The Lake Victoria fisheries situation in Uganda was investigated during three short missions undertaken in February, July and October-November 1987. Although these visits were brief, most of the major landing sites and fish markets around the lake shore were covered, from Majanji on the Kenyan border to Malembo, near Tanzania. Informal interviews were held with various fishermen and traders at all these places.
Additional information was gathered through discussions with fisheries administrators and other professionals involved in the fishing sector. Further background material was obtained through review of the archival record and statistical reports of the Uganda Fisheries Department (UFD) and the Uganda Freshwater Fisheries Research Organization (UFFRO).
The Uganda part of Lake Victoria covers 31,000 km2, or about 45% of the total surface area (Welcomme 1972). Though of significant size, these waters are rather lightly fished in comparison to the Tanzanian and especially the Kenyan parts of the Lake (Ssentongo and Welcomme, 1985).
The events surrounding the wars of 1978 and 1986 have had a disastrous effect on Uganda and the administrative capabilities of the state. War damages, looting, and a crippled economy have resulted in the fishery administration being, de facto, unable to operate since 1978 (cf. Bernacsek, 1986). As a result, very little quantitative information is available, and the catch data from this time is judged most unreliable by the Fisheries Department itself. It is therefore quite difficult to portray accurately recent trends in the fisheries of the Ugandan sector of the lake. The following treatment is a “best approximation” of the situation based on what information is available.
4.2.1 The Situation of the Fisheries During the 1970s
The fish biomass of Lake Victoria during the 1970s consisted mostly of cichlid fish (Haplochromis, Oreochromis), as well as a number of non-cichlid fishes like: Bagrus, Synodontis, Clarias, Protopterus and Barbus. The relative abundance of these major genera in the Ugandan waters of Lake Victoria is indicated in Table 4.1. Haplochromis represented over 80% of the standing biomass during this period. In order of importance, the other major species were: Bagrus docmac, Clarias mossambicus, Synodontis victoriae, Oreochromis esculentus and Protopterus aethiopicus. Among the introduced species -- O. niloticus, O. leucostictus, Tilapia zillii, T. rendalli and Lates niloticus -- only O. niloticus has become (and remained) relatively abundant in spite of heavy fishing. Until the late 1970s, the abundance of Lates remained very low. It contributed only about 0.15% of the biomass in the early 1970s and still less than 1% in 1981 (Okaronon, Acere and Ocenodongo 1985).
The total catch from the Ugandan waters of Lake Victoria decreased dramatically throughout the 1970s as indicated in Table 4.2. From 38,000 t caught in 1971, total catch reached a lowest level of 11,100 t in 1976 before stabilizing at about 15,500 t in the 1977–79 period. While not well documented, there are probably two major reasons for this rapid decline: (a) the overexploitation of genera like the Oreochromis following the progressive reduction of mesh-size in the predominant gillnet fishery, and (b) the economic crisis and war conditions which have affected Uganda since 1976. The second factor may have been more determinant during this period.
It is important to note that Lake Victoria, inspite of being by far the largest water body of Uganda, only contributed to about 10% of total national fish production in the late 1970s, down from about 25% in the late 1960s. This is explained by the decreasing abundance of traditional key species on Lake Victoria and by the spectacular development of fisheries based on the exotic species L. niloticus and O. niloticus in Lake Kyoga.
The catch composition by species for the Uganda part of Lake Victoria, is given in Table 4.3 for the second half of the 1970s. This table shows the major species groups as being: Bagrus, Clarias, Protopterus, Haplochromis and O. niloticus. Of importance is the decrease in the catch of O. niloticus and the limited contribution of Haplochromis to the total production. The catch of Haplochromis has indeed remained quite low in Uganda (about 1,500 t) compared to those recorded in Kenya (about 6,000 t) and Tanzania (about 25,000 t) during the same period. This follows from the very limited acceptability of Haplochromis in Uganda, where it is only consumed fresh by the poorest segment of the population in the narrow belt along the shore. They are otherwise fed occasionally to children in dry form, reportedly for medicinal purposes (against diphtheria). A certain amount of the catch was reported to have been processed into chicken feed.
The Ugandan Lake Victoria fishery during the 1970s was mostly based on gillnets with mesh size varying from 1.5 to 7 inch. Three mesh sizes were predominant: 1.5 to 2 inch for Haplochromis but also catching a significant amount of Synodontis, immature Tilapia and Oreochromis; 4 to 4.5 inch for larger size Oreochromis as well as Bagrus; and 5 to 7 inch for large size O. niloticus as well as Bagrus, Protopterus and Clarias. Longlines, beach seines, traps and weirs were also used for O. niloticus, Protopterus and Clarias. Unlike in Kenya and Tanzania, the use of scoop nets, ringnets or mosquito nets in the exploitation of the small-sized pelagic species Rastrineobola argentea has remained negligible mainly for lack of a market for this species. Table 4.4 indicates the distribution of effort by type of gear in the early 1970s.
A survey of fishing craft was undertaken in 1972, recording about 2,700 canoes fishing on the Uganda waters of Lake Victoria, 10% of which were dugout canoes. At about the same period, the number of canoes operating in Kenya and Tanzania waters was about 4,000 for each country. The number of fishermen was estimated at 8,100, operating from over 600 landing sites (Kanyike, 1972). It was also reported that gillnet fishermen were using around 60 nets for motorised canoes (6 HP), through the level of motorization remained quite low. For non-motorised canoes, the number of nets was about 40 per fishing unit.
The most recent survey (June 1977) recorded 3,002 canoes fishing on the Uganda waters of Lake Victoria, of which 8% were dugouts. However, the amount of gear used by each canoe was reported to have fallen drastically.
The situation of the fisheries of Lake Victoria overall was relatively stable throughout the late 1960s and 1970s with total catch fluctuating around 100,000 tons (Table 4.5). The slow but progressive expansion of the Haplochromis fishery is associated with a decline in the abundance of the most valuable species since the 1920s. This phenomenon has been well documented starting with Graham (1929). Among the species whose numbers have dwindled in catches over the last few decades are Labeo victorianus, O. esculentus, Shilbe mystus, Alestes jacksonii and Alestes sadleri. As discussed by Cadwalladr (1969) for the case of Labeo victoranus and by Mann (1970) for the case of the endogeneous tilapia species, the key factor in the gradual collapse of these fisheries has been a systematic overexploitation of most inshore stocks through the intensification of gillnet fishing and the progressive decrease in mesh sizes (also see Table 2.2).
For the Lake in general, the situation prevailing in the 1970s was therefore characterised by the heavy exploitation of the most valuable inshore stocks and the overwhelming presence of the Haplochromis stocks whose contribution to actual production nevertheless remained quite limited if compared with a minimum potential yield of 200,000 tons per annum, as estimated by Kudhongania and Cordone (1974). The situation prevailing in Uganda is on the other hand characterised by a significant diminution of catches and fishing effort following the various events which affected the Ugandan economy in the late 1970s. The lack of gear and the disruption of the established marketing channels were primarily responsible for this reduction in fishing activity.
