FISHERIES DEVELOPMENT ON LAKE KIVU (RWANDA) - CREDIT SYSTEM AND FISH MARKETING

by G. Hanek and T. Baziramwabo

MINAGRI/UNDP/FAO-RWA/87/012 Project

1. INTRODUCTION

The ongoing FAO/UNDP Fisheries Development Project on Lake Kivu is in its third and final phase. It started modestly in 1979 following what proved to be the successful introduction of a clupeid Isambaza (Limnothrissa miodon) from Lake Tanganyika to Lake Kivu in 1959 (Collart, 1960). Its main objectives were, and still are, a rational exploitation of Lake Kivu fisheries resources in order to increase the protein supply and to create employment and revenue opportunities.

Due to rather unique circumstances - no previous fishing tradition, virtually monospecific fishery and no local tradition of consuming fish - the project staff was, and to some degree still is, called on to intervene in virtually all areas of fisheries development. Thus fisheries biology, technology (fishing techniques, product development, treatment and preservation) and marketing were the main areas of intervention. Clearly these were complemented by the necessity of training local personnel in all areas of fisheries development, providing credit and establishing the basis for fisheries administration and management. It follows, therefore, that in this last project phase the emphasis is placed on consolidation and the ensurance of its proper transfer to nationals.

The case study describes the general situation of Lake Kivu fisheries, followed by project results and experiences in the areas of credit and fish marketing.

2. LAKE KIVU

2.1. Characteristics

Lake Kivu is located at 1°34' latitude South and between 28°50' and 29°23' longitude East in an altitude of 1463 m. It is situated in the west of Rwanda and in the east of Zaire, forming a natural border of about 100 km between these countries. Its total surface is 2370 km², of which 42 per cent belongs to Rwanda. There are some 150 islands on the lake with a total surface of 315 km², the largest being Ijwi island. The lake is deep, with an average depth of 240 m and a maximum depth of 490 m.

Lake Kivu's origin is volcanic, as is that of Lakes Tanganyika, Edward and Albert, which all form part of the Western Rift Valley. The littoral zone of Lake Kivu is very small, representing only 12 per cent of its total surface.

The presence of methane gas is another characteristic of the lake. It is present in large quantities below 200 m and its total reserve was estimated at 50 km³ (Degens et al., 1972).

2.2. Fish resources

The fish fauna is poor and ecologically immature. There are some 30 endemic fish species belonging to the Cyprinidae, Claridae and Cichlidae families (Beadle, 1974). For this reason an attempt was made to introduce two clupeids from Lake Tanganyika (Collart, 1960). Only Limnothrissa miodon has adapted and now represents its most important stock.

There were several acoustic estimates of the biomass of L. miodon, varying between 5 000 and 8 000 tonnes. Only recently has the latest acoustic equipment been acquired and the total biomass of L. miodon in April 1989 was estimated at 3 030 tonnes (Johannesson and Lamboeuf, 1989).

2.3. The fishery

The first attempts to exploit the stock of L. miodon were made in 1976, using small beach seines. Only with the start of the first phase of the project in 1979, however, was appropriate technology introduced, adopted from Lake Tanganyika. Initially a system of catamarans was introduced - a unit consisting of two canoes (each about 8 m long) connected by two wooden bars, one Colombus lamp (1 000 candles), a liftnet of 180 m² and five to six fishermen.

Gradually all catamaran units were converted into trimaran units. A typical trimaran unit consists of three canoes, two or more lamps, a liftnet of up to 420 m² and nine to ten fishermen. Over the last ten years the number of trimaran units has increased to 186 on both sides of the lake (Hanek et al., 1988; Lamboeuf et al., 1989).

Fishing techniques are identical with those used on Lake Tanganyika. It is strictly a night activity; fishing units leave in late afternoon and once the fishing site is selected the lamps are lit at nightfall. The net is lifted five to seven times a night. Captures vary depending on the season and on moon intensity, ranging from 40–70 kgs a night to 100–300 kgs a night per fishing unit. Fishing effort ranges from 18 to 24 trips per fishing unit per month with virtually no activity during the full moon cycle (Hanek et al., 1988).

