even the community purchase of forest land. At the same time, rural communities have viewed with dissatisfaction the rapid reduction in local employment associated with technological change in the forest sector and changes in the administration of forests which have often made forest managers more remote from the forests they manage.

Rural communities tend to place two, possibly contradictory, demands on forests: they wish them to provide employment; and they wish them to deliver amenity benefits to the local population. The extent to which forests can satisfy both of these aspirations is unclear, as are the mechanisms that will allow the representation of different and sometimes divergent stakeholder interests.

The Forestry Commission has recently (July 1999) launched guidelines on community involvement with forest-related decision making. The implications of adopting these guidelines will be considered and the extent to which they are likely to satisfy the demands of a citizenry which shows highly variable interest in forest-related matters will be assessed.

3) Policy makers decisions on farm and community forestry in the EU

Pentii Hyttinen, European Commission; Gerry Lawson, ITE; Terry Thomas, UCNW & a.n.o., Spain.

Forest policy decisions are influenced by a range of international, national and local issues, but it is seldom apparent that quantitative tools have been used to examine the consequences of financial and fiscal changes before they are applied.

This paper has 4 sections:

History of Government and local authority support for farm forestry and community woodlands in the UK, Finland, Spain and the EU as a whole. The climate, forest cover, farming pattern, and local-government structures in these 3 countries are quite different. Yet they now exist within a common Agricultural Policy and Forest Strategy. Which policies have been most beneficial to farm and community forestry in the past?

Regional variations in the approach to farm and community forestry and to the implementation of EU Regulations. Most countries in the EU devolve the major part of forest policy decisions to autonomous regions, and some also have significant involvement of local authorities. Forestry Regulations in the EU (e.g. Regulation 2080/92 on afforestation of farmland) have demonstrated a great degree of 'subsidiarity' or freedom for individual states or regions to decide on the degree and nature of implementation of guidelines. Is this flexibility likely to be continue under the CAP Agenda 2000, and what farm forestry incentives are we likely to see?

Capacity of current Farm Accountancy Data Networks to quantify current patterns of farm and community forestry. The 3 countries have different approaches to the collection of farm forestry statistics, but share common problems of: defining 'farm woodlands', less detail in censuses on forestry operations than on crops or animals, sparse information on part-time farm-foresters or on enterprises where forestry is the main 'business', and biased sampling patterns (e.g. voluntary returns).

Use socio-economic tools to assess the impact farm-forestry policy changes on individuals, communities and the environment. Given adequate data, can models of the profitability of individual representative farms can be scaled up to predict the likely uptake of farm-forestry incentives? Are these models useful for community forestry? How can the non-market benefits of farm-forestry be factored into policy-makers decisions?

4) Decisions on Farm Forestry Optimised using Linear Programming - Examples from three Agro-Ecological zones in Ghana.

Roy Fawcett, IERM, University of Edinburgh
Julian Smith NRE, Scottish Agricultural College
& Jim Wright, IERM, University of Edinburgh

This research illustrates why the effective modelling of decisions made in subsistence farming systems require the full integration of information on domestic household activities including leisure, recreation, field work, food processing, marketing, transporting, hand crafting, hunting, gathering and working for others.

Emphasis is placed on the ‘Livelihood Strategies’ employed by the stakeholders which express the complex web of requirements, opportunities, resources, tastes and preferences. These strategies can be quantified to provide a basis for optimising the spatial and temporal allocation of activities for individual family members as components of a household collective. Within the collective, individuals may have varying degrees of autonomy and control over land and labour resources.

A combination of household survey data and participatory planning exercises are required to validate the information base and establish the range of objectives and motivation expressed by individuals. The participatory planning activity is required for the finer detail in spatial and temporal resource endowment and use. In particular, it was discovered that the gender specific basis of labour allocation has to be flexible in response to the age sex composition of households.