4.2.2 The Rehabilitation of the Fishery Sector and the Early Development of the Nile Perch Fishery
The events surrounding the wars in 1978 and 1986 had a disastrous effect on the Ugandan economy as a whole, and the fishing sector in particular. Import restrictions and the lack of foreign currency led to shortages of gear, vehicles and spare parts, seriously disrupting fishing and marketing activities. The production of fishing nets by Uganda Fishnet Manufacturers Ltd. in Kampala nearly came to a halt due to lack of nylon yarn and poor management. The activities of the public fish marketing company TUFMAC were discontinued; numerous fishing and fish marketing cooperatives also suffered severe reductions of their activities.
The private sector, which is the dominant force in the fishing industry, managed to cope relatively well in view of the circumstances. Fishing activities appear to have fallen by about 35% during the late 1970s and early 1980s, some areas being more affected than others (TDRI, 1983). A major consequence of the crisis prevailing in Uganda during this period was a drastic reduction in external trade of fish products: from US$ 120–130 million in 1973–75 to US$ 4–5 million in 1980–83, according to FAO statistics. At the same time there was a dramatic increase in the trade of processed products as a consequence of the difficulties encountered in marketing fresh fish. According to a fish market and marketing study undertaken in 1984–85 by an Italian consulting company (ITG, 1985), the percentage of fresh marketed fish decreased from 70% in the early 1970s to about 40% in 1984–85. Sundried/salted products remained stable at about 10% while the percentage of smoked fish rose from 18–20% to about 50% of total marketed quantities.
A large element of uncertainty remains as to the magnitude of the decrease in fishing activity on Lake Victoria during this period. According to official statistics the catch might have gone down to around 15,000 tons per annum during the period 1975–1983 (MAIF, 1983). These data are now thought to have under-estimated actual catch levels (Biribonwoha, Commissioner for Fisheries, pers. comm.). FAO estimated that catch levels remained at about 30,000 tons during this period (CIFA, 1982; 1983). Other assessments estimated catch levels to have been around 29,000 tons in 1982/83 (TDRI, 1983) and around 27,000 tons in 1984/85 (ITG, 1985). The value of all these assessments is limited by the unreliability of official fisheries statistics following the disintegration of the data collection system. Furthermore, informal assessments undertaken by various development agencies during this period, while confirming a downward trend in catch levels, do not provide any reliable estimates. In this context, it may be reasonable to assume that total annual catch from the Uganda portion of Lake Victoria ranged mostly from 20,000 to 30,000 t between 1975/83; down by 25% to 50% with respect to levels observed during the late 1960s and early 1970s.
Several projects have been undertaken since the early 1980s in an attempt to rehabilitate the Ugandan fishing sector. The main projects were:
International Fund for Agriculture Development (IFAD)
This project of US$ 4.5 million focused on the rehabilitation of the fisheries of Lake Kyoga, providing essential inputs to small-scale fishermen and some logistic support to the fisheries administration. A facility for local training courses was also constructed.
World Bank (IDA Credit)
Two lines of credit were provided to cover the cost of importing nylon yarn and spare parts to enable a resumption of fishing net production by Uganda Fishnet Manufacturers Ltd. (US$ 1 million and US$ 0.7 million - 1983/84).
Euro Action Accord
This NGO provided fishing equipment and technical assistance to small-scale fishermen of Lake Albert and the West Nile Region. The project has been operational off and on since 1983 and is valued at over US$ 1 million.
People's Republic of China
China has provided for the construction of three ice plant and cold storage facilities in the country. The largest of these was completed in 1983 in Kampala, and is now producing substantial quantities of ice. Although use of the plant by the fishing industry was the original objective, this has remained very marginal so far.
European Economic Community
This project, aimed at the rehabilitation of small-scale fisheries, started in September 1984 and was originally meant to continue for two years. It came to a halt in October 1986 and is programmed to continue for a second phase in 1987/88 to complete some of the activities initially planned. With a global budget of ECU 6 million, the project is focussing on the provision of inputs to small-scale fishermen in areas not covered by the aforementioned projects. Other related activities include technical assistance, logistic support, the rehabilitation of the Fisheries Training Institute and various surveys. The second phase of the Project was supposed to start shortly after the last visit to Uganda by the present mission in October 1987.
This project constitutes the second phase of Project (a). Following signs of overexploitation of the major stock of Lake Kyoga (Nile perch), the second phase focusses on stock assessment and fisheries monitoring even though a provision is made to supply additional fishing gear and to rehabilitate some landing sites. The second phase of the project is executed by the World Bank in the context of a wider “Agricultural Development Programme” with a budget of US$ 2 million for fisheries activities over the period 1986/88.
Massive inflow of gear and other fishing equipment from assistance projects led to the recovery of the fishing sector after 1983, even if it was again affected by the 1986 war and by marketing constraints. As noted earlier, the disruption of the marketing channels led to a severe reduction in the trade of fresh fish. It also isolated the southeastern part of the country from the Western Lakes region which had in the past supplied large amounts of fish to most of the country and to Kampala in particular. For example, about 7 t of fresh fish from Lake George was delivered daily on the Kampala market in the early 1970s, before supplies were discontinued. Such disruption has led the densely populated and highly urbanised southeastern part of Uganda to rely increasingly on supplies from Lakes Kyoga, Wamala and Victoria.
The excess demand situation; the general lack of gear and the near collapse of the fisheries administration - de facto unable to enforce any fisheries regulation - led to a widespread use of illegal fishing gear and fishing practices during the late 1970s and early 1980s. Most common were the use of detrimental and previously discouraged mesh sizes, “tycoon” fishing or beating of the water to drive fish into gillnets and the active use of small mesh-size gillnets of large dimensions as surrounding nets, the latter having been used quite extensively in the shallow waters of Lake Kyoga. Combined with the massive inflow of gear in the early 1980s, this situation apparently led to the overexploitation of the major fisheries of Lake Wamala and Lake Kyoga, where the Nile perch stock was particularly affected. Fish stocks in Lake Kyoga have also become more vulnerable to exploitation with the gradual decline in the water level over recent years. As a result of all these factors, the contribution of Lake Kyoga to national production - which had been over 60% throughout the 1970s - appears to have decreased by about 30% to 40%, reducing the amount of fish available nationally by perhaps as much as 40,000 tons.
The development of the Nile perch fishery of Lake Victoria occurred in this context, characterised by an increase in the availability of gear after 1983 and by the excess demand situation resulting from the sharp decrease of the catch of Nile perch on Lake Kyoga. Catch data for this period are very scarce and not really reliable. The data provided by Okaronon, Acere and Ocenodongo (1985) for the Jinja area do, however, provide some insight on the situation prevailing on Lake Victoria in the early 1980s. The catch composition of commercial activities and trawl surveys is presented in Figure 4.1 for the period 1981/83. According to these data, the contribution of Nile perch to commercial catches increased very abruptly from 13.4% in 1981 to 75.1% in 1982 and 68.4% in 1983. On the other hand, the results of the trawl surveys show an increase of Nile perch from 0.9% of the catch in 1981 to 9% in 1983, with Haplochromis decreasing from 92.3% of the trawl catch to 80.8% over the same period. Judging from Table 4.6, the explosive increase of the Nile perch stock and the subsequent refocussing of fishing effort on this species seems to have occurred in Kenya one to two years earlier than in Uganda.
