Chapter 2 Economic issues in the research and control of tsetse and trypanosomiasis

Economic analysis of tsetse and trypanosomiasis control

The most important criterion in terms of tsetse control programmes is effectiveness rather than cost per se. Within each individual project area there will be some techniques that will be more appropriate and effective, others less so or completely inappropriate. Cost should only be a factor in deciding between appropriate and effective techniques.

The first major economic evaluation of T&T control, undertaken in Uganda (Jahnke, 1976), indicated that ground spraying of insecticide was cheaper and more effective in controlling tsetse flies than eliminating game. Different approaches were subsequently used to evaluate the benefit-cost ratios of meat production under scenarios with and without T&T control (prophylaxis treatment and/or the use of trypanotolerant cattle). Jahnke concluded that development of a tsetse control and trypanosomiasis strategy has to be integrated with land-use development and planning in Uganda. In the early 1980s, Putt and colleagues (1980) conducted a comparative analysis of insecticide ground spraying and drug therapy in cattle for the northern Nigeria savannah. Camus (1981) undertook a detailed study in Côte d’Ivoire to measure the impacts of trypanosomiasis on the productivity of four types of cattle and determined economic losses caused by the disease using a herd model. In addition, several limited economic studies and reports are available for eastern and southern Africa (Putt, 1985; Leslie, 1987; Putt, Leslie and Willemse, 1988) and for West Africa (Touré, 1981; FAO, 1977).

Brandl (1985), Shaw (1987), Shaw, Zessin and Munstermann (1994) and Itty (1995) made major contributions with the development and application of computer models for improved analysis of the projected benefits from T&T control in Burkina Faso, Côte d’Ivoire and Mali. Brandl (1988) compared the cost-effectiveness of the SIT, aerial applications of persistent insecticide and impregnated traps. His benefit-cost analysis examined tsetse control in pastoral areas of Burkina Faso and Côte d’Ivoire where projected livestock revenues derived from milk, herd growth, slaughter offtake and live sales. The profitability of the SIT was the lowest, and in general the viability of tsetse control depended on the scale of operation, the level of challenge and project life. However, there has been a major development in the SIT technique that has made it more effective. The initial cost of the SIT is steep at the outset but the accruing, long-term benefits from eradicating tsetse flies can justify the costs, as shown by the success in Zanzibar in 1997 (IAEA, personal communication). Economic analysis of the Tsetse Control Project in Côte d’Ivoire (1983-1994) resulted in a benefit-cost ratio of 3.2 and an internal rate of return of 23 percent (Shaw, 1993).

The economics of trypanosomiasis control using trypanotolerant cattle were investigated under the auspices of the Trypanotolerant Livestock Network of the then International Livestock Centre for Africa (ILCA), in Kenya, Ethiopia, the Gambia, the Democratic Republic of the Congo, Togo and Côte d’Ivoire (Itty, 1992). Itty and Swallow (1993) have shown that tsetse control appeared appropriate in situations with higher disease risk, and that imports of trypanotolerant stock (the Democratic Republic of the Congo, Togo) were not necessarily profitable. Itty (1995) developed a dynamic herd model to simulate projections for herd growth, meat and village milk production, and economic performance, from the base values collected on herd structures and productivity.

Other studies have variably documented approaches and results of cost analysis in T&T control activities. For example, the International Laboratory for Research on Animal Disease (ILRAD) socioeconomic unit compared the structure of the costs of trypanosomiasis at a national level in three countries (ILRAD, 1993). It showed that virtually all the costs of the disease in the Gambia (US$1.30/head of cattle at risk) were attributed to production losses; in Zimbabwe the costs (US$7.60/head of cattle at risk) were almost entirely the costs of control, while in Côte d’Ivoire the costs (US$4.66/head of cattle at risk) were attributed to production losses (90 percent) and control (10 percent). Barrett (1997) provided a comparative review of the costs of four major techniques for tsetse control based on case studies in Zimbabwe and Zambia where each technique has been used on a large scale. Treating cattle with insecticides was found to be the cheapest method of T&T control. The costs of using odour-bated, insecticide-treated targets compared well with traditional ground spraying, increasingly disfavoured on environmental grounds. In a more recent review, Shaw (2001) has addressed the issue of how to integrate economic criteria into the strategic planning process for T&T control and eradication in West Africa with focus on the guidelines for prioritizing projects on the basis of their economic performance.