3. FISHERY CREDIT

3.1. Introduction

As is apparent from the foregoing, this project was responsible for the introduction of fishing activities for L. miodon on Lake Kivu.

During the preparatory phase the ministry responsible for fisheries in Rwanda (MINAGRI) selected 18 interested persons and instructed them in night fishing activities on Lake Tanganyika. During the first project phase (1979–1982) the practical training started. The best of those chosen were given the opportunity and the necessary credit to purchase complete fishing units, each consisting of two canoes, one lamp and a liftnet. Gradually the number of fishing units increased, as illustrated in Table 1. The first column of the table provides details concerning the introduction of project-financed and -assisted units - those which benefited from the project's credit scheme. A considerable multiplier effect took place as a number of private liftnet fishing units on both sides of the lake entered the fisheries, as shown in the second column of the table.

3.2. Project credit scheme

A unique system has been introduced by the project, mainly due to reluctance on the part of both commercial and rural cooperative banks to provide loans to the fisheries sector in Rwanda. A simple contract was drawn up between the chief of the fishing unit and the project, committing the fishing unit to sell the entire catch to the project, which recovers the loans by deducting a certain percentage from the sale proceeds after crew shares and operating expenses have been met. The loans were free of interest and the repayment periods were flexible.

The size of loans varied as different types of canoes were tested during the first project phase. Table 2 provides details on the size of loans provided, repayment period and loan recovery.

An equivalent of only US$ 108 588 was used to fund the project's credit scheme, US$ 2 715 being the average loan per fishing unit. As apparent from Table 2, a reimbursement rate of 98.72 per cent was achieved. The death of borrowers is the reason why three fishing units have not fully reimbursed the project so far. The amount outstanding is less than US$ 1 400.

The period of reimbursement varied greatly; an average of 31.4 months was recorded, with a minimum of 13 months and a maximum of 83 months per fishing unit. As mentioned above, loan repayment was closely linked to fish marketing.

¹ FRW = Rwanda Franc. FRW 81 = US$ 1

4. FISH MARKETING DEVELOPMENT

There was no tradition of consuming fish nor fishery products in Rwanda until very recently. In fact there were a number of taboos - against the consumption of not only fish but also of eggs and both domestic and wild birds - related indirectly to cows. In Rwanda, the cow is a status symbol; there are some 600 000 cows in a country of 6 million people. Cows are given as a dowry, by the man to the woman's family, or as a special gift. It was believed until recently that the consumption of fish, eggs or birds would endanger the breeding of cows. For these reasons, the consumption of fish stood at only 0.2 kg per capita per annum some ten years ago, as only people living close to the border with Zaire or to other lakes were accustomed to eating fish.

In view of the above, and considering the fact that the project had to introduce a previously unknown fish species, it follows that considerable effort was necessary to establish a market. The project has carried out an intensive promotional programme, particularly during the last four years. It has developed a number of recipes and demonstrated how to prepare these dishes, highlighting the high nutritional value of Isambaza. Working with primary and secondary schools, nutritional centres, local restaurants and hotels and with market women, the project has gradually established a steady clientele.

In the beginning only two products were offered at the three landing, processing and sales centres operated by the projects: fresh and dried Isambaza. Later other products were developed, such as frozen, and several types of smoked, Isambaza.

Up to 1985 fish landings and sales steadily increased at the project's three regional centres. Table 3 shows the fish landings received at the centres.

