Previous Page Table of Contents Next Page



The mission review of the Tanzanian sector of Lake Victoria covered localities in Mwanza and Magu Districts of Mwanza Region, and Musoma and Tarime Districts of Mara Region. There was no opportunity to visit the more remote third region, Kagera, in the southwest corner of the Lake. The landing beaches observed included six in Mwanza District, one in Magu District, and three in Musoma District. Visits were also made to the municipal markets in Mwanza, Musoma, and Tarime. Several local markets and fish processing sites were toured in Mwanza and Mara Regions as well (see map, Fig.1.1).

Additional information was derived from discussions with Regional- and District-based Fisheries and Natural Resources officials, research officers of the Tanzania Fisheries Research Institute (TAFIRI) at Nyegezi (near Mwanza Town), members of the Haplochromis Ecology Survey Team (HEST) based at TAFIRI, Fisheries Training Institute (Nyegezi) personnel, and other public and private sector individuals involved in the fisheries. Statistical records, evaluation reports, and research documents were reviewed insofar as possible in an effort to gain an overall appreciation of developments.


It is difficult to compile a very detailed and accurate portrayal of the Tanzanian Lake Victoria fisheries from available statistical records, since these are often rather flimsy and unreliable. As one observer of the situation has simply remarked, “Statistical information on the fisheries is inadequate and not much can be deduced from it” (Katunzi 1985; cf. Bernacsek 1986; Bonzon et al. 1988; CIFA 1987; FAO/Norway 1984). At best, existing reports provide only a schematic account of past circumstances and recent trends.

3.2.1 Production Rates and Species Composition

Tanzania has jurisdiction over some 49% (about 34,700 km2) of Lake Victoria (Welcomme 1972), and the country relies heavily on these waters for fish production. Taking the official figures shown in Table 3.1 as a rough approximation of harvest magnitudes for the years 1969–1981, the mean contribution of the Lake to total national production of fish and marine products amounted to about 27%; for the fresh water catches alone, the Lake's contribution to total national production amounted to some 32% over the same period.

Historically, the Lake has provided Tanzania with a rich multispecies fishery exploited mainly by artisanal operators equipped with canoes and various types of gears including gillnets, seines, traps, and longlines. Fish stocks in the past were comprised mainly of haplochromine and tilapiine cichlids. Non-cichlid fish of commercial and subsistence value included Bagrus, Clarias, Protopterus, Mormyrus, Synodontis, and the migratory Labeo, Schilbe, Alestes, and Barbus. (Goudswaard 1987; Katunzi 1985).

However, a complex of changes occurring over the past decade or so has fundamentally altered this picture. Of particular significance, here as elsewhere on the Lake, has been the dramatic incursion and rapid ascendancy of the exogenous Nile perch, L. niloticus. Although it was first noticed some years earlier, Nile perch, known locally as sangara or chengu, began to appear in significant numbers around Musoma and Mwanza in 1982. As in the Ugandan and Kenyan experiences, it quickly became a dominant new feature of the Victoria waters of Tanzania (Bwathondi 1988; Goudswaard and Ligtvoet 1987; Ligtvoet, Chande, and Mosille 1987; Ssentongo 1985).

Indications of changes in catch rates and their species composition in recent years are found in data assembled by the HEST project (Goudswaard 1987). A comparison is made of the results of a trawl survey conducted in 1969/70 (Kudhongania and Cordone 1974a) with that of a HEST survey conducted in 1984/85. Although the two surveys were not exact duplicates in terms of gear and methods (Goudswaard 1987), it is obvious that several critical developments occurred during the interval that they cover. Serious declines are registered for the Haplochromis species stocks; O. esculentus -- previously the most highly regarded and commercially important tilapia species) is shown to have disappeared almost entirely; and strong reductions are indicated for Bagrus and Clarias. On the other hand, Nile perch do not figure at all in the 1969/70 trawls, but show up strongly in those of 1984/85. Table 3.2 provides a summary of the results of the two surveys.

The trends depicted by the trawl survey comparisons partly corroborate the official production figures covering the period 1975 – 1985, as presented in Table 3.3. While these figures should be treated as broad approximations only, what clearly stands out is the sharp upsurge of Lates in the catch from its incidental occurrence in the late 1970s and early 1980s to a proportion of nearly 40% in 1985. The decline of Bagrus and Clarias noted from the HEST survey above is not apparent. Some hint of the importance of the R. argentea (dagaa) fishery over recent years can also be seen in the official figures, though it is likely that dagaa have been greatly underenumerated. Finally, it appears that the exotic tilapia O. niloticus are faring rather well. It is possible that these fish are thriving in the presence of Nile perch due to their occupation of different ecological niches (Bwathondi 1988).

The latest but still preliminary landing figures reviewed for Tanzania waters show that overall production for 1986 has surged sharply, reaching a level of some 216,000 t of which Lates comprises 57%.

3.2.2 Fishing Effort and Fisherfolk

Background information on fishermen and others involved in the trading and service sectors of the Tanzania Lake Victoria fisheries is very sparse. Most of the reports and statistical records examined during the mission provide only a superficial view of circumstances, and can be briefly summarised as follows.

Although a small fleet of modern trawlers have been working in the Tanzania waters for some years, the fisheries are still heavily dominated by artisanal operators whose activities are mostly confined to the shallow inshore areas. As elsewhere on the Lake, fishing is practiced on both a full-time and occasional basis, and is frequently combined with farming as a complementary source of household subsistence and income (cf. Butcher and Colaris 1975; FAO/Norway 1984; Gerrard 1987; Katunzi 1985; Robinson 1984; Ssentongo and Welcomme 1985).

Available statistics (Table 3.4) show a rising trend in numbers of fishermen and vessels over the last fifteen years. For the interval 1970–74, there was a mean of 3578 canoes and 14,042 fishermen operating on the Tanzanian sector of the Lake; during 1975–79, the mean fleet size was 3803 canoes and the mean number of fishermen 16,538; and for 1980–85, there was a mean of 4232 canoes and 18,202 fishermen. Preliminary figures from the 1986 Fisheries Statistical Report indicate a steep rise in the number of fishermen to some 24,200, and in the number of canoes to around 7400.

Of the three Lake Regions, Mwanza has the greatest concentration of fishing effort. In 1983, for example, some 50% of the Tanzania Victoria fishermen were operating in Mwanza Region, as against 34% in Mara and 16% in Kagera. For the same year, about 44% of the total canoes were enumerated in Mwanza, compared to about 23% in Mara and 33% in Kagera (Ssentongo 1985).

The record also suggests an overall rise in Catch Per Unit of Effort (CPUE expressed as tons/boat) over the 1970–85 period. Means of 12.5, 14.1, and 18.6 tons/boat can be calculated for the intervals of 1970–74, 1975–79, and 1980–85 respectively. Considerably higher catch tonnages in the latter years coincide with the “boom” in Nile perch, and the improved CPUE evidently reflects this fact.

Aside from fishermen themselves, the Lake directly and indirectly provides a livelihood for their household dependents as well as myriad traders, boatbuilders, gear artisans, transporters, and others who offer support services connected with the fisheries. While it seems likely that the population of fisheries-dependent people along the Tanzania lakeshore amounts to several hundred thousand, there is unfortunately no solid data to which one can refer.

