Decision Making of Rural Farm Households in Namibia: Lessons Learned From Multi-Annual Programming Optimisation Models

Summary

Farming activities can be assumed to be one of several driving forces which lead to continuous deterioration processes of tree and bush savannahs in the Kavango Region of northeastern Namibia. However, these farming activities can be assumed to be closely connected with securing subsistence needs of peasant farmers. In this context, the present study aims to a) empirically identify objectives of peasant farmers, b) quantify their most prevailing and other possible on-farm and off-farm activities, c) identify optimal farming strategies under important system constraints and their impacts on environmental and socioeconomic aspects and d) identify policyinduced changes to the previously identified optimal farming strategies. Results of this study are intended to a) serve as a solution point for similar problems in related ecological or economic systems and b) be relevant for policy makers to investigate policy impacts on peasant farming strategies.
Chapter 2 addresses some major challenges of the Namibian economy. In particular, the agricultural sector and the peasant farming system of the Kavango Region are described. In this environment, agricultural activities are predominantly subsistenceoriented and embedded in a semi-commercial system of imperfect markets. The prevailing farmingsystem comprises of three major elements: a) crop production, b) livestock production and c) natural resource production. By a varying degree, all elements contribute to food security, cash income or domestic usage patterns. Hence, an important internal driving force which causes (over)usage or depletion of natural resources is the obligation to secure substantial living needs. This aspect is further discussed by giving an outline of a) environmental and socioeconomic threats as well as b) possible coping strategies.
Chapter 3 illustrates the theoretical background of the present study by discussing two different methodological approaches. Peasant farmers in the research area can be as-sumed to be both producers and consumers. This aspect determines the first considered methodological approach. Several concepts of farm household (FHH) models reflect non-separability under specific assumptions and imply certain benefits and limitations. This study uses a FHH model which is primary based on Barnum and Squire. However, it is supplemented by features from Chayanov and Low. Economic farm activities are influenced by the natural resource base and vice versa. This aspect determines the second considered methodological approach. In order to reflect economic and environmental dynam-ics, bio-economic models (BEM) are wellknown tools. Hence, important facets of BEMs and their applications to various research tasks and areas are briefly outlined. This study develops a multiannual programming optimisation model (MAPOM) which is based on theories of FHHs and BEMs. MAPOM represents a typical village with two non-separable and non-interacting gender-specific FHH categories and includes some bio-physical features. Both FHH categories are equally equipped with several farm assets but follow different objectives.
Utility is maximised for both FHH categories by consuming several goods and services which are generated by participating in different farm and off-farm activities. One major constraint which needs to be respected is food security. Biophysical dynamics are addressed by a) including rainfall in all on-farm production functions, b) updating the resource base each year and c) calculating a ‘native biomass loss’ index for each farming element.
Chapter 4 consists of two parts which describe the data base of MAPOM. Primary data is predominantly obtained by a case study on farming systems (CSFS). Outcomes of the CSFS build the basis of all relevant parameter calculation processes (e.g. static input-output relationships). Moreover, a review of region-specific publications is supplemented to validate calculation outcomes and to fill data gaps. In a second part, this chapter de-picts the methodological approach of a traditional conjoint analysis (TCA). This method is used to identify objectives of peasant farmers. A brief outline of the research design is specified after a description of the theoretical foundation of TCAs and their applications in developing countries. Results of this analysis generated weighting factors for the arguments of the objective function used in MAPOM. Since these results showed some gender specifications, MAPOM distinguishes in this point between female and male-headed FHH.
Chapter 5 presents the model framework and the mathematical model formulations. Generally, it embeds the previously determined static input-output combinations in dy-namic ecological and economic relationships, in terms of equations. These equations are specified for each farming element and off-farm activity in MAPOM. Moreover, the most important consumption patterns are outlined along with production functions. As an example, equations are specified for cattle dynamics, annual growth of native biomass and population dynamics. Linkages of the different farming system elements become apparent in exchange relationships specifically between livestock and crop production. In addition, attention is paid to the description of system-related constraints. They are imple-mented in MAPOM to mirror the prevailing conditions in the research area. A final section presents the development of the objective function.
Chapter 6 illustrates the results of four different scenarios simulated by MAPOM. In general, all different farming elements and off-farm activities are represented to a varying
degree in the optimal solution of the baseline scenario. Cattle production plays a minor role in optimal farming strategies. Results indicate that land and labour are limiting fac-tors. A trade-off between food security and native biomass conservation is apparent. To some degree, native biomass could be potentially conserved without tremendously threat-ening food security. Scenario 2 involves a modification of the cattlekeeping element in the objective function. Then, cattle production is slightly intensified at the costs of crop production activities. Simultaneously, overcompensations of nutrition requirements and potentials of native biomass conservation are reduced. Scenario 3 addresses the enforcement of a fee system for natural resources usage and considers an increased land endowment. Impacts of these changes on optimal farming strategies are rather small. However, in some cases farmers use even more land for cultivation, instead of conserving native natural resources. In a final scenario, weighting factors in the objective function are modified. Generally, average weighting factors (over both genders) are suitable to reflect optimal farming strategies of both genders to some degree. Contrarily, equal weighting factors underestimate or overestimate optimal production levels. Hence, they do not reflect actual native biomass destruction potentials. Results of all scenarios indicated that FHHs are highly attracted to labour-saving technologies or activities which demand little labour. Hence, family labour is one of the most prominent limiting factors. This factor might, however, be threatened in the future by several socioeconomic developments. Therefore, it seems to be of high relevance to promote labour-saving devices for the different farming activities of peasant farmers in the research area.