Much of this paper has looked at the methodology of assessing the economics of individual tsetse and trypanosomiasis control projects. Many of the studies listed here adopt a similar approach based on comparing discounted benefits and costs for with and without project scenarios, and assessing benefits using herd models. The broad similarity of the approaches used does point to there being a general consensus as to how the problem should be tackled. However, small differences in the way the problem is handled do mean that the results from different studies are difficult to compare and difficult for non-economists to assess. For this reason, it is strongly recommended that a standardized methodology be adopted for analysing the economics of tsetse and trypanosomiasis projects, which could be set out in a small booklet.
This should incorporate the following features:
basic benefit - cost analysis methodology, including the use of an agreed discount rate;
presenting various with and without project scenarios, using the partial analysis layout to identify benefits and costs;
as part of this, an analysis of the strategies currently in use, usually by farmers and livestock keepers, to control the disease;
presentation of the full costs of disease-control techniques, incorporating a standardized list from the elements listed in Figure 3, probably omitting only research and unrelated overheads;
a dynamic cattle herd model which, as well as charting herd growth and meat and milk production, either incorporates some of the main outputs from cattle into crop farming, i.e. draught and manure, or uses another method to link growth in livestock numbers to increased output from agriculture;
a benefit - cost analysis model, which could be linked to a herd model and which also incorporates the costs saved category from the partial analysis, for example the fall in trypanocide use if tsetse challenge is reduced through control measures or habitat change;
the ability to simulate a wide range of situations and outcomes and thus deal with aspects of risk and uncertainty.
While individuals will always tend to analyse a problem using their own preferred tools, there would be great advantage in their also using a generally adopted methodology that produced comparable results. Although it would inevitably handle some situations better than others, and incorporate some biases, as long as these were acknowledged and discussed before interpreting the results, this would not invalidate its usefulness.
In order to permit some historical comparisons, since even over a few years some prices change significantly, it would be useful if the convention of noting unit prices for key components of costs and benefits was maintained. These would be for such items as trypanocides, insecticides, aircraft hire, local staff salaries, price per kg liveweight for livestock, price per litre of locally produced milk, etc.
Linked to this, but outside the scope of this paper, is the nature of the data to be collected and the methods to be used to do this (see also Chapter 4, Techniques and approaches used in quantifying benefits, page 37). Again, there would be great advantage here in standardizing. For example, the socio-economic questionnaires appended to Doran 2000 or the approaches used in Kamuanga et al. (2001a) could be adapted for more general use. Hendrickx (1999) has outlined cost-effective ways of collecting data on the disease in cattle. A short list of essential quantitative and qualitative data items needed to feed the simulation model and undertake sensitivity analyses could be drawn up. The ways in which the different types of data could be collected (PRA techniques, questionnaires, trypanosomiasis surveys) and suggestions for sampling methods could also be set out. It should be clearly linked to the filtering process for priority setting that has been developed using GIS techniques.