The dearth of knowledge on socio-economic aspects of the Lake fisheries in Tanzania, as in the case of the other riparian states, severely handicaps policy and planning exercises and can only be corrected through a concentration of research effort. Information on the fisheries trading and support sectors is especially lacking. Field studies of a local-level, detailed sort are needed to help fill this gap. Gerrard (1987) has reported in preliminary draft form the findings of one such study recently undertaken in the fishing village of Shirati-Sota (in Tarime District, close to the Kenya border). But extensive research of a similar nature is required along other reaches of the Tanzania lakeshore.

3.2.3 Gear and Equipment

Data on overall changes in gear usage amongst fishing operators in the Tanzanian waters of the Lake is also generally weak. Recorded numbers of some gear types vary drastically from year to year and there are simple arithmetical descrepancies in the statistical summaries (see for example the records cited by Ssentongo 1985 and Bwathondi 1988).

Gillnets are the basic gear used by artisanal fishing operators (FAO/Norway 1984). A recent survey by the HEST group around Mwanza found that for a sample of 80 canoes observed, 84% used gillnets, while beach seines and longlines were each used in less than 10% of the cases (Ligtvoet, Chande and Mosille 1988). The survey also found that most canoes are specifically targetting Lates, and that many fishermen have recently switched over to this fishery from others (especially from the tilapia fishery).

Local fishermen for some time have faced a problem of obtaining suitable gear, and this has been viewed as a retarding influence on Nile perch fishing especially (CIFA 1987; HEST 1986). Netting materials produced by domestic manufacturing plants often have been in short supply; at the same time, their prices have tended to be high and their quality inferior. Alternative supplies of manufactured gear have not been available due to import restrictions (Robinson 1984; cf. Bonzon et al. 1988). Because of these constraints, artisanal operators have resorted to other materials to meet their needs. Many nets are now fashioned locally out of the synthetic fibre obtained from tyre manufacturing wastes, or separated strands of nylon cords, or the polyethylene twine weaving of discarded fertilizer bags. The resulting profusion of these home-made, non-standardized gears makes it difficult to use net numbers as a simple fishing-effort index (Ligtvoet, Chande and Mosille 1988).

Official figures show 189,762 nets in use during 1981, 121,674 during 1982, 164,203 during 1983, 92,524 during 1984, and 65,946 during 1985 -- an overall decline punctuated by sharp fluctuations for this five year period (Bwathondi 1988). But for 1986, the figures for Mwanza Region show that gillnets total more than 111,000 (Table 3.5) -- a count for one region that far exceeds the 1985 count for all the three regions together. Thus the record does not clearly indicate numerical trends and is suspect in its irregularities.

It appears however that there has been a broad shift towards the use of larger gillnet mesh sizes over the last several years as the relative abundance of species stocks has changed and fishing operators have increasingly targetted Nile perch. In the Mwanza area, for example, the use of very small mesh nets (25.4–47.6mm or 1–1.9 in., stretched mesh) has declined from an occurrence of about 20% of the net fleet in 1979 to almost nil in 1986; at the same time, the use of very large mesh nets (152.4mm or 6 in. and above) has increased from a nil occurrence in 1979 to about 45% in 1986 (Ligtvoet, Chande and Mosille 1988; cf. Bwathondi 1988).

There are also indications that larger meshed material is now being employed in beach seines around Mwanza in order to harvest Nile perch more effectively (Ligtvoet, Chande and Mosille 1988), and that longer seines are being used (Goudswaard and Ligtvoet 1987). Beach seines, dagaa seines, and longlines are the other principal gears besides gillnets in the Tanzania Victoria fisheries. But it is difficult to say much about trends in their use either. As in the case of gillnets, the enumeration appears to be of questionable validity. Data from the Mwanza Region, for instance, show fairly erratic fluctuations in recorded numbers of these gears from year to year (see Table 3.5).

Nor is much known about the extent to which local operators are using engines instead of the usual sails and paddles for propelling their fishing boats, and how such usage has changed over the last ten or fifteen years. Information from the Mwanza Region (Table 3.5), again as an example, is incomplete for some years, and appears to vary erratically between others.

A notable development during the 1970s was the introduction of industrial trawlers to exploit the then abundant stocks of Haplochromis species. Beginning in 1975, several trawlers operated to serve the Nyanza Fishing and Processing Company in Mwanza with hauls of haplochromines for the production of fishmeal and oil. Trawling was mostly confined to the Mwanza Gulf and inshore waters. Within a few years concern was being raised about overfishing and adverse effects on the artisanal gillnet fishery (Bon 1988). By the early 1980s it was obvious that haplochromine stocks were not withstanding the combined pressures of heavy fishing and predation by the newly insurgent Nile perch (Ssentongo 1985; Witte and Goudswaard 1985). The industrial Haplochromis fishery has now collapsed, and the Company trawlers have in recent years been targetting Nile perch, as have the boats of other public and private organisations operating in Tanzanian waters (Bwathondi 1988). Sources indicate that the trawler fleet has expanded of late and now numbers some 18 wooden and steel-hulled craft, mostly based around Mwanza (Table 3.5).

3.2.4 Prices, Marketing and Value

Although detailed information is lacking, it is reported that fish prices may have risen more rapidly in recent years in comparison with other foods (Robinson 1984). Official estimates (CIFA 1987) indicate that in 1970, the overall Tanzania ex-vessel price of fish amounted to TShs. 0.65/kg; by 1981, it had risen to TShs. 4.43/kg. This nearly seven-fold increase would seem to be only partly attributable to normal inflationary pressures, since the annual inflation rate during the 1970–81 period in Tanzania was about 12% (World Bank 1983). No data of any significance were found for national retail fish price trends in available reports.

Information on price evolution specific to the Lake Victoria fisheries of Tanzania is similarly deficient. Complete ex-vessel price records for the 1976–86 period were found for the Mara Region only. These are summarised in Table 3.6. The overall Mara ex-vessel price was TShs 1.72/kg in 1976, and TShs 15.60/kg or nine times higher in 1986. Limited data for the Mwanza Region show fairly marked increases in ex-vessel prices during the four year period of 1983–86. Overall average prices there rose from about TShs 11/kg in 1983 to over TShs 18/kg in 1986 (Table 3.7).

Breakdowns of ex-vessel prices in Mara (Table 3.6) and Mwanza (Table 3.7) indicate the following trends for main commercial species:

Information on retail (local market) prices of fresh fish by species is not consistently recorded in the regional reports examined. In Mara, there is coverage only for the years 1976 and 1977. Data on Mwanza cover 1983 only. These latter show that retail prices ran from two to three times higher than ex-vessel prices for fresh fish.

Gross statistics (i.e. not broken down by species) on consumer prices for all fresh versus all processed (dried and smoked) fish are available for Mara covering the years 1982–86. They indicate that local market prices for fresh fish are some 50% to 100% higher than beach prices, and that processed fish are generally somewhat more expensive (20% to 30%) than fresh fish.