Anecdotal evidence has shown that households rarely behave as profit maximisers. There are many societal, household and individual goals, needs and wants which must be satisfied in addition to the objective of ‘earning a living’. Techniques of Multiple Criteria Decision Making (MCDM) are used to effect compromise in the face of conflicting objectives. Maximising the marginal value productivity of disposable time subject to numerous constraints representing individual tastes and preferences lies at the core of the modelling activity. Results show that taste and preferences have a significant influence on the optimal deployment of disposable time.

The complex spatial distribution of property rights requires participatory mapping in order to construct the appropriate matrix coefficients to express the vector geometry. Distance or time to a particular location without a direction component is insufficient to solve the time allocation problem.

Where food and medicines from the wild are still a significant component of domestic activity, there is a dearth of information on productivity trends and the impact that increasing population density will have on these common property resources. It is suggested that locally based management plans are required to conserve and maintain the productivity of these resources because they play a significant role in the local carrying capacity of the local farming system.

Output from the modelling activity could provide prescriptive support to professionals in the field and extension services supplying stakeholders with advice. Portfolios of possible enterprise activity mixes can be developed for a range of scenarios thus providing support information for a selection of ‘what’s best’ strategies which have been matched to local situations.

5) Modelling Strategic Decisions at the Tropical Forest Frontier

Mandy Haggith & Robert Muetzelfeldt. IERM, University of Edinburgh

FLORES (the Forest Land Oriented Resource Envisioning System) is an ongoing process aimed at creating a model of land use change at the forest frontier, with the ultimate goal of providing policy-makers with a tool for exploring alternative policy options. One of the fundamental principles of the FLORES modelling process is that land use change results from the decisions made, and actions taken, by individuals and households, and thus their decision-making processes must be included within the model if we are to accurately simulate the impact of policy decisions on land use.

The model consists of an ‘outer wheel’ dealing with the core issues of strategic decision making and land allocation which are decided on an annual basis, and an ‘inner wheel’ of operational decisions taken weekly. The strategic level covers 5 types of process:

• Population dynamics: births, deaths and marriages.
• Land tenure and inheritance: tenure rules and rules of inheritance.
• Strategic planning: planning to acquire/dispose of land, or migrate.
• Land and labour re-allocation negotiations: negotiations between households; or households and villages.
• Land use planning: planning to clear/cultivate plots; yearly workplan.

The weekly processes by households include:

• Perception of conditions on their plots of land.
• Choice of crops, where appropriate.
• Labour allocation to activities based on the household’s strategic plan.
• Consumption and conversion of yields from the land.

The overall structure of the model, and the form of its components, is intended to be as generic as possible. The model includes about 20 settable parameters and definable functions to enable the model to be callibrated for particular sites. A set of policy levers and indicators provide user inputs and outputs. To a large extent the model specification is defined in terms of heuristic rules which model decision-making in terms of agents reasoning towards conclusions. To complete the specification, site-specific heuristics and parameters have been derived from research papers on the Rantau Pandan area of Sumatra.

The model will be evaluated by including it as part of FLORES and calibrating the whole FLORES model for the Rantau Pandan area. It will then be tested in other areas of the tropics.

6) Community Decisions on Forestry in Karnataka

Mike Harrison & Dermot Shields LTS International, Bush Estate Edinbrugh EH26 0PH

Reform of forest sector management in India has taken the form of Joint Forest Management following the New Forest Policy of 1998, which recognises and promotes the role of local communities in the management and protection of forest resources. The rationale behind the approach is that, under ‘benefit-sharing’ arrangements between the state and the local community, the net returns to the local community are sufficiently attractive to ensure a convergence between local interests and national policy. Ultimately, this should ensure national forest conservation, whilst providing a sustainable flow of products locally.

The paper addresses this issue directly, by considering the impact of the new institutional arrangements on the livelihoods of four local communities: two in the Western Ghats Forests of Northern Karnataka and two others, one in the ‘sal’ forest region of West Bengal and the other in Gujurat State. The paper goes on to consider whether these incentives apply separately to different user groups within the community, and overall whether the returns are sufficient to ensure the long-term sustainability and conservation of the forest.

