41. Regarding the issue of balance or imbalance, the available evidence suggests an imbalance exists between rangeland use and its regeneration. A general indication is revealed by the fact that regional stocking rates exceed the potential carrying capacity. In addition, and perhaps more important, farmers themselves consider lack of grazing as the most important constraint to livestock development. It was found that this view also existed before the drought period during which field-work was conducted. A historical analysis of both regions - the results are not discussed in this paper - shows that the imbalance in rangeland use dates as far back as the 1950s (Arntzen, 1989). In other words, the imbalance has been in existence over a considerable period. Finally, mixed farming may be seen as a spontaneous adaptation to the mounting pressure on rangelands. However, there are no data available which could shed light on the nature of the imbalance in terms of rangeland productivity.
42. Regarding the role of destabilising factors, strong national and international incentives have encouraged the expansion in livestock numbers, adding to the established comparative ecological advantage of livestock production in semi-arid areas. Little attention has been paid by government to incorporating environmental considerations into macroeconomic and livestock development planning. Most general and sectoral policies do not (yet) differentiate for the constraints facing specific groups or ecological zones, and therefore may not address the needs of specific areas and rangeland users. This overall context has probably hampered existing activities and policy instruments from being directed specifically towards improving of rangeland management.
43. Socio-economic stratification has proved to be an important determinant of household strategies and constraints in general and for livestock development in particular, and of adjustment opportunities in terms of expansion and intensification. For example, the ownership of household assets and the management practices pursued by farmers were found to be clearly stratum-related: ownership of transport means, hiring labour, provision of fodder and bonemeal, fencing of fields in order to prevent crop damage by cattle. Moreover, data for Kgatleng showed that the upper stratum (over 40 head of cattle) has better access to groundwater than the middle stratum. This is typical for Botswana as a whole, although data for the second study region showed more egalitarian access to groundwater and surrounding rangelands because of the existence of wells with considerably lower investments and running costs. Thus overall, small-herd owners are increasingly being marginalised, and may be forced to drop out of the livestock sector. Finally, our field-work showed that labour is an important constraint to rural activities.
44. The government's awareness of environmental matters is rapidly increasing and a number of measures have been made or suggested. These include the preparation of a comprehensive environmental policy. Regarding the livestock sector, changes in the livestock pricing system are under consideration, the slaughtering capacity has been expanded, land-use planning and enforcement is being strengthened, and recently it has been suggested to target subsidies and give due consideration to external environmental impacts of livestock development. These measures are steps in the right direction; the systematic integration of environment in macroeconomic and sectoral development planning is especially considered important.*
* Editor's note: A Government White Paper on Agricultural Development Policy approved by Parliament in February 1991, includes most of the points raised here by the author.
45. This paper suggests additional (or more detailed) areas for consideration. First, more explicit use of economic instruments (e.g., reconsideration of the present pricing and subsidy schemes, negotiable grazing licenses also issued to non-cattle holders and adjustable to the carrying capacity). Such instruments could assist in controlling rangeland pressure and increase the effectiveness of existing stock and conservation orders. Second, the role of water planning as a control instrument of rangeland use could be further enhanced (e.g., fees for water use). However, proper coordination with land use planning is required. Third, the terms of trade between rural sectors could be manipulated to promote qualitative improvements in livestock production instead of expansion in numbers. This requires an assessment of government support and external costs of individual sectors. Fourth, some important livestock development issues are: safeguarding the marginal position of small-herd owners (unless sufficient alternative income opportunities exist), resolution of the unrestricted access to communal areas by freehold and leasehold ranchers (dual grazing rights).
46. The survey established a relationship between socio-economic stratification and access to rangelands as well as a relationship between socio-economic stratification and household strategies and the nature and scale of their activities. Government policies need to take these relationships into account by targeting towards specific strata and physical conditions of regions, not only in terms of carrying capacity but also in terms of access to grass and groundwater.
Map 1: Botswana and the study regions