At the continental and international levels, evaluation of T&T control programmes is turning more and more to ascertain the benefits and economies of scale from a much wider perspective. Under the auspices of the International Livestock Research Institute (ILRI), Kristjanson and colleagues (1999) measured the costs of animal trypanosomiasis and the potential benefits of control at the continent level. The study used geographical information systems (GIS) to spatially link a biophysical herd simulation model with an economic surplus model. Results indicate that the potential benefits of improved trypanosomiasis control in terms of meat and milk productivity alone amount to US$700 million a year in Africa. Trypanosomiasis costs livestock producers and consumers an estimated US$1 340 million per year in Africa without including the indirect benefits such as manure and animal traction. The authors estimated the internal rate of return of a project for research on trypanosomiasis vaccine development at 33 percent.

All in all, the recently completed study, commissioned by the Department for International Development (DFID), on Tsetse and Trypanosome Research and Development since 1980 (Budd, 1999) has provided an evaluation of the whole “International Research Programme” from the standpoint of cost-effectiveness of the product to the individual user, community, nation or region, the breadth of applicability, and research cost-effectiveness. The study concludes that over a 20-year period the ratio of total benefit to total cost would be as high as 2.6. If trypanosomiasis were to be eradicated, 40 percent of the population in sub-Saharan Africa would benefit and 55 000 deaths per year from sleeping sickness would be avoided. The potential economic effectiveness of this international research programme compares very favourably with other Africa-wide, research-development initiatives such as those for the Larger Grain Borer and the African Cassava Mosaic (Budd, 1999).

Private, public and mixed goods and characteristics of the benefit streams generated by trypanosomiasis control

The distinction between private and public goods has important implications in the implementation procedures to ensure the sustainability of T&T control. Services or goods are private when the individual who uses or consumes them captures their full benefits (Leonard, 1993). Likewise, public goods or services exist when the benefits “spill over” to other members of the community. From the viewpoint of market intervention, concepts of “excludability” and “non-rivalry” of consumption distinguish public from private goods (Cornes and Sandler, 1987) and have been used to characterize the public-private nature of animal health inputs. An input is rival (subtractable) if its use by one person reduces its value to others; it is excludable when its owner or provider can withhold its benefits from some people without incurring any cost. Thus, exclusion applies when it is relatively easy to deny access to a service to those who have not paid for it.

No particular public-cooperative-private balance is appropriate for all animal health inputs. Umali, Feder and de Haan (1994) suggest that the most efficient method for delivering an input depends upon the way its benefits are distributed. Private firms should finance an input whose benefits accrue to a particular livestock owner, while an input whose benefits are diffused through a larger population should be delivered by public agencies. Hence, an individual livestock farmer will not be willing to pay for aerial spraying to control tsetse flies because it would require spraying not only his farm but all other adjacent farms, wildlife reserves and other habitats favourable to the fly’s survival. Such a service has to be provided by the government that can “coerce” all the beneficiaries to pay for it.

Many animal health inputs are neither purely private nor purely public. For example, some inputs generate benefits to well-defined groups of livestock owners; these are called local public goods. Other inputs generate both public and private benefits; these are called mixed public-private goods or impure public goods (Cornes and Sandler, 1987). The prospects for efficient private sector delivery of a mixed public-private good depends, everything being equal, upon how the potential beneficiaries perceive the benefits and costs, and how those perceptions compare with the benefits and costs to society (Swallow. Woudyalew and Leak, 1995).

In the ex ante process of selecting appropriate techniques for combating trypanosomiasis, it is necessary to take a three-pronged approach with respect to the level of the benefits, the difference in costs and the characteristics of the benefit stream to be generated. The most obvious differences in benefits will relate to the techniques’ effectiveness in reducing livestock production losses, while the most obvious differences in costs relate to foreseeable economies of scale, the unit costs of inputs and the amount of capital, labour, management and consumables needed to generate the particular levels of output. Less obvious are the differences related to the characteristics of the benefit streams in how they are affected in nature, level and distribution to beneficiaries. Swallow (1994) developed a conceptual framework that distinguishes control techniques with respect to multiplicity, exclusivity, externality and security of the benefit stream (Table 1). The implications for socio-economic research and public policy are briefly reviewed.