The analysis is based on construction of a community forestry simulation model, using (a) existing users and the status of the sites used, (b) assumptions about the growth of forest and associated non-forest products on each site; and, (c) financial and economic values, derived from both market and imputed values. The simulation model is used to estimate the incremental net returns to different benefit-sharing and management decisions. The distributional impact of specific management decisions on different classes of forest users is determined based on assumptions about product flows. As well as simulating the resource returns to the community, the model is also used to consider the cash flows into the community fund, and the revenue flows to government.

The models show positive returns under certain conditions. However, there is considerable variation between sites indicating wide variation in effect. This is especially true with respect to the distributional impact of improved forest management where reduced access can have an immediate negative effect on the poorest groups such as fuelwood ‘headloaders’ and people collecting leaf litter.

The approach used has resulted in greater understanding of the impact of policy decisions and institutional arrangements by identifying critical factors and explaining variation in outcome. This should contribute to and improve the current policy debate.

7) Policy-Makers Decisions on Farm and Community Forestry in India

Neil Thin, University of Edinburgh, Department of Social Anthropology

This paper will analyse the processes of forest policy makers' decision-making by looking at one particularly important and explicit policy arena: Joint Forest Management in India (and related policy issues such as eco-development, 'social forestry', and forest management in general). It will discuss theoretical concerns about the processes by which policies emerge, and about the policy instruments by which objectives are negotiated and announced, and by which strategies are devised and implemented. It will focus mainly on the articulation of policy at State, national, and international levels, looking at the influences on both explicit and implicit policies.

To relate policy processes to the specific theme of decision-making about forestry, the analytical framework for the paper will address four sets of issues: choices, people, objectives/outcomes, and processes:

1. Choices. To analyse decision-making we need to identify and select among the topics about which choices are available and hence about which decisions are made - in this case: land-use alternatives; rights over land (ownership, use/management rights, access); revenue-sharing and incentives to various stakeholders for various management alternatives; species and species mix preferences; forms of institution and inter-institutional relations for forest planning and management (including policies on recruitment, training, and staff management in the Forest Department); harvesting regimes; labour allocation and employment provision; regulation of prices and marketing; regulation of forest-based industries; transport regulation. To make sense of this diversity of topics about which decisions need to be made the paper will draft a hierarchy of choices related to the cause-effect hierarchy which planners would use to identify problems and strategies for achieving objectives.

2. People. We also need to conduct basic analysis of the stakeholder categories relevant to policy decisions: decision-shapers, decision-takers, those about whom decisions are made, and those (including unborn generations) who are affected by the outcomes of decisions. And our analysis of policy decisions needs to look at relations among these stakeholders, their overlapping and competing interests, and the at the need for changes in those relationships if policy outcomes are to be effective and fair. The paper will attempt to show how our understanding of decision-making can (and must) be informed by analysis of the political ecology of knowledge management and policy making.

3. Objectives and outcomes. The paper will also discuss the extent to which policy makers make explicit the objectives which theoretically should guide their decision making, and whether they are yet showing signs of being interested in actual outcomes. Distinctions will be made between maintenance policy and gradual improvement policy and radical improvement policy, and between intended outcomes and unintended outcomes.

4. Processes. Rather than taking for granted that decisions are identifiable and explicit moments when one alternative is preferred over others, the paper will examine decision-making as a set of processes which include the articulation, concealment, arbitration and management of conflicts, and efforts to reach consensus. Focusing on policy makers' decisions, the paper will take a broad definition of 'policy' to include systematic patterning of ideas and behaviour directed at specific outcomes - whether the policy is explicit or just implicit, clear or ambiguous, formal or informal, sudden or emergent, agreed or contended. Two related issues will be the relation between Forest Department policies and those of other stakeholder categories (other government agencies, industries, civil society organisations), and between decisions which are explicitly taken and those which are either not taken or are taken by default through negligence or avoidance. 

Copyright © 2005 ECTF.