13. The introduction of a system of full cost recovery for publicly provided services is essential to protect emerging private initiatives from unfair competition. Furthermore cost recovery is crucial to enable governments to reduce their financial burden, make the service sustainable and independent of outside (international) financing, and ensure that the service is efficiently used by the beneficiary. Cost recovery becomes more justified as the private benefit of the services increases.
14. Cost recovery for public sector supplied veterinary services is probably the reform most widely introduced in SSA over the past decade. This comes from a growing awareness that the producer is quite willing to pay for good services and that additional revenue must be found for these services from the beneficiaries themselves since national budgets throughout SSA are under the pressure of financial austerity. Cost recovery introduced under Bank-funded projects did not cause the demand for veterinary services to decline. Producers were quite willing to pay a beneficiary contribution.
15. On the contrary, the total availability of the services seems to improve and poorer people seem to gain greater access to the services when cost recovery was introduced. Leonard (1984) already showed that when staff began charging for their curative visits, the work output increased significantly, inequality in distribution was reduced by at least one-half, and the more fully commercialized veterinary staff graduated their charges according to the recipient's ability to pay. Similar results were also found in the Bank/IFAD/ADF funded livestock project in the Central African Republic (CAR), where the introduction of full pricing led to improved drug availability, especially among the poor, who purchased, on average, 50% more drugs per animal than the wealthier livestock producers.
16. A partial contribution is nowadays charged to beneficiaries in most SSA countries for voluntary vaccinations against less contagious diseases of little three: to the national herd (theiluric diseases, anthrax, blackleg and pasteurellosis, FMD in non-meat-exporting countries, and all poultry vaccinations). This contribution as about US$ 0.05-US$ 0.25 per vaccination, which covers the vaccine and some of the other variable costs, but not staff salaries and other overheads. The recovery of these costs would double the charge. In order to gradually arrive at full cost recovery, several Bank-funded projects have included specific cost-monitoring exercises and an annual adjustment of the fee in the program. It is critical to arrive as soon as possible at full cost recovery for those non-compulsory vaccinations.
17. Most compulsory vaccinations (Rinderpest, CBPP and FMD in the meat-exporting countries of SSA) were traditionally financed fully out of general taxes and provided free to the producer. However, the private benefits of these vaccinations seem significant enough to ask for a direct contribution from producers and now a number of countries have introduced partial cost recovery. Although incomplete, initial data on 10 countries seem to corroborate the effectiveness of this approach, as no depressing effect could be found from cost recovery: Rinderpest vaccination coverage in the five countries that had introduced a vaccination fee was 58%, against 60% in those countries the: had maintained a free vaccination policy. These findings are not too surprising as sociological research in several countries has shown that producers are quite willing to pay for good services and that even where the vaccinations are officially free the producer is (illicitly) made to pay. Although the Bank recommends some degree of cost recovery for compulsory vaccinations, it generally acknowledges that other reforms (cost recovery for voluntary vaccinations and all clinical interventions and withdrawal of "privatizable tasks") are of higher priority and should be pursued first.
18. Cattle dips to treat ticks and tick-borne diseases (especially East Coast fever in East Africa), produce predominantly private benefits. If participation is low, however, the population of ticks resistant to the acaricide may increase and pose a threat to all farmers, including those participating in the program. There is thus a public element in a dipping program. While dipping is economic and essential in intensive production systems using exotic crosses, its economic justification in extensive production systems is doubtful. A study in the CAR livestock project in 1986 indicated that an average investment of US$ 15 per head (which equals the annual revenue per animal) and a recurrent annual charge of US$ 2 per head would be necessary to recover all costs, including depreciation. This was more than the herders were willing to pay, and the program was discontinued. In the Ituri Project in Zaire, newly established dips had an ex-post economic rate of return (ERR) of 0 to 15% because of the high investment costs, and the treatment produced little change in productivity. Studies undertaken during the preparation of the Kenya Animal Health Project in 1986 indicate that an increase of Ksh 0.30 to Ksh 1.50 per animal per dipping would be necessary to make the dips self-financing, and the mortality figures would have to drop from 10 to 8 percent per year to justify this fee increase. This reduction is probably feasible in the heavily infested East Coast Fever areas, but not elsewhere.