As noted earlier, very little is known about the characteristics of fish traders in the Tanzania sector of the Lake. There are some indications that men are much more heavily involved in trading here as compared to Kenya. Gerrard reports that in the Shirati-Sota area of Mara Region, women dominate the small-scale, micro-level trading of community markets, whereas men dominate the more profitable, longer-distance trade to larger markets, being generally more mobile and better capitalised (Gerrard 1987).

Fresh fish are usually marketed within lakeshore communities and the immediate hinterland. More remote destinations are very difficult to supply given the long distances and rather poor transportation infrastructure. Frozen fish products are similarly hard to distribute to the principal inland consuming centres due to the lack of an efficient cold chain system (cf. Bonzon et al. 1988; FAO/Norway 1984; Robinson 1984). The railway corporation has recently tried to improve the situation by adding four new refrigerated wagons to service on the Mwanza - Dar es Salaam line, reportedly for the transport of Nile perch. Dar retail shop prices for whole Nile perch have lately risen to TShs 90/kg (Bernacsek 1986).

In the absence of adequate transportation and cold handling facilities, most of the fish marketed beyond the lakeshore region are processed in some form, usually as dried or smoked products. Shipped by bus, lorry, or rail, these products reach virtually every part of Tanzania, and considerable quantities are exported to neighbouring countries as well. The traffic in dried dagaa is known to be particularly significant, but steady annual measurements of the domestic and international movement of these and other fish from the Lake Victoria regions are unfortunately not available (FAO/Norway 1984; Robinson 1984).

Fisheries office records on product movement for Mara were found only for the years 1976, 1981, and 1982; and for Mwanza only for the years 1982–1985. Based on this very limited information, it seems that export traffic from these regions has grown substantially in recent times. Exports of dried and smoked fish from Mara stood at some 25.6 tons in 1976; by 1982 they had risen to some 92.5 tons. Exports from Mwanza amounted to about 1,069.5 tons in 1982, and 2,073.4 in 1985. The Mwanza figures for 1985 are presented in Table 3.8, broken down by destination, type of product, weight, and value. It can be seen that dagaa account for the substantial proportion of total exports, and that Mwanza fish products are circulated across an extremely wide portion of the country. Dar clearly serves as an important consumer centre for Victoria fish, and the large tonnages which reach places like Songea, Bukoba, and Kigoma suggest a flourishing export traffic to neighbouring countries (e.g. Mozambique, Uganda, and Zaire).

Available data probably understate the true quantities of regional exports by a considerable margin. The Mwanza reports give no enumerations for dried tilapia, for instance; nor are there regular figures on the quantities of fresh fish exported each year to points outside the region. The real value of regional export traffic in fish may therefore be much higher than is reflected in the official figures (cf. Bonzon et al. 1988); but these are quite substantial even as they stand. As shown in Table 3.8, for example, the total value of products distributed beyond Mwanza Region in 1985 amounted to over TShs 45 million.

Overall, heightened production in combination with price increases have apparently brought significant financial benefits to local fisherfolk. Data on catch values by species covering the 1976–1986 decade for Mara are reproduced in Table 3.9. They indicate an enormous rise in total annual value from a level of TShs 13.9 million in 1976 to TShs 263.8 million in 1986, even while not including enumerations for dagaa. In recent years returns from the Lates and O. niloticus fisheries have become predominant. For 1986, Nile perch contributed over 43% of the total catch value, and O. niloticus some 42%. Similar data for the past decade could not be found for Mwanza Region, but the total value of fish caught in 1986 is reported to be incredibly large -- some TShs 2047.1 million, of which Nile perch accounts for about 61%.


Impressions gathered from the brief series of field observations around the Tanzania part of the Lake yield a complex picture of circumstances. Local fishermen, processors, and traders have rapidly adjusted to recent dramatic transformations in the fisheries and turned them to advantage. A transition from the broad multispecies fishery of pre-Lates days into a more narrow one based on relatively few dominant species is evidently in progress, although it may not be so advanced as in the case of Kenya. Capture and trading activities have increasingly focussed on dagaa, O. niloticus, and, especially, Nile perch. Direct testimony from local people confirm that these fisheries have all grown vigorously of late and have produced substantial economic benefits for many.

The appearance of Nile perch has occasioned more fundamental adjustments in both the harvest and post-harvest sectors because of its peculiar qualities. O. niloticus is an exotic species, but one which has been known and exploited for many years in the same way as indigenous tilapia. Dagaa fishing in the modern style requires some special arrangements because it is a nocturnal operation conducted with lanterns and seines; but the fishery again is a well-established one. Moreover, dagaa appear to be more or less equivalent to Haplochromis in terms of processing and consumption patterns. Lates, on the other hand, very recently and almost explosively appeared on the scene as an acquatic novelty. It is clear that Nile perch has now found a large measure of acceptance by fisherfolk and consumers alike; but it is equally clear that lakeside communities face certain problems associated with the fish, as the following discussion shows.

3.3.1 Fishing Operations, Gear, and Investment

Of the seven landing sites toured in Mwanza Region, three are located within or close to Mwanza Town, at the head of Mwanza Gulf: Chakula Barafu, Shede, and Mwaloni Kirumba. Three others, Kisaka, Igombe, and Semba, also within Mwanza District, lie along the outer waters of the Lake at the southern entrance to Speke Gulf. The remaining site, Nassa Beach, is further inside Speke Gulf along the southern shore, in Magu District. In the Musoma District of Mara Region, three sites were visited. These included Nyarusuria and Mwigobero, within the precincts of Musoma Town, and Makoko, a little further to the west along the shores of Mara Bay (map, Fig. 1.1).

Not all of the landing sites serve as “home” bases for fishing canoe fleets. Chakula Barafu and Mwaloni Kirumba principally serve as receiving stations for fish caught from other points, and as beach markets for a wide range of other goods. Canoes putting into Chakula Barafu ferry catches from various fishing grounds around the entrance to Mwanza Gulf. Mwaloni Kirumba is a very large receiving station for dried and smoked fish which are shipped in sailing dhows from Ukerewe Island, Bukoba, and other points.

Lates was much in evidence at all of the fresh fish landing sites, making up by far the largest part of catches seen on the days of the visits. Conversations with boat crews and observations of gear being employed corroborated reports by other researchers that most fishermen are now aiming for Nile perch. At several beaches, particularly within Mwanza Gulf and around Musoma, locals declared that Lates currently provide the only catches of any real significance. It was noticed however that a variety of other species were represented in the landings at most sites. These included tilapiines, Bagrus, Clarias, Protopterus, and Barbus, along with some Haplochromis. Some species continue to be relatively abundant in certain areas, and remain the object of specialised fisheries. The Magu Bay area in Speke Gulf, for example, supplies Mwanza Town with fresh Bagrus, Clarias, Synodontis, Schilbe, Barbus, Labeo, Alestes, and Mormyrus. Quantities of all these fish were seen on display at the Mwanza Municipal Market during early August 1987. They were being sold along with loads of fresh and processed Nile perch, tilapia, and dagaa which reportedly came from landing sites closer to the town.