Compared with the results of experimental trawl surveys conducted in 1968/71 (Kudhongania and Cordone, 1974), the explosive increase of the Nile perch stock led to a sharp reduction of the mean catch rates. Total catch rate decreased from 797 kg/hour in 1968/71 to 360 kg/hour in 1982/83. Table 4.6 shows the evolution of these rates for the major species. It clearly indicates that, with the exception of O. variabilis, O. niloticus and L. niloticus, the abundance of the other major species had already decreased significantly by 1981; i.e., before the explosion of the Nile perch stock. The 1982 data show the dramatic increase of the Nile perch stock and its effect on Haplochromis especially. The abundance of Haplochromis appears to have decreased dramatically in 1981/82 to less than half its 1968/71 level. While a further decrease in the abundance of Haplochromis may have occurred after 1983, Ogari (1985) has shown that Nile perch has been able to rapidly switch from preying predominatly on Haplochromis to preying on other species (Caridina shrimp and R. argentea).
4.3.1 Total Catch
At all landing sites visited by the mission, fishing activity was reported to have boomed since 1982. According to field staff from the Uganda Fisheries Department, fishermen, and fish traders interviewed during the survey, total catch is now two to three times higher than in the early 1980s at most landing sites.
Most fishermen have rapidly switched from traditional target species to Lates and O. niloticus as noted by Okaronon, Acere and Ocenodongo (1985). While Nile perch was caught mostly by tilapia gear in 1981, larger mesh sizes aimed specifically at Nile perch were becoming predominant by 1983. In addition the fishing sector has attracted investors from other sectors of the economy. Although variations were noted from beach to beach, the new investments seem to have occurred especially during the 1984/85 period.
Although no official data have been published by the Uganda Fisheries Department since 1983, total catch for 1984/86 was thought to have about doubled from the late 1970s-early 1980s levels, reaching perhaps 50,000 to 60,000 to per annum. However, a thorough analysis of catch records for the first six months of 1987 provided an estimate of about 40,000 t for this period alone (Orach - Meza, UFD, pers. comm.). Assuming that there is no marked seasonal variation, the annual catch level for the Ugandan sector of the Lake may have reached 80,000 t.
During the mission a rough estimate of total landings was made on the basis of reported and estimated annual landings for all landing sites reported to be “major”. The Ugandan section of Lake Victoria is divided into five fisheries regions: Sesse Islands, from which fresh and processed fish are brought to the mainland; Jinja, for the eastern Lake; Entebbe, for the central Lake; Masaka, for the western Lake; and finally Tororo, a minor region which includes the small easternmost portion of the Lake extending to the Kenya frontier. Table 4.7 provides a breakdown of the number of “major” and “minor” landing sites comprised within each of the four mainland regions.
From observations on the number of canoes (fishing canoes and transport canoes bringing fish from the islands) and the number of pick-up vehicles operating from the “major” landing sites, it is estimated that about four tons of fish are landed daily at each of these places. For most of the sites, this was corroborated by available statistics for recent months. This represents a total of over 39,000 t being landed annually for a reported total of 27 “major” landing sites. If one further considers the quantities landed on “minor” landing sites and on the reportedly very numerous occasional landing sites as well as the fish consumed on the islands and by the fishermen themselves, total landings from the Ugandan section of the Lake may indeed have reached the level of 80,000 t per annum.
4.3.2 Effort and Catch Rates
The results of the last two canoes surveys undertaken in 1972 (2,700 canoes) and 1977 (3,000 canoes) suggest a relative stability of the fleet from the early 1970s up to the initial development of the Nile perch fishery in 1981/82. In spite of the lack of data, the likely scenario seems to have been a general decrease in the size and quality (and thus efficiency) of gear used over the period due to a lower and quite irregular supply. With canoes lasting about five years, it is quite possible that fleet size did not decrease significantly after 1977. Indeed, as it is generally the case in fisheries, lagged investment/disinvestment responses level off the impact that poor performances have on fleet size. However, as a large proportion of the canoes were already owned by non-fishermen, who are more readily able to re-invest their capital in other economic activities, it is likely that a number of boat-owners did stop fishing operations during this period of limited and erratic gear supply (1976–1980 in particular).
For the sake of comparison, the catch per canoe may have decreased from 12.5 t/year in 1972 (for an annual recorded catch of 33,900 t) to about 8.3 t/year in 1977 (for an estimated catch level of about 25,000 t). Available statistics for Tanzania show catch rates per canoe remaining fairly stable through the 1970s (means of 12.5 t/year for 1970–74, and 14.1 t/year for 1975–79), at levels quite similar to Uganda under the more normal conditions of the early 1970s. By comparison the catch rates that can be projected for Kenya during the early 1970s are much lower. With a fleet size estimated at 4,000 canoes, they amounted to about 4 t/year. This lower figure essentially reflects a lower abundance and greater rate of exploitation as it can be assumed that gear efficiency is at least the same in Kenya as in the other two countries.
Since the development of the Nile perch fishery, catch rates have increased to about 20 t/year in Tanzania (Table 3.4) and 16 t/year in Kenya (assuming a fleet size of 6,000 canoes; see Section 2.2.2). In Uganda, interviews conducted at the beach level indicate that the number of active canoes has about doubled on major landing sites since 1980/81. This is confirmed by field staff of the Fisheries Department and by the boat builders interviewed during the mission. Thus, present fleet size may have reached 4,000 canoes. For an annual catch of 80,000 t, this would indicate a catch rate per canoe of about 20 t/year.
The development of the Nile perch fishery in Uganda has also led to a significant change in gear composition. At the landing sites visited by the mission, nearly all the fishermen were using gillnets with only a few still fishing with longlines. Because of the ban on night fishing and beach seining introduced in June 1987, dagaa (R. argentea) gear and beach seines were not observed at the time of the mission. The evolution of the gear composition was reported as follows:
Beach seining has generally been on the decline since 1981, and was banned in June 1987 because of its adverse impact on juvenile Nile perch;
Dagaa nets have increased since 1981 but have not officially operated since June 1987 because of the temporary ban on night fishing introduced for security reasons;
Traps and weirs have been on the decline since 1981 following the decline in abundance of traditional target species and better opportunities in the Nile perch fishery;
Longlining has also been on the decline since 1981 for the same reasons as above although longlining for Nile perch can theoretically be quite profitable (Solomon, 1987). The difficulty of getting live bait in sufficient quantities is reported to be a major constraint. The work involved and its comparatively lesser remuneration by boat owners may be another;
Gillnetting is the predominant fishing technique and is mostly directed toward Nile perch using 7 to 10 inch (stretched) mesh size. It is also directed towards O. niloticus using 6 inch mesh size (stretched). The use of smaller mesh sizes has become insignificant.
Gillnetting involves a variety of fishing units which can be characterised as follows:
Motorised canoes: These are generally operated from the mainland although a few motorised canoes were reported to be operating from the islands. In addition to the canoes strictly used to transport fresh or processed fish from the islands to the mainland, further distinction can be made for:
the canoes strictly used for Nile perch fishing, which are equipped with small to medium size outboard engines (up to 15 HP), using a large number of nets (70 to 100) with mesh size (stretched) ranging from 8 to 12 inch;
the same unit as above using less gear with mesh size (stretched) of 5 to 7 inch used alternatively in the Nile perch and O. niloticus fisheries (on light and dark nights respectively);
the canoes used in conjunction with non-motorised units fishing around the islands, and equipped with medium-to-large outboard engines (up to 25 HP) but with less fishing gear, since they are also bringing the catch of non-motorised units (generally 2 or 3) from the islands to the mainland;
Non-motorised canoes: These canoes are mostly in the range of 6 to 7 m compared with 8 to 9 m for motorised fishing canoes. Paddling is the most common form of propulsion although small sails are sometimes used to complement it. All the canoes used in the Nile perch fishery, are planked canoes, mostly of the “Sesse” type. Most of the non-motorised canoes operating from the mainland are fishing for O. niloticus, while those operating from the islands are fishing for Nile perch and/or O. niloticus. The number of nets used by these units varies from 12 to 40.