From 1979 to 1984 virtually all production from the Rwanda side of the lake was marketed through the project's centres. From 1985 this was no longer the case. Firstly, a number of fishing units introduced by the project had repaid their loans and thus had no longer any obligation to land their catch at the project landing centres. Secondly and more importantly, a large number of private liftnet fishing units had entered the fisheries and virtually all of these sold, and continue to sell, their catch in increasing numbers of small markets. The total fish landings on the Rwanda side of the lake is estimated at three times the landings shown in Table 3 for 1988. The total number of those markets was 18 in July 1988 (Diquelou, 1988) and increased to 30 in October 1989 (Lamboeuf et al., 1989). Some 3000 women fish vendors regularly buy Isambaza in these markets. If one adds the number of regular project clients, one can conclude that various products of Isambaza are actually sold by some 3 400 persons in Rwanda, the vast majority being women, as shown in Table 4.

5. CONCLUSIONS

5.1. Credit

In terms of credit utilization and recovery, the project's credit scheme was a success. However, as the external technical assistance is due to terminate at the end of this project phase, in December 1991, the need for a more formal credit scheme is evident. The project has already taken a number of steps in order to prepare for a smooth transfer and continuity after December 1991. Firstly, two consultancies were executed specifying both the credit requirements for the fisheries sector (Tietze, 1988) and proposing suitable mechanisms for credit delivery and recovery (Tietze and Bru, 1989). Secondly, the responsibility for importation of the necessary fishing gear and equipment has already been transferred to a national company (Hanek, 1988). It remains, therefore, to define the form and/or status of this project after December 1991. It is obvious that the project will continue after this date, most likely functioning as a national project managed by nationals with a limited external input to allow for specified consultancies.

The funds already at the project's disposal will allow for the continuation of its credit scheme even after its official conclusion in December 1991. It is, therefore, essential that the project's recommendations (Tietze, 1988; Tietze and Bru, 1989) are accepted by all parties concerned and a more formal credit scheme established without delay.

5.2. Marketing

After ten years of efforts to develop fish marketing in Rwanda it is apparent that its population now accepts and consumes different Isambaza products. While it is quite remarkable to note that fish consumption has tripled over the last ten years (to 0.6 kg per capita per annum), it is still far below the African average of about 10 kg per capita per annum. It would seem that a further tripling of fish consumption per capita is conceivable within the next three to four years, provided that the rational exploitation and management of the fish resources of Lake Kivu and other water bodies continues.

The continuing increase of fish markets outside the project and the ever-increasing demand for Isambaza products, are clear signs that a complete privatization of Isambaza marketing in Rwanda can be accomplished well before the end of this project phase.

References

Beadle, L. 1974. The inland waters of tropical Africa. Longman, London. 297p.

Collart, A. 1960. L'introduction du S. tanganicae (Ndagala) au Lac Kivu. Bull.Agr. Congo Belge, 51(4):975–985.

Degens, E.T. et al. 1972. Lake Kivu: structure, chemistry and biology of an East African Rift Lake. Geologisch Rundschear, N°.62. 245p.

Diquelou, J. 1987. Aide-mémoire concernant les marchés paralleles. RWA/87/012/TRAM/10. 8p.

Hanek, G. 1988. Aide-mémoire concernant l'armement futur des unités de pêche au Lac Kivu. RWA/87/012/TRAM/35. 5p.

Hanek, G, et al. 1988. La pêche d'Isambaza (Limnothrissa miodon) au Lac Kivu, Gisenyi. RWA/87/012/DOC/TR/06. 118p.

Johannesson, K. et M. Lamboeuf. 1989. Estimation accoustique de l' abondance de stock d'Isambaza (Limnothrissa miodon) du lac Kivu, par méthode basée sur le comptage des échos de poissons, Gisenyi. RWA/87/012/TRAM/13. 64p.

Lamboeuf, M. et al. 1989. Rapport de voyage: recensement des unités de pêche du lac Kivu. RWA/87/012/TRAM/44. 6p.

Tietze, U. 1988. Système de crédit bancaire pour les pêcheurs du lac Kivu auprès des Banques Populaires di Rwanda. RWA/87/012/DOC/TR/05. 43p.

Tietze, U. and H. Bru. 1989. Rapport du cours de fonction en crédit bancaire pour les pécheurs du lac Kivu, Rwanda. RWA/87/012/DOC/TR/12. 44p.