Gillnetting is the mode of fishing at nearly all of the fresh catch landing sites visited. Observation and interviews verified previous reports of net supply problems. Extremely few nets of industrial manufacture were seen. Virtually all canoe fishermen were working with locally-made nets fashioned from tyre waste materials or fertilizer bag fibres; and virtually all of those interviewed were concerned about the difficulties of obtaining suitable netting materials. “Factory” or “duka” (shop) nets were said to be scarce and expensive; furthermore, they were said to be inappropriate for Nile perch because they tend to be of lighter ply and smaller mesh size (2.5 – 4 in., stretched mesh). Handmade nets for Lates seem to be produced in mesh sizes of 6 in. or greater, judging from those examined at the beaches. But comprehensive and reliable information on mesh size, length, and costs could not be obtained through brief observation of a limited number of canoes, nor through casual discussion with fishing teams. Fishermens' estimates of their net fleet characteristics varied widely. Many operators quoted dimensions which apply to industrially manufactured nets (e.g. 90 m lengths), but local net weavers do not work according to rigid standards and actual measurements are difficult to ascertain without direct inspection on a net-by-net basis. Cost quotations differed from boat to boat and place to place. Tyre fibre net costs were said to run from as little as TShs 1000/- to as much as TShs 2500/- complete (i.e. fitted with buoys, ropes, etc.). Numbers of nets per canoe similarly varied widely. Some canoes were seen to operate with as few as ten nets; others with as many as 30. Some fishermen claimed to be using 50 or more nets in their operations, though no actual counts were possible.

Longlines are also operated in the canoe fishery, though apparently to a far lesser extent than gillnets. Unfortunately it was not possible to interview any longline fishermen in person, so information on their operational expenses, daily routines, and other special circumstances could not be gathered directly. Longline canoes only constituted some 5% to 10% of the total canoe fleet on most of the beaches visited. Along the Speke Gulf, it was reported that some beaches “specialised” in longlining and therefore contained higher concentrations of these types of gear. Target fish for longliners are Protopterus, Clarias, Bagrus, and Lates, with the latter making up most of the catches at present. Longliners were said to operate with as many as 1500 hooks divided between lines of various lengths, and to use Haplochromis as the chief bait.

Beach seines and dagaa seines are widely distributed about the Tanzania lakeshore, with concentrations being found at particular fishing grounds. Dagaa is heavily fished around Ukerewe Island and in Mara Bay, for instance, although there was no opportunity to witness any operations directly. During one evening in Musoma, local people pointed out the “floating city” of lights used by dagaa fishermen operating on the far side of Mara Bay. The huge consignments of dried dagaa seen being transhipped at Mwaloni Kirumba were said to originate mostly from Ukerewe, indicating that a most productive fishery exists in those waters. (The magnitude of the traffic in dried dagaa observed in Mwanza alone makes the official statistics on landings of this fish extremely questionable. The dagaa catch is certainly many times greater than is reported.) Beach seines are used quite extensively at such beaches as Kisaka, Semba, and Igombe, where shorelines are open and sandy. They are expensive gears to construct and dimensions vary according to how much money their owners can afford to invest. The largest are around 1000m in length, cost several hundred thousand shillings to construct, and require teams of 30 or more people to operate. One such seine was seen in operation at Semba Beach, as will be described below.

Industrial trawling operations for Nile perch have been expanding in the Tanzania sector of the Lake, as previously indicated. The misgivings which have been expressed in other contexts about the growth of an industrial trawl fishery seem justified. Trawlers operating along inshore waters pose an immediate threat to the artisanal fishery because they put additional pressure on fish stocks and also destroy gear of small-scale fishermen. However, the intensity of trawling may not be as great as official records depict. During the mission one research trawling operation was witnessed (a trip by the HEST trawler Kiboko in Mwanza Gulf) and interviews were held with the managers of two pair trawling concerns, one in Mwanza and the other in Musoma.

Nile perch fishing offers potentially high levels of return for boat and equipment owners, as indicated by the example cost-earnings projections for gillnetters shown in Tables 3.10 and 3.11. It should be noted, however, that these projections are based on certain assumptions (equipment costs, number of fishing days, catch levels, crew size and remunerations, individual rather than multiple ownership of nets and canoes, etc.) which may not universally apply. As in other parts of the Lake, fishing enterprises differ substantially in scale, profitability, and organisational detail. Brief observation confirmed that most fishing activity at the landing sites visited involves small-scale, non-mechanised units and rather low investment levels. But it was also evident that even such artisanal operations vary considerably in their features. An appreciation of “typical” units would require the perspective of more intensive field investigations. For the present it is only possible to provide a few case examples. These cover the following general enterprise types: small-scale gillnet fishing units comprised of owners-operators and crew; larger labour-intensive fishing businesses involving an owner and many employees; and mechanised trawling operations.

Case Example Tl: Gillnetting at Shede Beach

Justin      , Joseph      , and John       are gillnet fishermen based at Shede Beach in Mwanza.

  1. Joseph and John cooperate as partners and have been fishing in the area for most of their lives. Both are about 40 years in age and have farms in the immediate vicinity, on which they raise rice and maize. While farming takes up much of their time during planting and harvesting seasons, they devote most of their energies to fishing and depend upon it for their principal income. Unlike the majority of Shede Beach fishermen, these two partners did not switch over to Nile perch (sangara) when the latter became abundant a few years ago. Instead, they continue to concentrate on tilapia (sato), claiming that despite some decrease in numbers over recent years, these fish can be just as profitable as sangara. Prices for sato have risen quite steeply (to about TShs 60/kg), and they reckon they earn more from these fish than ever before. Also, they think that sangara fishing presents more problems, in that gear is often stolen and is difficult to get. Joseph and John share the ownership of one canoe, and usually operate 7 to 10 nets of 6 inch (stretched) mesh, which they fashion and repair themselves. Each net costs around TShs 1700/- when completely assembled with hauling ropes and floats. Their canoe is now some years old and requires a good deal of maintenance to stay operational. Like most of the other 20 or so canoes based at Shede, it is a craft of some 25 feet and would cost about TShs 35,000/- at today's prices. The partners fish four or five days a week, setting out with two helpers in the late afternoon and staying with their nets all night. Nets are pulled in the morning and the team returns with their catch to sell at the beach. The two helpers are paid by being allowed to take the catch every fourth day. Catches are unpredictable, but a “normal” day brings about 20 fish which can be sold for TShs 100/- each. Joseph and John estimate that their operation yields proceeds of roughly TShs 20,000/- per month, after taking costs of food provisions, workers' shares, and equipment repairs into account.