19. Cost recovery for dipping services has been disappointing in terms of regularity and numbers dipped, but these results may be due in part to an increase in hand spraying. In Kenya, a dipping fee created a disincentive for regular and widespread dipping, which was not overcome by making the dipping compulsory (Sandford, 1983). Meanwhile, Zimbabwe authorities argue that their strictly enforced compulsory dipping program is a success because it is free. However, in specific Kenyan projects, with good services, farmers have shown clearly to be willing to pay for dipping services-especially for crossbred animals. A fixed fee per head to cover all dippings over a longer period seems to produce better revenues and herder participation than a system of payment per dipping. This was the case in the Central Province in Kenya, where revenues doubled after a lump sum for four months per animal was introduced; also, the leakage of funds diminished once dip attendants were no longer required to handle small amounts of money continuously. With several new technologies emerging (direct application of acaricides with long residual effects, vaccines) and because of disappointments in profitability and management of dips, the best strategy seems to be to use for the time being simpler means (knapsack sprayers), until these more modern control techniques become available. This is the current strategy of the World Bank, which has stopped financing the construction of new dips, although several Bank-financed projects are still involved in rehabilitating existing facilities with emphasis on cost recovery and the privatization of the management.
20. Although the public good element of artificial insemination (AI) is quite small, cost recovery for this service has not yet reached a reasonable level. This is partly because in most of SSA AI has only recently been introduced. During the first years, the cost per farmer of a daily AI run, with only a handful of farmers participating, is excessive, and some initial subsidy is unavoidable while demand builds up. However, it has proved difficult to phase out the subsidy after the initial phase. According to calculations in Kenya - the only SSA country in which the World Bank finances AI - in 1986, the real cost per insemination varied from Ksh 20 to Ksh 250, depending on the livestock density of the region, but the government charged only Ksh 4 per insemination. Farmers have obviously been willing to pay more than Ksh 4 as private veterinarians were recently charging Ksh 30 per insemination. But in many regions the real cost price through government services was still substantially higher than the farmer was willing to pay. Consequently, a profitable, self-standing AI service is not likely to materialize in the medium term, except in areas of high density and intensive production where transport costs are shared with fixed clinical runs and on some private "elite" farms that are breeding bulls to be sold for natural mating.
21. Transfer of the revenues generated by the cost recovery to a special account or revolving fund, under the responsibility of the livestock services, is frequently sought by the departments and donors alike. Through the establishment of such a Livestock Development Fund (LDF): (i) livestock services would become independent from erratic central budgeting processes and cumbersome systems of financial control; (ii) staff are better motivated to collect and users to pay the fee, because there is a direct return in the form of more means to operate and better services rendered; and (iii) livestock services would be assured of long term financial sustainability. LDFs have been included in a number of recently approved bank-funded projects. The supplementation of the LDF with the revenue of special earmarked taxes, although included in a number of earlier projects, is generally opposed by finance ministries and institutions which are macroeconomic oriented such as the International Monetary Fund. They argue that because there is no longer a direct link between services rendered and revenue generated, it is more efficient to transfer the revenues of such taxes to the central treasury where they can be allocated across all sectors.
22. A decentralization of the decision making on the use of funds generated by cost recovery, for example, to the provincial level has been argued (World Bank, 1987) because it would: (i) help improve the utilization of resources since local staff generally have a better idea of the needs of their unit than some remote planner at headquarters: (ii) encourage wider community participation in development activities: and (iii) help bring down administrative costs, inasmuch as any shift to a higher level of revenue collection increases the administrative load, and the savings could be used for other productive purposes.
23. At the same time, the efficiency gains expected from decentralization need to be balanced against the possibility of leakage. Experience has shown that recovery tends to be low (30-50%) and that revolving funds become rapidly depleted under a decentralized system of administration (i.e., at the level of the vaccination yard, dip, or veterinary post). This is clearly shown in many externally financed projects by the dismal performance of veterinary pharmacies operated by livestock posts. Therefore it seems prudent to keep administrative functions at an intermediate (i.e., provincial) level where adequate control can be maintained and ample motivation provided for local participation. As more emphasis is placed on cost recovery, project designers need to give increasing attention to the administrative management and control of cost recovery revenues.
24. LDFs now exist in six SSA countries. Most of the LDFs operate at the national level, except in Cameroon, which has created three provincial LDFs. Thus far, experience with these LDFs has been positive: revenues are in the range of US$ 100,000 - 500,000 per year, and some LDFs have become important operating tools of livestock services in the countries concerned. The accounts have by and large been certified by external audits. However, LDFs have not yet had an opportunity to operate after external financing (and external protection) has ended. Where that happens, it is likely that the political pressure to reintegrate the revenues into the eternally cash-strapped Treasury will be stronger, although the need for these funds will also be much greater.