Catch per unit of effort and its evolution are difficult to assess. In the tilapia fishery, Nyholm and Whiting (1975) report a catch rate of 0.53 fish/net or 0.45 kg/net over the period 1969/70 for gillnets of 5 inch mesh. The analysis of catch and effort figures available for Kigungu fish landings illustrates the probable evolution of catch rates since this period.
In 1974 the average fishing unit carried about 86 nets (3.75 to 6 inch stretched) with a yield estimated at about 4.0 kg/net; in 1978 the number of nets was reduced to 48 nets (2.5 to 5 inch) with a slightly higher yield per net; in 1985/86 the average number of nets per canoe was back to about 80 nets with yields of about 2 kg/net in the tilapia fishery (mostly 5 inch) and yields of about 3 kg/net in the Nile perch fishery (mostly 8 inch).
Information on catch rates gathered from the site visits yields the following picture:
For the tilapia fishery, catch rates range mostly from 1 to 2 fish/net with weight ranging from 1 to 2 kg/fish depending upon the mesh size used. The average yield may be slightly less than 2 kg/net;
For the Nile perch fishery, catch rates range mostly from 0.5 to 1 fish/net with weight ranging from 3 to 5 kg/fish depending upon the mesh size used. The average yield may be slightly above 3 kg/net. An investigation of 32 fishing units conducted by Semakula (1986) indicated yields of 3.3 kg/net for non-motorised canoes and of 2.9 kg/net for motorised ones.
If it is conservatively assumed that the average number of nets per canoe is 30 and the average catch rate is 2.5 kg/net/day, the catch rate of the canoes involved in gillnetting would amount to 18.75 t/canoe per year of 250 fishing days. This gives a conservative estimate of annual catch of 75,000 t for a fleet size of about 4,000 canoes. While such figures involve a fair amount of guesswork, they tend to confirm the results of our assessment of total catch based on observed and reported landings at major sites (see Section 4.3.1).
4.3.4 Fish Processing and Marketing
Fish processing and marketing in the Uganda Lake Victoria area is almost exclusively in the hands of private operators. The major part of total supply of fish from Lake Victoria comes from the islands in either fresh or processed form. It is carried to the mainland by transport boats, and then either distributed locally by bicycles for retail sale or taken to major consumption centres by wholesalers using pick-ups or lorries. Fish trade for traditional species as well as for Nile perch is relatively well organised. The industry readily adjusted the sudden and dramatic increase in the landings of Nile perch. A very large proportion of the fish catch - perhaps 60% to 70% - is traded fresh without any apparent constraints at marketing or demand levels. The main reasons for this rather remarkable evolution are certainly the high demand for fish in the central region of the Lake, and the fact that Nile perch is much appreciated by the local population.
184.108.40.206 Demand and major product flows
There are various reasons why demand for fisheries products is very high in the central region of Lake Victoria: the presence of major population centres around the Lake (Kampala, Entebbe, Jinja, Masaka); a purchasing power generally higher than in the rest of the country; and a traditional preference for fish, reinforced by the relative scarcity of meat.
Demand for fish in this region was far higher than the quantities which could be supplied from Lake Victoria before the Nile perch boom. As a result, large amounts of fish were being brought to the region from Lakes George, Edward (mostly tilapia), and Kyoga (tilapia and Nile perch). Lake Victoria is the only lake in Uganda from which fish production is essentially consumed locally. For other lakes, a large proportion of the catch is traded outside of the immediate region. These factors also explain the higher price (from 50 to 100% according to species) received by Lake Victoria's fishermen as compared to the price received by fishermen operating on other lakes, (TDRI, 1983).
In this context, the market forces were very favorable to the development of the Nile perch fishery after the stock increased dramatically in the early 1980s. This is especially so as many people in the central region of Lake Victoria were already used to consuming the Nile perch of Lake Kyoga, which is reported to be of a lesser quality.
The evolution of the Kampala fish markets is a chief case in point. Towards the end of the 1970s the wholesale market of Katwe (fresh fish) was supplied by about 43 pick-ups delivering fish daily from Lake Kyoga (20), Lake Wamala (15), Lake George (5) and Lake Victoria (3). Each pick-up was supplying about one ton of fish daily. In 1981 a second fresh fish market was opened in Kawempe for trading of tilapia with Katwe being reserved for the Nile perch trade. In 1986/87 there were about 12 pick-ups bringing Nile perch daily to Katwe from Lake Victoria and about nine pick-ups bringing mostly tilapia to the Kawempe market from Lake Kyoga. In addition, processed fish from both lakes are brought regularly to the wholesale markets of Kibuye and Nakivubo. The evolution of the wholesale market for fresh fish clearly shows the increasing role of supplies of Nile perch from Lake Victoria (in replacement of Nile perch from Lake Kyoga and of tilapia and other species previously brought from other lakes). It also shows a reduction of the “organised” wholesale trade in fresh fish as an increasing amount of fish no longer passes through the wholesale market. This is a direct consequence of the proximity of the fish supply.
Except for Lake Kyoga, the interaction between the supply of fish from Lake Victoria and from other lakes is now insignificant. With the drastic reduction in the abundance of Nile perch on Lake Kyoga, the interaction between the production of both lakes is more balanced. Significant amounts of fresh and processed tilapia are still brought from Lake Kyoga to Kampala, Jinja, and Busia. Some smoked Nile perch is traded south of Lake Kyoga as well. The production of Lake Victoria also involves a significant long-distance trade (over 100 km). A recent study by an Italian consulting company (ITG, 1985) estimated that about 60% of the quantities available for internal trade (i.e., production minus exports, autoconsumption and losses) was traded over long distances. Long-distance trade mostly involves the following consumption/trading centres:
Masaka: A large proportion of the fish caught in the area extending from the south of the Sesse Islands to the Tanzanian border is brought to Masaka in either fresh or processed form. From there, some is channelled to Kampala/Entebbe, with a portion of the processed fish being marketed further inland. Fish trade with Tanzania appears to be limited with the exception of significant imports of dried R. argentea which are ultimately re-exported to Sudan and Zaire.
Kampala/Entebbe: As the principal market of Uganda, Kampala/Entebbe is supplied from most of the major landing sites of Lake Victoria. Large trading centres like Jinja and Masaka are intermediary links in this trade. The major source of supply remains however the northern part of the Sesse Islands.
Jinja: Fish is brought to Jinja - Masese from the nearby islands as well as from landing sites situated eastward, between Jinja and Bugoto. From Jinja, fresh fish is transported to Kampala, and to the Tororo Region.
Tororo/Mbale: These markets are important outlets for the Nile perch in the Jinja/Majanji area. The trade from Lake Victoria involves both fresh and smoked Nile perch and some sundried R. argentea.
