Farm household level optimal resource allocation. An explorative study in the lime stone area of East Java.

The role of agriculture changes markedly when an economy is transforming from developing to developed. The dramatic structural transformation in the development process has consequences for the policies to be implemented. Acknowledging differences between countries, generally speaking, four phases can be distinguished: (1) getting agriculture moving, (2) agriculture as a contributor to growth, (3) integrating agriculture into the macro-economy, and (4) agriculture in industrial economies. In many low income countries, the agricultural sector is still in the 'getting agriculture moving' phase. Due to policies that discriminate against agriculture in many low income countries the sector has not been able to play its potential role in economic development and hence hampering overall economic development. This awareness gives extra reason to evaluate the possibilities for agriculture. Agricultural research is one of the instruments that can be used to enhance the role of agriculture in development. Those that will eventually have to accomplish the goals that are set for agriculture are farm household members. To determine the research agenda at farm household level, agricultural research institutes have relied increasingly on farming systems analysis methods.Fanning systems analysis methods, however, have not been free from problems (Chapter 2). These problems are: fanning systems analysis is vulnerable to subjectivity, has been too qualitative, is mainly farmer oriented, has been mainly crop oriented, has suffered from institutional problems, has been confronted with time conflicts, lacks gender differentiation, and has seen no unification of methods. To overcome some of these problems a new methodology for farming systems analysis was designed. Because this new methodology is more quantitative than most farming systems analysis methods it is named quantitative farming systems analysis. The new methodology encompasses the analysis of the bio-physical and socio-economic components of farming systems. The information that is generated through these analyses is used in a farm household level optimal resource allocation procedure.The working method for the farm household level optimal resource allocation procedure includes three stages and seven steps (Chapter 3). Stage 1: model preparation including (1) goal variable definition and constraint determination, (2) system definition and time horizon determination, and (3) generation of data requirements. Stage 2: (4) construction of the FLORA model. Stage 3: model utilization including (5) computing the playing field, (6) conducting sensitivity analyses, and (7) scenario construction.The study area is situated in the limestone area, South of Malang, East Java, Indonesia (Chapter 4). This area was selected because farm households in the study area are confronted with relatively low incomes, low crop productivity, and high levels of soil loss. Both the local government and the Brawijaya University, where the project was located, share efforts to improve the welfare of farm households in the study area.Results computed during Stage 3 of the farm household level optimal resource allocation procedure are presented in Chapter 5. The manner in which the procedure was developed deviated from the original design. The reasons for this are described (Chapter 6). Also the contribution of the FLORA procedure in overcoming the problems facing farming systems analysis methods is evaluated. Due to institutional conditions the procedure to establish goals for research interactively with stakeholders had to be postponed, but this dissertation enables goals for research to be presented to stakeholders and trade-offs between goal variables to be demonstrated. This dissertation concludes that future efforts to implement QFSA to establish goals for agricultural research should focus on: clear research objectives, selective data collection, better interdisciplinarity, phasing of research activities, and simple and quick modelling procedures