  2. Justin operates independently as a Nile perch fisherman and canoe owner, and appears to be somewhat better off than his sato fishing colleagues. Now about 35 years old, he originated from the Usimbiti area of Mara Region, where he grew up in a fishing family. After secondary school, he went to work for some years in the Railway Corporation, but left this employment in 1985 to settle in Mwanza, where he found more lucrative prospects as a sangara fisherman. He claims to have done quite well. Along with his almost new canoe, he owns about 35 Nile perch nets of 7 and 8 inch (stretched) mesh, woven from tyre fibre material, with a reported worth of about TShs 2000/- each, fully assembled. Working with two helpers, who take the catch of every fourth fishing day as payment, he sets out from Shede around 4PM each afternoon (except Sundays and holidays) to set nets and stay overnight at fishing grounds in the Mwanza Gulf entrance. Nets are pulled in the morning at about 5AM, and Justin's team returns to the beach by 7AM or so to sell their fish. Demand for sangara is very high, and these days there never seem to be enough fish to supply all potential buyers. Nile perch prices are currently running about TShs 30/kg, but fluctuate somewhat and are usually negotiated on a day-to-day basis with buyers. Justin is reluctant to fix a specific figure for his “average” income per month, pointing out that fishing success depends on the season (being particularly poor during the dry seasons), that it can be good one year but bad the next, and that at all times it is “a matter of luck”. Nonetheless, based on his remarks about usual catches, and crew provisioning and payment, it seems likely that he earns around TShs 24,000/- or more during good months. Out of these earnings, however, must come the funds required for gear replacement and maintenance. These can amount to considerable sums. The wear and tear on locally made nets necessitates constant expense, and gear is often lost due to theft, which Justin identifies as “the number one problem” for fishermen around Mwanza.

The fishermen interviewed at Shede Beach would seem to be making quite good incomes -- probably rather more than is usual around the shoreline, judging from the many other operators who were observed to be less well equipped. Be that as it may, the composition of the Shede fishing units follows the pattern most frequently encountered at other landing sites; i.e., canoe and gear owners working with one or two hired crew on a share-of-the-catch basis.

It should be emphasised that those who work as helpers and crew constitute a large mass of people who must also be counted as fishermen even though they do not own much if anything in the way of productive equipment -- canoes, nets, etc. These workers are often in relatively poor economic circumstances, though the more fortunate can still earn an income which is substantially higher than minimum unskilled wage levels (ca. TShs 1300/- per month in 1987). Justin's two helpers, for instance, theoretically earn one quarter of the total catch landed by his canoe each year, which they divide between themselves. This amounts to an average somewhere in the neighbourhood of TShs 3000/- per month for each.

A number of canoe and gear owners do not directly participate in fishing. At Nassa Beach, for example, it was found that about 25% of the canoes were owned by people who either supervised their crews from the beach or, especially in the case of those with outside employment, simply let their craft out on hire to others. Most fishing enterprises observed at the various landing sites appeared to be of modest consequence. But this description hardly applies to the operations of certain entrepreneurs who own equipment and have others work it for them. Some have seized the opportunity presented by Nile perch to reap rather impressive financial benefits.

Case Example T2: Beach Seining at Semba

Kazungu hails from a fishing family in Ukerewe. Now in his mid-fifties, he is a person of considerable means, with business interests that include trading in dagaa at Mwaloni Kirumba, a taxi service, and a shop at Pasiansi (near Mwanza Town). He also owns a farm in the Uzinza area. It is chiefly through fishing that Kazungu has established himself, however. In 1967 he started fishing around Mwanza, using a beach seine for Haplochromis (furu). By 1983 he had switched to Nile perch, as the latter became more plentiful. At first he simply carried on using his furu gear for the sangara, even though it was not very suitable for the larger fish. But in 1985 he was able to buy two heavier seines, and these now form the basis of his operations at Semba Beach. There are two other seine owners stationed at Semba, but Kazungu runs by far the largest enterprise. He claims his nets are each some 1000m long, and cost him about TShs 450,000/- apiece to buy. Another TShs 100,000/- was reportedly spent on each net's hauling ropes and buoys. Counting the two canoes that he owns and deploys when setting and retrieving the nets, Kazungu estimates that he has invested well over TShs 1,000,000/- in his seining operations. His running costs are quite considerable as well, as many helpers are required. On the day of the mission visit, one of Kazungu's seines had been set out from the beach by his canoes at 4AM. Hauling commenced around 8AM. Two teams of 12 young men apiece were stationed several hundred metres apart, each pulling in a wing of the seine. Chanting and whistling in rhythms of work, the teams pulled the long wingropes and eventually the wings of the net itself for a continuous period of six hours. As the seine was slowly retrieved, Kazungu's canoes patrolled along the outer edges, the crews watching for any fish floating loose after being entangled in the mesh. As each end of the seine came close to the beach, the teams of pullers began drawing together, so that the catch was confined to an ever decreasing circle of net and could be landed in a narrow area between the teams. More helpers appeared in the final stages of the haul. Young boys waded out into the water along each side of the seine to keep the buoyed top and weighted bottom lines from fouling and prevent any fish from floating free. By the time the codend of the seine appeared, some 45 men and boys were involved with the haul, performing various tasks. Kazungu hires most of his pullers on a casual, daily basis. Each puller is given TShs 100/- per day and a meal. Some get as much as TShs 250/- a day if they are hardworking and have been with the operation for longer periods. Those who perform smaller tasks, such as tending the lines as they clear the water or folding the seine as it is passed back from the pulling teams, are rewarded with some fish. In addition to his pulling teams, Kazungu hires the assistance of an accountant, clerk, and foreman, along with crews for his canoes and people to carry his catches to the road where they can be collected by the vehicles he hires for delivery to Mwanza. Kazungu's more senior associates are remunerated through catch shares, each taking half of the catch on designated days. Although he recites a long list of expenses for his operations, and laments the high prices of material and transport services, etc., Kazungu admits that he does very well indeed from his beach seining. On the day of the mission visit to Semba, 267 Nile perch were landed, along with a few tilapia and Barbus. The entire haul was estimated to be about 1.5 tons, but Kazungu says that he commonly lands 2 tons, and that he can get as much as 4 tons in one haul. Using his own estimates for running costs (labour, repairs, transport, catch disbursements, etc.), and assuming conservative catch levels (20 to 25 tons/month), it can be roughly calculated that Kazungu realises anywhere between TShs 75,000/- to 100,000/- a month. If these figures are realistic, and even though major reinvestments in ropes and nets must be taken into account, then sangara has made Kazungu a wealthy man.

Beach seining on the scale witnessed at Semba Beach necessitates vast amounts of capital investment by local standards. But even the smallest industrial trawling units require considerably greater outlays of funds -- on the order of several million shillings. Trawling operations are therefore the preserve of large state- or privately-run business concerns. From sources contacted during the mission it was learned that 12 (rather than the officially reported 18) trawling units are currently working around the Mwanza area, and that only 6 of these belong to private commercial interests (4 large trawlers and 2 units of small pair trawlers). The remaining trawlers are operated by state or parastatal agencies but most do not fish on a daily basis because of fuel supply problems. There are a few additional trawlers owned by parastatal companies which are out of service due to mechanical defects, and several new trawlers are now under construction at local boat yards. Should all of these boats eventually be repaired or commissioned, the Mwanza Region trawl fleet will be of considerable size, with probable serious implications for the artisanal fishery. In the Mara Region, one large trawler recently started operations from Makoko. It is owned by a Catholic seminary but works well below its capabilities because of difficulties in marketing. The Mara Fisheries Development Centre at Musoma is also involved with trawling in a small way. The Centre operates a pair trawling unit with two boats imported from Denmark. These boats are underpowered (10HP engines) and cannot operate in heavy winds or deep waters. Trawling is therefore restricted to shallow waters within Mara Bay. Catch levels are low and the Centre does well just to meet the costs of its operations.