TABLE 1
Characteristics of benefit streams derived from different techniques for controlling trypanosomiasis

Source: Adapted from Swallow, 1994.

Multiplicity of benefits

The range of benefits derived from the use of traps and targets includes:

• effective reduction of the density of one or several species of flies;

• reduction in the density and nuisance of biting flies and ox-peckers.

Pour-ons generate a variety of benefits to the individual livestock keeper including:

• reduced rates of trypanosome infection (knock-down effects);

• reduction of ticks and tick-borne diseases;

• reduction in the irritation caused by biting flies and birds that feed on ectoparasites.

Eradication of tsetse flies can be obtained with the use of the sterile-male technique under certain conditions without the likelihood of tsetse re-invasion. However, it is interesting to balance its high initial cost with the advantage of species-specificity and its nonpolluting nature. Indeed, it would also be interesting to compare the cost per square kilometre or unit of reclaimed land under areawide techniques with the costs of other techniques over the longer term, say ten years.

The acquired ability to control anaemia and other diseases is a non-negligible benefit of trypanotolerant breeds in Africa (d’Ieteren et al., 1998). In some countries trypanotolerant cattle are the only breeds available for livestock development programmes (e.g. the Gambia, Guinea-Conakry, Guinea-Bissau in West Africa). There is an urgent need for methodological refinement to account for the nature and multiplicity of the benefits in assessing the benefits of T&T control.

Exclusivity of benefits

This refers to the size of the actual or potential group of intended beneficiaries and the ease with which “non-members” can be excluded from the benefits. Drug therapy and trypanotolerant cattle generate benefits that are exclusive to the owner or holder of the particular livestock and therefore are “private goods” to be financed by private firms (Umali, Feder and de Haan, 1994). Tsetse suppression that can result from the effective use of traps and targets is a “local public good” that produces benefits to anyone who keeps animals within an area of reduced trypanosomiasis risk. The exclusivity of tsetse control benefits will thus depend on the size and the exclusivity of the control area itself (Swallow, 1994).

The public nature of tsetse control means that some form of local organization or public agency is probably the most appropriate way of organizing its finance and delivery (Umali, Feder and de Haan, 1994; ILRI, 1997).

Indeed, various experiences with traps and targets during the last 15 to 20 years have proved the need for some sort of participation from the intended beneficiaries. People must at least participate passively by refraining from vandalizing or removing targets or traps placed around their villages or grazing areas. Lack of this minimum level of support has led to the failure of some tsetse control experiments and interventions (Willemse, 1991; Swallow and Woudyalew, 1994). Community participation through voluntary contributions has been attempted for financing tsetse control (Gouteux and Sinda, 1990; Swallow and Woudyalew, 1994; Echessah et al., 1997). Active involvement by farmers in tsetse-intervention schemes has also been noted for Burkina Faso (Bauer et al., 1992), Côte d’Ivoire (Kientz, 1993) and Uganda (Okoth, Kirumira and Kapaata, 1991).

Externalities

Externalities occur when the activities of one economic agent affect the activities of another agent in ways that are not taken into account by the operation of the market. Traps and targets, sterile males and the use of trypanotolerant livestock appear to generate relatively small external effects. An important positive externality of the application of pour-ons to the animals owned by one farmer - to be valued as a social benefit - is that they kill tsetse flies that would otherwise transmit trypanosomiasis to other livestock in the area. However, the use of pour-ons has been controversial because of the range of external costs that are suspected. For example, ticks may develop insecticide resistance, young cattle may not have the opportunity to acquire resistance to tick-borne disease, and residual amounts of the insecticides may appear in cow milk. In recent years, toxicological tests of certain veterinary drug residues in food have been conducted on cypermethrin and a-cypermethrin, the pyrethroids most used in topical applications to cattle (WHO/FAO, 1993).

The most obvious negative externalities are the environmental costs generated by aerial and ground insecticide-spraying techniques that were in vogue from the 1950s to the 1980s, particularly in southern and eastern Africa, with the use of persistent organochlorine insecticides such as DDT and dieldrin, although recently replaced by low residual insecticides such as endosulphan (Barrett, 1997).