Busia: This border town has an important fish market. It is supplied by Majanji (fresh and smoked Nile perch, some smoked tilapia) and from Lake Kyoga (fresh and processed tilapia). Barter trade with Kenya is important, mostly involving fresh and processed tilapia, as well as processed Clarias and Protopterus. Significant amounts of tilapia and some Nile perch are also directly exported to Kenya from the islands south of Majanji. Currently no fresh tilapia is being landed in Majanji. In July 1987, special measures were being taken by the Government of Uganda to limit such trade and eventually organise official export procedures.
The ability of the fish trading industry to adjust to the drastic changes which have affected both Lake Victoria and Lake Kyoga is quite impressive. This is especially so if one considers that the working conditions are still extremely difficult (poor roads, lack of parts, etc.). Nevertheless the trade routes and markets are still adjusting themselves to the new situation. Numerous fish traders were reported to frequently change their route and investigate new markets, looking for better opportunities.
220.127.116.11 Consumer preference
Consumer preferences for traditional species, around the Uganda shore, are reflected by the relative prices presented in Table 4.8 for the early 1970s. The most striking difference concerns: (a) Bagrus, highly appreciated in the central region of Lake Victoria but far less appreciated by the populations living near the Tanzanian and Kenyan borders, and (b) Protopterus, which on the contrary is quite appreciated in the border regions but not in the central region.
Similar price data are not available for recent years. However, the mission did note a growing preference for tilapia, which is now fetching a retail price quite similar to the price of the traditionally most preferred species. Nile perch on the other hand is widely accepted all around the Uganda shore. In the Central region, tilapia fetches a price which is about 20% to 30% higher than Nile perch; this difference becomes more important near the borders, reaching a maximum of about 50% on the Kenyan border.
Considering the quantities of Nile perch which are reaching consumers, it is clear that people are not only buying fish because of its lower price but because it is widely appreciated even in areas where it was hitherto largely unknown. Actually the fact that the growth of the Nile perch fishery of Lake Kyoga had apparently not been limited by a lack of demand in the 1960s is another indication of the adaptability of people's taste to new but otherwise quite palatable fish species.
It was further noted during the mission that marketing practices certainly have a significant impact on the pricing of currently secondary species like Bagrus, Barbus, Clarias and Protopterus. A priori, and accounting for regional differences, one would have expected the current retail price of these “preferred species” to be much higher than the price of tilapia; reflecting their very limited supply. But reported and observed prices at each marketing level actually show no significant differences between tilapia and the other “preferred species”. This may be explained by: (a) the fact that O. niloticus, now the prevailing tilapia species, is considered as palatable as the traditionally “preferred species”, (b) lesser demand on the part of fresh fish wholesalers, who appear reluctant to deal with small quantities of species other than tilapia and Nile perch; and (c) the direct sale of these small quantities from transporters/wholesalers to urban retailers or consumers.
As indicated in Table 4.9, the price of smoked products gives a better reflection of species preferences.
A very important resource nowadays is R. argentea, known locally as mukene. However, its exploitation appears to be limited due to the lack of demand. Although this fishery was closed during our field visit in July 1987, the regional fisheries officers of the UFD reported very little local demand for fresh or sundried mukene. Small quantities were reported to be sold in Kampala, Jinja, and in the Tororo Region. The main source of demand appears to be the West Nile Region, Sudan and Zaire. Most of the national production of mukene was reported to be used as animal feed after local processing into fish meal. Major wholesale markets for this product are Jinja and Masaka. It was reported that 2,000 to 3,000 bags (30 kg each) are passing monthly through Malembo/Masaka (most of it from Tanzania). About 700 bags are similarly traded each month in Jinja.
Finally, something should be said about the demand for Haplochromis. Unlike Kenya and Tanzania, where Haplochromis represented over 30% of total catch in the mid-1970s, its production remained limited to about 10% of total catch in Uganda. This reflects a lesser demand both for human consumption and for animal feed. Extensive test marketing conducted in the period 1967/71 by Nyholm and Whiting (1975), showed Haplochromis as “probably the least popular of all species for direct consumption in East Africa as a whole, its general acceptability being limited to a narrow belt along the lake shore and to the Luo peoples of Nyanza Province in Kenya … [Results] from test marketing were uniformly discouraging”.
Inspite of the fact that Haplochromis can yet be caught in some quantities, only very small amounts were being landed or traded at the time of the mission visits. Haplochromis was reportedly still being caught by beach seines (before they were prohibited) and mostly traded in sundried form. Fisheries officers from the UFD reported that during the 1970s Haplochromis remained the least popular species and was mostly consumed fresh in fishing villages or near landing sites and in sundried form in the narrow belt along the Lake shore (see also Dhatemwa, 1982).
Haplochromis is still very much considered a “poor man's food” or food of last resort in Uganda. In sundried form, it is now mostly used to prepare a soup traditionally fed to children reportedly for medicinal purposes. Prices observed at landing sites or on wholesale markets represented 10% to 20% of the price of tilapia.
18.104.22.168 Prices and price margins
The evolution of prices in recent years is difficult to assess and interpret because of the lack of reliable data and the effect of inflation. It was nevertheless reported that retail fish prices around Lake Victoria have slighty decreased in real terms since 1981. Average prices relating to the Kampala fish markets are shown in Table 4.9, which shows a price margin between landing prices and those charged by wholesalers of about 50% for fresh fish from Lake Victoria compared to over 100% for fresh fish from Lake Kyoga. This mostly reflects higher transport costs in terms of distance and road conditions. The distinction between urban wholesale and retail trade is often blurred as both are practiced at different scales. The overall urban wholesale and retail mark-up is about 60%, i.e., relatively lower than mark-ups generally observed in similar African fisheries. The retail mark-up for Nile perch is higher than for tilapia as more work is involved. The overall mark-up between landing and consumer prices is in the order of 150% for Lake Victoria products and 225% for Lake Kyoga products. These mark-ups also appear to be reasonable and to reflect a rather competitive situation.
TDRI (1983) shows that mark-ups for processed fish sold on Kampala markets are generally lower due to the lesser risk involved in transporting and trading processed products. Wholesale prices are given in Table 4.9 for the first semester of 1987, i.e., before the change of currency. It confirms an important distinction made by the population of Kampala (but reportedly applicable to the whole southeastern part of Uganda) between the Nile perch and tilapia coming from Lake Kyoga and those coming from Lake Victoria. The products from Lake Kyoga are reported to be too lean, of smaller size, and to have a muddy taste as compared to those from Lake Victoria. Thus the price difference observed on both fresh and processed fish markets.
Compared to meat prices, fish is still by far the cheapest source of animal protein. The retail price of average-quality beef was about 100 U.sh/kg in Kampala at the time of the mission, compared with 45 U.sh/kg for Nile perch. In Jinja, these prices were respectively 80 U.sh/kg and 35 sh/kg. In general, the price of beef is similar or above those of other commonly consumed meats (goat, chicken, pig) all around Lake Victoria, with the best quality fish remaining cheaper than beef. This is, however, not the case for the eastern part of Uganda. In the Tororo, Mbale and Busia areas, beef is cheaper than chicken, goat or fresh Clarias but still slightly more expensive than fresh tilapia. In general, the comparative advantage of processed fish is larger still, especially in terms of fresh weight equivalent.