Case Example T3: Trawling Operations in Mwanza Gulf

The       Co. of Mwanza started pair trawling operations in mid-1987 as part of a much larger Nile perch processing project (about which more will be said later) undertaken on a joint venture basis with a Norwegian-based corporation. Financing is provided through the Norwegian corporation and credits from the Tanzania Investment Bank (TIB 1986). The Company has a temporary base close to the Fisheries Training Institute - Nyegezi from which it operates two pair trawling units composed of four Danish boats each equipped with a 20HP diesel engine, and costing about TShs 1.2 million apiece, completely assembled. Each pair usually trawls for several hours per working day around the entrance to Mwanza Gulf. Operations are still in a preliminary stage and it is hoped that they will become more efficient in the near future. Original plans called for a 5 ton per day catch level from each pair, but to date the highest daily catch has amounted only to 1.2 tons per boat. The usual catch level per boat thus far has stayed in the 400 – 500 kg. range, comfortably above the current break-even point of 200 kg per boat per day. Each day's catch is generally sold to local buyers at Nyegezi, though the Company also fills special orders when placed. Nile perch are sold ex-vessel for TShs 20/kg, and by-catches of tilapia go for TShs 50/kg. Eventually the trawlers will work exclusively to supply the Company's fish processing plant, once the latter has become fully operational. The Company managers report that demand for Nile perch is excellent and that thus far no problems have been encountered in selling fish.

3.3.2 Perceptions of Fishing Operators

A common nickname used for Nile perch around the Tanzania lakeshore and in the hinterland areas is mkombozi, the Kiswahili for “saviour”. That local people should use such a designation leaves little doubt about their high regard for Lates. As in Kenya, there was initially a strong negative reaction to the fish, followed by a period of adjustment and then of positive reception. Unlike Kenya, the transition from one extreme (rejection and annoyance) to the other (acceptance and approval) seems to have occurred very rapidly -- within a span of only a few years. It also appears that the current enthusiasm for Nile perch is somewhat more pronounced and widespread amongst Tanzanian fishermen than amongst their Kenyan counterparts. Perhaps Tanzanian fishermen found that the market for the fish developed rather quickly, with consequent dramatic improvement in their catch earnings over pre-Lates days.

Fishermen along the Tanzania shore were occasionally encountering Nile perch in their catches by the early 1970s (Bon 1988). The Mara Regional Fisheries Officer reports that their appearance caused some consternation locally. Urgent messages were received from remote landing sites about strange new “monster” fish, and the fisheries staff went to considerable effort to assure people of Lates' identity and edibility. But local operators found little to admire in this creature, which tore up nets and had no particular appeal as a food item. When it became associated with the disappearance of “normal” fish like tilapia and Haplochromis, there was all the more reason to despise it. Worse than useless, Nile perch was seen as a downright menace.

There are still fishermen these days who hold to such views; but judging from interviews at the landing sites, they seem few and far between. In most instances fishermen expressed gratification with Nile perch, and gave one basic and practical reason for their position: the fish has proved to be a good source of income. It is still widely believed that sangara has been a major cause of decline in certain other fish stocks, especially amongst the haplochromines. But this no longer seems to be commonly regarded as a great misfortune, and rarely excites complaints. For one thing, fishermen point out that furu were never such a profitable item as Nile perch. For another, there is the realisation at least on the part of some operators that overfishing of certain species has hastened their decline. Finally, the persistence and even expansion of some fisheries (especially for O. niloticus and dagaa) in the face of the Lates boom has helped to dispel early notions about Nile perch as a totally destructive predator.

While generally positive testimony was collected from fishermen at the various sites toured in the Mwanza and Mara regions, it was also quite evident that the Lates fishery is fraught with problems. Some of these are common to all the artisanal fisheries, and some are more severe in certain places than in others.

Case Example T4: Nassa Fishermen

Fishermen who work out of Nassa Beach on the southern shores of Speke Gulf in Magu District mainly operate gillnets, though there are a few longline specialists as well. The number of active canoes varies according to the time of year. Peak times are said to be during the rainy seasons, from March to June and again from September or October to December, when as many as 80 are in operation. During the mission visit in August 1987, about 65 canoes were fishing. Five of these were equipped with outboard engines. These are regarded as far more effective than other units since they could work across wider areas, although they are sometimes subject to periods of reduced activity due to fuel supply shortages. Most of the fishing units at Nassa are of the owner/operator type, in which canoe and gear proprietors are directly involved in operations with their crews, who participate on a share basis. As indicated earlier, about a quarter of the owners do not go out fishing themselves, choosing instead to contract the work through various arrangements with others. The Nassa fishing grounds yield several species, including Bagrus, Clarias, and O. niloticus. But Nile perch is most abundant and is the target fish of nearly all operations. The Assistant Fisheries Officer of Magu District and the Nassa Fisheries Attendant report that catch levels of Lates have been high in recent years, and that fishing, trading, and processing activities have intensified in the area as a result. The views expressed by Nassa fishermen themselves are decidedly in favour of Nile perch. They readily blame sangara for the decline of Haplochromis and some other species stocks, but do so almost in passing, without dwelling on it very much. Some of those interviewed simply shrugged off such concerns, suggesting that “It is a problem for the furu.” Fishermen at Nassa are obviously far more troubled about the constraints hindering their ability to exploit fully the rich resource Nile perch has now become for them. Chief amongst these constraints is the lack of good netting material. One veteran fishermen, interviewed as he sat on the beach repairing his worn gear, gestured towards a nearby pile of nets and lamented (in reference to the material used to weave them), “See, we are fishing here with old fertilizer bags. These are not proper nets, but they are all we can get.” Observation of gear at Nassa confirmed that most of the nets were indeed constructed of the plastic fibre patiently unravelled from discarded drygoods sacks. The “tyre nets” now common to the Tanzanian waters were also much in evidence. These latter are said to stand up better than the “bag nets”, but neither type seems particularly appreciated by local fishermen. The preference for heavier ply nets of standard manufacture was repeatedly expressed. Good nets of whatever description are expensive and time-consuming to make, and this may be a factor contributing to the high incidence of gear theft which greatly troubles Nassa fishermen. It is alleged that net thieves are sometimes not even fishermen themselves; they prowl fishing grounds at night and look for unattended gear, steal it, and then take it to other places to sell. In order to counter this problem, Nassa fishermen often stay out with their nets overnight. (Net theft is also discouragaged to some extent by the vigilance of Sungu Sungu, the traditional informal groups of neighbourhood “police” common to the Sukuma ethnic community which occupies the region adjoining the southeastern shores of Lake Victoria.) Loss of gear is not only due to theft. Large commercial trawlers operating in Speke Gulf are alleged to destroy the nets of canoe fishermen quite often, and are strongly resented by local fishermen as a result of this. Fish prices are another source of concern at Nassa, since they are regarded as both low and unsteady. Mwanza is the principal market but is some distance away, and buyers pay much less for fish ex-vessel at Nassa than they do at landing sites closer to the Town. On the day of the mission visit, for instance, fresh Nile perch were selling for approximately TShs 13/kg at Nassa, as compared to the TShs 30/kg then current in places like Shede or Chakula Barafu. During times of peak Nile perch catches, Nassa prices can drop to very minimum levels -- as little as TShs 4–5/kg. When buyers from outside are few and the fish many, there is no choice but to sell for such amounts.