The use of trypanocides in ways that cause trypanosomes to develop drug resistance is an obvious external cost (Codjia et al., 1993; Swallow, 1994; Ouédraogo, 2001). There is growing concern that future effectiveness of trypanocides will be severely curtailed by widespread drug resistance. Geerts and Holmes (PAAT, 1998a) have documented the current situation for the chemotherapy of trypanosomiasis in African livestock and proposed guidelines for delaying the development of drug resistance, including recommendations on ways of controlling drug resistance when it occurs. Theoretically, if the service involves externalities, public intervention is economically justified to reduce (for negative externalities) or raise (for positive externalities) utilization to optimal social levels (Umali, Feder and de Haan, 1994).

Security of the benefits

The certainty of the benefits will vary with each technique. Generally, farmers will have the most secure expectations about benefits when they use or adopt control techniques that generate benefits of a private nature. In this respect the benefits from drug therapy will only be limited by the accuracy of the diagnosis and the dosage, and by the possibility that the drug may be adulterated. The use of pour-ons is associated with relatively less secure benefits, as the presence of a critical mass of local users is required. There is even greater uncertainty surrounding the use of T&T control measures that generate benefits of a public nature. Theft, fire and the presence of new fly species reduce the certainty of traps and targets being effective, while the uncertainty of acquiring capital and insecticide limits the benefits of all types of insecticide spraying. In areas where farmers are familiar with trypanotolerant livestock, they constitute a more secure investment in disease control than other breeds. Where trypanotolerant cattle are being introduced farmers may lack the information on which to base the large investment needed for their purchase (Itty and Swallow, 1993).

The consensus that is emerging from several studies and new technological developments is that the bait techniques (targets and treatment of cattle with insecticides) are cost effective and compete with the long-established techniques of ground and aerial spraying of insecticides. They provide new approaches to T&T control that are flexible, more sensitive to land-use improvement and they are suitable for community participation on larger scale (Barrett, 1997).

The DFID-commissioned report (Budd, 1999) concluded that international research and development on T&T control has reached the stage where the range of available tsetse control techniques, used either alone or in combination, could be widely implemented with the confident expectation that trypanosomiasis could be virtually eradicated from much of Africa.

Socio-economic planning and appraisal of tsetse and trypanosomiasis control operations

The decision to implement a T&T control programme, as with any project, must rest on the premise that physical measurements (increased meat, milk, traction power, manure), monetary measurements (market and shadow prices, opportunity costs) and the value of non-marketed inputs and production reflect social values and costs. The flow of discounted benefits and costs is balanced and should net out in favour of the livestock farmer and the community as a whole, as shown by the values of standard quantitative indicators, i.e. internal rate of returns, benefit-cost ratio and net present value (Gittinger, 1972; Little and Mirrlees, 1974). In simpler terms, project analysts have to demonstrate that the net value of livestock production forgone in the pre-control situation (the without case) is more than compensated by the net benefit streams of the T&T control programme over its expected lifetime.

The selected economic studies reviewed above (e.g. Jahnke, 1976; Brandl, 1988; Shaw, 1993; Shaw, Zessin and Munstermann, 1994) are ex post evaluations of completed operations, providing insights into strategy and policy. The power of economic analysis, however, rests in improving not our hindsight, but our foresight (Barrett, 1997). The money spent on economic analysis would be wasted if it does not improve the ability of planners to make better decisions regarding the implementation and sustainability of T&T control programmes. In this respect, a distinction has to be made between “macroplanning”, which takes a telescopic view of the entire life of the programme to derive quantitative indexes of its profitability, and “microplanning”, which deals with the practical functioning of the programme as constrained and modelled by local institutions, social rules and norms, once it has been implemented. Ex ante analysis of T&T control programmes will be further improved when account is made for non-tangible and non-quantifiable benefits and costs that will set up and sustain the dynamic and impetus of control activities. The involvement of social scientists in socio-economic planning will help to anticipate the emergence of compatible organizational forms and minimize social costs (Goodell, 1984), while effectively translating the economic research into the world of the tsetse control practitioners.