Landing prices on the mainland are generally 3 to 5 shillings higher than those prevailing on the islands. This basically reflects transport costs as most fishermen from the islands jointly bring their catch to the mainland. Some wholesalers, however, do buy fish directly on the islands. On the mainland, the fish is generally auctioned, either by weight or by numbers of “standard” fish. Landing prices typically decrease as the distance between the landing site and the major market centres increases, the condition of the roads being an added factor. In the case of Nile perch, this leads to the following indicative price structure:
|Regions||Landing prices (U.sh/kg; July 1987)|
For Nile perch and tilapia, prices fluctuate widely following variations in the supply of both products. However, their relative substitutability tends to reduce these fluctuations, especially as high and low landings of both species do not seem to coincide very often. Typically, when fish is sold for an average of 15 U.sh/kg, prices will mostly vary from 10 to 25 U.sh/kg, and only fall exceptionally to 5 U.sh/kg. This reflects the impact of processing which tends to reduce the downward evolution of prices when fish is plentiful. Processors generally try to buy fish when prices are below average. Together with lower marketing margins, this explains the fact that processed fish (in fresh weight equivalent) is often much cheaper than fresh fish.
22.214.171.124 Fish processing
In the early 1980s it was estimated that about 70% of total landings from Lake Victoria was sold fresh, compared to 40% for Uganda as a whole (TDRI, 1983). With the dramatic increase in landings, it is possible that a slightly larger proportion of the catch is now processed. Now that the supply of Haplochromis is very much reduced, processing mostly involves smoking. Smoking is largely conducted by women. It is especially common on the islands, not only because they are further from major markets but because of plentiful firewood resources. It is reported that deforestation is not likely to become a serious problem in the foreseeable future. Careful monitoring is nevertheless called for, because of the vast increase in quantities of fish being smoked.
Fish smoking is undertaken both because there is a demand for the final product and as a means of preservation for medium- and long-distance trade. Smoking involves most species including Nile perch, tilapia, Bagrus, Clarias, Protopterus, Barbus and Synodontis. Nile perch smoking is, to a certain extent, limited by its oil content. Above a certain size, Nile perch cannot be smoked whole or split in two without becoming a fire hazard. As the limit seems to be around 8 kg, it only affects a portion of total Nile perch landings (mostly from longlines and larger mesh size gillnets of 10 to 12 inch). Larger fish of 6 kg and above are often cut into chunks of about 400 gr before being smoked. Most of the Nile perch being processed is reported to be lightly smoked with a shelf-life of three to seven days.
At the wholesale/retail level, Nile perch is often fried in pieces of about 250 gr, both as a means of preservation and to satisfy consumer demand. This is more commonly done on the mainland using larger size fish. As these are fried in their own oil, the process is quite economical. In large cities as well as in smaller rural towns, there is a large and apparently growing market for “bite size” fried Nile perch sold throughout the day and in the evening as “snacks”.
Fat is also extracted from Nile perch for use as cooking oil. Unlike in Kenya and in some parts of Tanzania, swim bladders are not yet collected for export abroad.
Ice is not used for the storage and transport of fish, even in Kampala where it is available. In fact, even if the price of ice was not prohibitive, its use does not appear to be justified in most cases. At market level, here as elsewhere in Africa, the consumer is not ready to pay more for a fresher product. Furthermore, transport time does not require the use of ice except for a very limited number of locations. The local market for already filleted fish is insignificant, even if the consumer will generally have his fish filleted by the retailer after having bought it.
Fish processing is almost exclusively a woman's activity. Women are also very much involved in small-scale wholesaling and retailing of processed fish. Their involvement in the fresh fish trade is more limited.
4.3.5 Costs and Earnings of Representative Fishing Units
A financial appraisal of costs and earnings was conducted for gillnet units fishing for Nile perch and tilapia. The results are presented in Table 4.10 for non-motorised units (Nile perch, tilapia) and for motorised units (Nile perch). These results should be considered as indicative of the investment requirements and of expected returns to labour and capital for representative units on the mainland. “Representative units” refer to fishing units which are commonly used, and not to average units, even if average yields and fish prices are used.
126.96.36.199 Major inputs
The canoes used by fishermen are relatively standard: 8 to 9 m for motorised canoes, 6 to 7 m for non-motorised canoes. Their durability depends however on the type of wood out of which they are constructed. Accordingly they will last between 3 and 6 years. Boatbuilders and boat owners interviewed during the mission reported that in recent years most orders were for canoes built with good quality wood. These canoes are expected to last 5 to 6 years with proper maintenance.
For non-motorised fishing units, gear is by far the most important element of investment and running costs. The number of nets used was observed to vary mostly from 12 to 40 in. the Nile perch fishery and from 15 to 35 in. the tilapia fishery. Fishing units operating from the mainland usually carry more nets than those operating from the islands. Very few motorised units were reported to be used on the islands, or in the tilapia fishery for those operating from the mainland. A fair number of motorised units operating from the mainland are, however, fishing alternatively for Nile perch and tilapia. For motorised fishing units, the investment cost of the gear and the outboard engine is within the same range. In the Nile perch fishery, these units are generally using 60 to 100 nets. Various engine brands are used with engine size varying from 6 to 20 hp. Many units are either under-powered or over-powered, reflecting the relative scarcity of engines.
Gear and outboard engine costs used in Table 4.10 reflect prevailing market prices. These are now greatly influenced by the sales of the EEC project which have generally brought market prices down by increasing availability and selling at cost. (While most of the equipment was presumably sold directly to the fishermen, a significant proportion is being traded by intermediaries.) More equipment will be provided by the Project over the next two years. Afterwards, shortages and higher prices may be expected unless the fishnet factory (UFM Ltd) is able to resume work at full capacity and if regular imports of inputs are available. It is to be noted that gear can be imported tax-free as they are considered “agricultural inputs”. However, post-import sales taxes are still imposed on raw goods such as netting fibre (yarn). This can pose a problem for local manufacturing since costs are increased, and the final product may have to be priced at high and non-competitive levels.
The imported nets and those made by UFM Ltd. are generally adequate for tilapia fishing but not strong enough for use in the Nile perch fishery. Such nets (e.g., 7 inch/9 ply) have the advantage of reducing the investment cost but are more costly to use in the long run. They are reported to last about eight months in the tilapia fishery and six months in the Nile perch fishery, although they could last longer with better maintenance and more systematic repair. These aspects deserve further investigation by the UFD eventually to insure that appropriate nets are imported.
Most of the brands of outboard engines used on Lake Victoria are represented by agents in Kampala. Fuel is generally available locally or brought by the fish wholesalers/transporters. At 30 U.sh/litre (US$ 0.5), fuel is comparatively cheaper than in neighbouring countries. The major constraint seems to be the occasional lack of spare parts and the rather limited selection of engines.
188.8.131.52 Occupational and ownership patterns
Butcher and Colaris (1975) reported that in the early 1970s the fishing industry in Uganda was already quite specialised with little linkage to farming. This situation still seems quite prevalent with most of the people actively involved in fishing, processing or fish trading reporting full-time involvement in these activities except for small-scale gardening. In this context, the Nile perch boom has brought additional wealth to the fishermen's communities without greatly affecting the former occupational pattern. Indeed their choice is basically limited by the lack of alternative employment or investment in other sectors like farming or small industries.
The investment cost required for entering the Nile perch fishery is rather similar to that required in the tilapia fishery. Therefore, although no data are available on ownership patterns and their evolution, one would not expect the Nile perch boom to have had much effect upon the formally prevailing pattern. Indeed, most of the boat owners interviewed were already involved in fishing before 1981. Butcher and Colaris (1975) reported that about 55% to 60% of the boats operating on Lake Victoria at the time of their survey were owner-operated. Other patterns noted were: straight hire on a monthly basis, owner hiring of crew on a cash basis or for a share of the catch and occasional hire against payment in kind (fish).