Fishermen working out of beaches immediately around Mwanza Town also complain about seasonal price fluctuations, but the problem is not so acute for them as it is elsewhere. Given the Town's high concentration of buyers and consumers, and the fact that it serves as the most important centre on the Tanzania part of the Lake for distribution of fish to inland points by road and rail, producer prices remain somewhat more stable. The experience of Nassa fishermen in this respect is therefore more representative of all other areas besides those adjacent to Mwanza Town. Even in Musoma, the second most important centre on the Tanzania lakeshore, price fluctuations can be rather serious. Fishermen in all outlying areas face price disadvantages like those in Nassa because of inadequate storage and transportation facilities.

The problems with gear supply and net theft noted for Nassa Beach are shared by fishermen at every one of the other landing sites visited. The latter problem could well be a symptom of the former in all localities. It is noteworthy that nets fashioned of tyre manufacturing wastes and nylon rope strands seem to be more common to fishing beaches closer to the major market centres of Mwanza and Musoma, where there are denser concentrations of fishermen and where ex-vessel fish prices tend to be higher. Nets of these types are more costly to weave than those constructed of plastic bag fibres, and the fishermen at more remote sites like Nassa may not so easily afford them. It may also be that the other materials are more readily available at major centres.

Destruction of gillnets by trawlers was mostly reported by artisanal operators in the fishing grounds around the Mwanza Gulf entrance and within the Speke Gulf. This is not surprising as most trawling activity takes place in these waters. It is understood that trawlers are not supposed to work in certain inshore areas, but it is not clear how such restrictions operate or whether they are enforced or can be enforced. What does seem clear is that conflicts between the canoe fishery and the industrial trawl fishery are occurring and that local fishermen harbour deeply felt grievances about the situation.

On another plane, there are potential conflicts of interests inherent in the relationships between larger-scale canoe and gear owners and the crews and workers who serve them. Not enough is known about the circumstances of hire fishermen, but in some cases at least it appears that they remain trapped in the role of cheap labourers. Their work may not allow them to achieve any long-term financial security personally, but it certainly provides for the well-being of their employers. The young beach seine pullers at Semba, for example, might be seen in this light.

3.3.3. Processing and Trading

The impact of Nile perch on the Lake Victoria fish trade in Tanzania, as in Kenya, has been both profound and ambiguous: a combination of welcome benefits and worrisome costs. But the impression was gathered during the mission that the post-harvest sector differs between the two countries in some striking ways. In Tanzania men appear to take a far more prominent role in the fish trading business. Also, greater quantities and wider varieties of fish seem to be processed in that part of the Lake. Finally, for several reasons large-scale industrial processing of Lates in Tanzania has not developed to the same extent as in Kenya. Transportation infrastructure is lacking and the Lake is very remote from the major consuming centre and (for export traffic) port installations at Dar es Salaam. Industrial processing is advancing along certain lines in the Mwanza area, however. It can thus be treated as a distinct aspect of the post-harvest fisheries sector, which is otherwise the domain of artisanal processors, local traders, and intermediary agents who operate enterprises of various sizes.

3.3.4 Local Trading Operations

At nearly all sites visited around the Tanzania lakeshore men were found to be heavily involved in the fish trade -- as processors as well as sellers. Indeed, judging from what was seen, men largely control this business. The women traders encountered tended to run smaller operations, their commitments often divided between business and household and farm duties. More thorough investigation of gender demarcation in the fish trade is needed. There were indications from some sources interviewed that when Haplochromis were abundant, women were the ones mostly responsible for their processing and trading at the local retail market level. Men traded at the retail level in other fish which offered more lucrative returns but required greater capital commitments, and also dominated the bulk wholesale markets. When Lates started to assume important commercial significance, they were the ones who became its principal business agents. In this connection it is interesting that at the major municipal markets in Mwanza and Musoma, the trade in fresh, dried, and smoked fish was mostly handled by men, with the exception of dried dagaa, Haplochromis, and juvenile Lates, which were mainly being sold by women fishmongers. A noteworthy point is that the name “furu” seems to be increasingly used locally to refer to all of these small fish which are processed by drying in bulk and have a relatively cheap unit value in the marketplace. The implication is that a certain size and price category of fish is being designated -- a category of fish which particularly suits the low capital - high turnover requirements of very small-scale women retail traders.

Dagaa has now replaced haplochromines as the main small fish stock exploited in the Tanzania waters. A remarkable wholesale traffic in dried dagaa can be witnessed at Mwaloni Kirumba, a landing station on the northern outskirts of Mwanza Town which serves as the principal entrepot on the Tanzania lakeshore for processed fish of all types.

Case Example T5: Trading at Mwaloni Kirumba

There are some 150 fish traders who do business at Mwaloni Kirumba. Mzolo       is one of these. A young man in his late twenties, Mzolo turned his hand to various types of casual work before he settled into the fish business four years ago, starting with TShs 5000/- as capital. He has been quite fortunate in his ventures since then and now runs a thriving enterprise of medium size -- nowhere near the scale of the very large operators at Kirumba Beach, but probably bigger than most. Mzolo's business specialises in bulk consignments of dried dagaa. Most of his supplies come from Ukerewe Island contacts, although he also does business with agents in Bukoba. Fish shipments reach Mwaloni Kirumba from these points aboard large sailing dhows with cargo capacities of from 10 to 15 tons. A huge volume of goods are landed at the beach, and it bustles with activity. On the day of the mission visit, dozens of trading dhows were seen drawn up along the shore, offloading cargoes of fish, firewood, charcoal, and farm produce. Each fish trader rents a small plot on the beach from the municipality (TShs 150/- per month), where shipments are bulked atop coconut trunk platforms before being sold off to other buyers or removed by transport agents. Dagaa are shipped and sold by the gunny bag, weighing around 35kg each. There was no way of counting the number of bags seen at the beach on the day of the mission visit. Stacked in long rows and sometimes piled as high as a small house, they covered an area nearly the size of a football pitch, and obviously amounted to many hundreds of tons of fish. Mzolo handles three to four hundred bags every week during dry months when supplies are high. (Supplies decline during the March-June and October-December rainy seasons because fishing falls off, sun-drying is more difficult, and people have more greens and other food available to eat.) A bag of Ukerewe dagaa fetches an ex-vessel price of about TShs 750/-, to which must be added incidental costs such as for municipal tax (TShs 40/-) and porterage (TShs 5/-); If Mzolo resells the fish locally to Mwanza buyers, he makes a profit of about TShs 100/- per bag. But far greater profits can be made if transport can be arranged to ship dagaa to inland points. In Dar, for example, one bag can sell for as much as TShs 2000/-, which means a profit of TShs 500/- or more, depending upon transport costs. Some of the larger-scale businessmen at Mwaloni Kirumba handle two or three thousand bags of dagaa a month, use their own dhows to ship fish from Ukerewe or elsewhere, and their own lorries to transport loads inland. Mzolo looks forward to the day when he too can run things on such a scale. In the meantime, to protect his own interests in the highly competitive trade at Kirumba Beach, he carefully cultivates his business contacts. Dagaa fishing crews and dhow owners from Ukerewe are offered “incentives” such as kerosene lamps, fuel, and food in order to ensure the flow of supplies. Though mostly a dagaa trader, Mzolo also deals in smoked Nile perch and to a lesser extent in dried tilapia as well as smoked Bagrus and Synodontis. Supplies of smoked Nile perch from Ukerewe are fairly constant, and Mzolo claims he sometimes handles as many as 100 loads in a month. Loads of sangara and other larger processed fish are shipped by the tenga, consisting of about 200 pieces stacked in a circular arrangement, often with a layer of grass around the outside, and bound tightly between two pieces of flat basketwork. One tenga of Ukerewe Nile perch can be bought ex-vessel at Mwaloni Kirumba for TShs 5000/-, or the equivalent of TShs 25/- per piece. These same pieces, if sold off individually to other wholesalers at the Mwanza market, could fetch about double this amount. However, as with dagaa, much better profits can be realised by shipping them inland, providing that transport is available. In Dar, for instance, pieces of smoked sangara sell for TShs 75/- each. Mzolo and his fellow traders at Kirumba Beach are clearly finding business to be quite lucrative.