Impressions gathered during visits around the Uganda shoreline, however, suggest a relative decrease in the number of owner-operated units (i.e., where the owner is a crew member). This is especially the case on the mainland where fishing units usually carry more nets and where the rate of motorisation is much higher. There is no doubt that high returns on investment have led many “non-fishermen” to enter the Nile perch fishing industry. Also important is the fact that boat owners who usually fish when returns are low tend to stay home and hire a crew leader when returns are high.
It is not always correct to refer to non-fishing owners as “absentee” owners, which is often meant to imply that they are exploiting the fishermen without taking an active role themselves. In most cases, owning a medium to large unit or several smaller ones involves many managerial tasks (supervision of crew, selling of the catch, insuring prompt supply of inputs, repairs and maintenance services, etc.). The management of two large fishing units can be considered a full-time job. In this context, it can be said that most fishing units are still owner-operated, especially as multiple ownerships are rather common and as many units are operated by relatives.
Co-ownership is also fairly common, particularly on the islands. Typically, a boat owner who also may own some nets teams up with other crew members who themselves own some nets. Under such an arrangement, the catch of each set of nets goes to its owner, with about 10% of the catch of the crew members going to the boat owner.
Hired fishermen are usually young and single individuals coming from the islands. For those who will eventually come to operate their own fishing units, employment as hired fishermen can be considered as apprenticeship. Numerous fishermen remain crew members all their lives, however, never acquiring the productive resources to mount their own enterprises.
184.108.40.206 Remuneration for the crew
The share systems observed during the mission are quite heterogeneous, varying both from fishery to fishery and within each fishery. In the Nile perch fishery especially, crew remuneration varies widely according to locations, catch performances, fishing season and unit size, with many ad hoc periodic readjustments. This is typical of a new fishery which is not yet stabilised.
Crew size varies from 3 to 6 according to the number of nets carried by the fishing unit. However, the actual number of fishermen going fishing at any time is generally limited to 3 to 4 for non-motorised as well as motorised units. The difference in crew size is mostly related to the amount of work involved and to the rate of use of the unit. Accordingly, crew size is increased to 4 to 6 fishermen working on a rotation basis. Indeed, the largest and best run fishing units have 2 or 3 crews working alternately.
In the gillnet fishery, the most common share system is the payment of straight salaries with ad hoc bonuses when the catch is high. Salaries vary within the crew depending on seniority and experience, and from unit to unit, depending on overall performances. This system is flexible enough to offer a definite incentive to catch more.
For non-motorised canoes, the crew may alternately be paid a percentage of net revenue which is itself calculated in various ways. Estimated daily running costs may include or exclude some elements of depreciation. When depreciation is excluded, the percentage paid varies from 15% to 25% of net revenues. Such share systems are often used on the islands but seldom on the mainland.
When salaries are used, they usually correspond to 10% to 15% of net revenues in the Nile perch fishery and to 20% to 25% in the tilapia fishery. In general the payment of salaries is more often used in the Nile perch fishery than in the tilapia fishery. When crew members are paid a percentage of net revenues, it is higher for the tilapia fishery than for the Nile perch fishery. This presumably reflects: (a) comparatively lower net revenues in the tilapia fishery (the lower cost of nets not compensating for lower gross revenues) and, (b) more intensive work as tilapia fishing grounds are more limited.
In the case of co-ownership, the catch is shared according to the contribution of each crew member as described before. It is to be noted that co-ownership is more often an occasional arrangement than a long-lasting partnership. Also and quite unexpectedly, it mostly involves smaller non-motorised fishing units. One would have excepted co-ownership to be primarily used to pool financial resources for the acquisition of larger, motorised units. Individualism or the fact that investing in a large but non-motorised unit can be quite profitable may explain this situation.
Finally, the differences observed between the tilapia and Nile perch fisheries are now becoming less and less marked as an increasing number of tilapia fishermen are exploiting both fisheries alternately. Tilapia catches are indeed greatly reduced on light nights while the Nile perch catches seem to be far less affected by the moon cycle. One important restriction is, however, that tilapia fishing grounds are limited, both in number and surface area. No particular advantage is gained by exploiting them with very large gillnet fleets. A fleet of 20 to 40 nets is usually sufficient, depending on the location and size of the fishing grounds.
220.127.116.11 Returns to capital and labour
Returns to labour may seem low in comparison with returns to capital but this difference reflects the relative scarcity of both inputs. The monthly cash income of fishermen (600 to 1000 U.sh) compares quite favorably with the reported monthly salary of farm labourers (around 250 U.sh) and with the medium-scale salary of civil servants (around 500 U.sh). In addition, the fish received by each crew member after each trip represents an additional revenue of about 400 U.sh/month. If it is further considered that fishermen are fed twice a day when fishing, their earnings can be considered to be well above the average received by semi-skilled labour. Earnings on the islands are less but living expenditures are also less than on the mainland.
Returns to capital are comparatively very high, even for smaller units. This reflects not only the risk usually associated with fishing but the need to keep ahead of a very high rate of inflation which especially affects imported inputs. The situation is reinforced by the fact that running costs are extremely high compared to investment costs.
The number of nets required to break even is indicated in Table 4.10. It is around 15 for non-motorised canoes and 40 for motorised ones. Actually, non-motorised canoes carrying around 15 nets are seldom seen on the mainland although they were reported to be quite common on the islands. These units are rarely able to carry more than 35 to 40 nets because of physical limitations (space) and working time limitations (working time being shared between paddling, handling, and setting). The maximum monthly income which the owner of such a unit may expect is therefore limited (within the range of 10,000/15,000 U.sh according to Table 4.10), even if returns to investment are high.
Considering the investment cost and the relatively high fuel cost associated with motorisation, it only becomes comparatively more profitable to fish with a motorised unit if over 60 nets are used, although in some locations close to major markets less may be required. This is why most of the motorised units using 40 nets or less are generally operated in connection with non-motorised units, both for fishing and transporting the total catch of the owner's canoes.
The comparison between the tilapia and Nile perch fisheries shows that the latter is in general more profitable - although this comparative advantage would vary with the location of the fishing units, vis-à-vis fishing grounds and markets.
Overall, the profitability of the fishery appears to be extremely high. Good returns can be achieved with 20 to 25 nets, which is about half of the number of nets which was required before the Nile perch boom. The Nile perch fishery can therefore be seen as having created ample opportunities for both small-scale and large-scale investors.
4.3.6 Development Prospects
Review of the fisheries sector situation in the Uganda sector of Lake Victoria has shown a very dynamic industry taking full advantage of the opportunities created by the Nile perch boom. Sizeable Government and donor agency assistance has been provided in terms of gear and fishing equipment but the development of the industry has also depended significantly on the initiative of the private sector, supported by a very strong demand for Nile perch and other fisheries products.
No major constraints to further development of the industry were identified, except of course for the resource base. There remains, however, some uncertainty as to the provision of gear and engines in the near future. The market study undertaken by TDRI (1983) for gear and other fishing equipment needs to be updated in view of the drastic changes having affected the fisheries of Lake Victoria and Lake Kyoga. On this basis, a strategy should be elaborated with the aim of ensuring adequate supply of fishing equipment.
The industry could benefit from the improvement of feeder roads, which for some landing sites are in quite poor condition and intermittently flooded. Numerous landing sites and fish markets could also be improved through the provision of very basic infrastructure (enclosed, roofed and cemented areas for auctioning, water supply, marketing stalls, etc.). The mission has noted significant differences in the condition of the landing sites and markets. Some are fairly well equipped and managed as a result of the initiative taken by local officials, local representatives of the UFD, or by the fishermen or fish traders themselves; others are not, either because of poor management or because the investment required is too large (e.g., access roads). These improvements would not only contribute significantly to the welfare of fisherfolk communities but reduce post-harvest losses and increase the efficiency of the industry.
There are several areas which potentially carry important development prospects and which deserve to receive greater attention from the part of the UFD. First and foremost is the possibility of developing the mukene fishery. The stock has become very important and supports a significant and rapidly increasing fishery in both Kenya and Tanzania. In Uganda, low demand for this product - at least for human consumption - is presently the main limiting factor. However, significant markets exist in neighbouring countries which could complement local demand for fish meal and human consumption (West Nile Region). This fishery and the potential for its development should be investigated, together with the possibility of improving fishing techniques based on the experience acquired by fishermen in Kenya and Tanzania. Other areas which should be investigated are the technical and economic feasibility of improving fish processing, e.g., through the introduction of the Chorkor oven for fish smoking, and of promoting longlining for Nile perch as a way to exploit the stock more rationally.
If development prospects are good and apparently not limited by any major constraints, the main issue is whether or not additional investment into the fisheries (tilapia and Nile perch fisheries in particular) is justified. It can be expected that additional capital will be invested in the fisheries of Lake Victoria as long as returns are high and as long as inputs are available at reasonable cost. Furthermore, two important development projects are in the process of being implemented. These are:
A joint venture created in mid-1987 between the Government of Uganda and a fishing company from the People's Republic of China, initially for the operation of two steel trawlers on Lake Victoria. The start-up investment cost is estimated at US$ 2.3 million. The trawlers have been ordered in China and are expected to start operating early in 1988. A progressive build-up of the fleet is planned. The fish is to be sold initially in Kampala using the Sino-Uganda ice plant as a base. The possibility of exporting part of the catch will also be investigated. Daily production of each trawler is expected to be about 2 t.
Establishment of a fish processing and marketing network. This project was agreed upon early 1987 and is financed for about US$ 14 million through a loan provided by Italy. It aims at establishing fish collecting, processing and distribution centres in the southeastern part of Uganda. As such, it covers both the southern part of Lake Kyoga and the northeastern part of Lake Victoria. The objective of the project is to improve standards of fish quality and extend fish marketing possibilities in the region by using modern processing technology (for fish smoking especially) and by using ice and refrigerated lorries. The UFD will initially co-manage this project with Italian experts, although the creation of a parastatal body is contemplated. A complementary project for the construction of a boatyard is still being negotiated. Since obtaining an adequate fish supply from the small-scale fishermen of Lake Victoria is likely to be a problem, a linkage with the trawling project noted above may be called for, although it is not yet envisaged. In general, the project is based on an interpretation of the present state of the fishing industry which differs from the mission's aforementioned conclusions on the constraints which the industry is presently facing. Furthermore, a more flexible approach to the improvement of fish processing and fish trading at the community level may prove to be more appropriate.
The issue of whether or not to encourage trawling in Lake Victoria is common to the three countries and as such will be addressed in the conclusion. It should be noted, however, that the UFD has so far dealt with the issue with all the caution it requires, being fully aware of the possible conflicts between trawling and small-scale fishing activities. As a result, and in spite of a growing demand by private individuals, no licences have been given for trawling. The forthcoming Sino-Ugandan fishing operation is in this context considered as a controlled experiment.
The Uganda Fisheries Department is also fully aware of the fact that proper management measures will be required if the Nile perch and tilapia fisheries of Lake Victoria are to be exploited rationally. Entry into these fisheries is at present unrestricted. Taxes paid at various levels (canoe and trading licences, flat rate income tax, landing and market dues) are very small and have no impact on entry, although the Government may consider increasing them to reduce entry in these fisheries when appropriate. A more immediate measure taken recently by the Government was to ban the use of beach seines. Although it was done in an abrupt way dictated by various reasons, this measure is certainly appropriate in reducing growth in fishing, especially for Nile perch.
The basic points brought out by the mission review of the Ugandan fisheries of Lake Victoria can be summarised as follows:
Statistical coverage of various aspects of the Ugandan fisheries has since 1976 become quite inadequate, giving only very limited indications of developments. However, additional information gathered from documentary sources, interviews and direct observation during the field visits provides a further basis to assess the recent evolution of the fisheries.
Within the last five years or so, Nile perch has gained a predominant role in the fisheries in terms of both tonnage and commercial value. Also gaining in importance is the tilapia fishery, now mostly based on O. niloticus. The contribution of these two fisheries has become extremely important. Total catch is now estimated to be about 60,000 to 80,000 t per year, i.e. more than the highest catch levels observed during the 1970s. R. argentea has increased very significantly in abundance but has so far remained largely underexploited due to limited demand.
Fishing is strictly conducted by small-scale operators using both motorised and non-motorised canoes. Fleet size and consequently the number of fishermen have increased significantly over the last five years. This development has to a large extent been made possible by massive imports of gear by externally funded projects, motivated by the extreme shortages which affected the fishing industry in the late 1970s. Gear supply may again soon become a problem as those projects are coming to an end.
If fish supply from Lake Victoria has risen very steeply over the last five years, prices at both ex-vessel and retail level have nevertheless remained relatively high, reflecting a very strong demand for both Nile perch and tilapia and decreasing long-distance trade from Lakes Kyoga, George and Edward. The combination of high prices and steeply rising catch rates has provided for very high returns to both capital and labour in the fishing industry. There are indications that both small-scale and larger, better-capitalised operators have benefitted from this opportunity.
Fish processing generally involves smoking and frying and has become a vigorous business, especilly on the islands. Along with small-scale marketing and distribution, it mostly remains a women's activity. The demand for fuelwood has consequently increased significantly. Wood supplies reportedly remain plentiful on the islands, but close monitoring of the situation is indicated.
Marketing and distribution within and beyond the lakeshore are well organised and appear as a thriving industry showing a remarkable capability to overcome infrastructural deficiencies (poor road conditions especially). Foreign trade is limited to neighbouring countries and mostly involves barter trading with Kenya.
The benefits of the Lates boom are very substantial and are perceived as such by fisherfolk and consumers as well. Unlike in Tanzania and Kenya, consumer acceptance of Nile perch was already well established by the time of the boom, as large quantities of Lates from Lake Kyoga have been marketed in the central lakeshore region since the early 1960s. The benefits from the fisheries appear to be widely and, as far as the mission could judge, quite equitably distributed among the various types of operators involved in the industry with large benefits accruing to consumers as well as to fisherfolk.
No major constraints to further development were identified except of course for the sustainability of the resource. However, thorough consideration should be given to the ways and means of developing the relatively untapped resources of R. argentea. Furthermore, in-depth investigations and careful monitoring are required concerning: the resource, the socio-economic performance of the industry, the supply of gear and fishing equipment, the supply of fuelwood, and the introduction of small trawlers.