Traders at Kirumba Beach occupy a second tier in the local marketing structure, receiving consignments of dried and smoked fish from myriad artisanal processors in outlying areas and channelling them through other agents for distribution in Mwanza and inland, to virtually all regions of Tanzania and into neighbouring countries as well. Most of those involved in the fish trade appear to operate on a far more modest scale of investment and return than these intermediaries. In lakeshore communities and areas immediately inland, the traffic in fish is largely in the hands of petty traders, who move between the landing sites and neighbourhood markets on foot, by bicycle, or using public transport where available to buy and sell in very limited quantities. Whereas an agent like Mzolo counts his business turnover in the tens of thousands of shillings and deals in tons of fish, the multitude of very small-scale enterprises work in terms of a few hundreds of shillings and headloads of fish.

Buyers of fresh fish purchase directly from canoes as the crews pull into landing sites each morning from the fishing grounds. At larger sites like Chakula Barafu in Mwanza, where fishing canoes are landing alongside freight and passenger boats and an extensive outdoor market in goods of all description exists, large numbers of people assemble and buying and selling activity can be intense. Traders crowd the water's edge, vying for the attention of fishing crews to strike quick bargains. Fish are hurled through the air as they are off-loaded from canoes, and onlookers must be prepared to dodge nimbly to avoid collision with large Nile perch aimed at the feet of a buyer. Catches are sold according to size and type of fish. Weighing scales exist only in the larger municipal markets but even there are rarely used to fix prices. If large fish are to be sold fresh, they are often first cut into chunks which can be offered individually at more accomodating prices to buyers. The market for fresh fish is best in bigger population centres like Mwanza, Musoma, Bunda, and Tarime, which are either in close proximity to landing sites or served by regular public transport links. Occasional shipments of Nile perch can be arranged by larger trading concerns using the few refrigerated wagons of the railway, though this mode of transport reportedly has not been as effective as originally planned. Some more enterprising operators use contacts in the air services to arrange special shipments of fish to Dar es Salaam. Fishmongers in the Musoma Municipal Market know that on days when a plane is due to depart for Dar, they will sell an extra 400 to 500 kg. of fresh sangara.

A great proportion of fish handled by local traders is processed in some fashion because the catches cannot be moved to marketplaces expeditiously nor put into coldstores, which are virtually nonexistent. Prices for processed fish run 20% or more higher than for fresh at retail markets, except for small fish like haplochromines, dagaa, and juvenile Lates which usually are dried as a matter of course because this is the easiest method of handling them. Dagaa are simply spread out on beaches in the sun, sometimes directly on the sand or ground and sometimes atop layers of cut grass. Juvenile Lates are treated in the same way as haplochromines, in a manner witnessed at Kisaka Beach. The fish are allowed to dry in the sun for several hours atop layers of grass, and scorched by fire as the grass is set alight. The processed tilapia seen during visits around the lakeshore were usually sun-dried but sometimes smoked or fried in fat. Kiln smoked Bagrus and Synodontis were also seen.

By far the greatest number of processing operations observed during the mission were concentrating on Nile perch. Frying and kiln smoking are the methods employed. Larger specimens are cut into chunks prior to processing, while smaller ones are laid open lengthwise to lie flat on the smoking screens. In some cases the smaller fish are simply fried whole. Fish frying is widely done with oil recovered from the fatty deposits of large Nile perch. Such oil has become an important commodity in local markets. At Chakula Barafu Beach, for example, many traders were selling it in bottles and tins for a price of TShs 60/- per litre. Consumers use it generally as a substitute for industrially manufactured cooking oil. Its use in soap making and medicinal applications, as has been reported by others, was not directly witnessed. Nile perch swim bladders are also recovered by some processors. Cleaned and dried, they are reportedly sold for TShs 40/- to 60/- per kg. to dealers who bulk and resell them to agents in Tanga or across the border in Kenya.

Case Example T6: Processing at Igombe Beach

Igombe is a landing site for some thirty or so gillnet canoes and is also a centre for beach seining operations. A few shops stand in back of the beach around a small market square and a number of dwellings and homesteads are clustered nearby. Fish processing is a major occupation in the village and mud-walled smoking kilns, frying hearths, and drying tables are scattered throughout the neighbourhood. Igombe fish processors are a mixed lot. Some are local Sukuma people who combine their business activities with farming. Others are immigrants from as far away as Luoland in Kenya. Those who own kilns often rent them out to the immigrant processors when not using them personally, collecting a fee of TShs 70/- for each sheet of wire mesh hired. Almost all of the twenty or so processors seen on the day of the mission visit were men, though women were observed to be cleaning fish at the lakeshore and operating a few kilns on their own. One woman was busy cleaning Nile perch swim bladders obtained for free from other processors, who had no time to deal with them. After drying, she planned to sell them to a buyer from Mwanza. Nile perch heavily predominate in the catches landed at Igombe. Ex-vessel prices depend on the size of fish, and were estimated to run between TShs 10/- to 15/- per kg. After being cleaned and either laid open (smaller fish) or cut into chunks (heavier fish), sangara are placed on tables to dry in the sun prior to smoking or frying. The fish are fried in their own fat, but the firewood needed either for this process or for smoking must be purchased from outside suppliers and is expensive. One smoking session for two or three racks of fish (100 kg. or more) costs the processor about TShs 250/- for firewood at Igombe. Once a sufficient quantity of fish have been prepared and packed for shipment, arrangements must be made for transport to markets. This can be done either by the processors themselves or through other agents. Most of the Igombe fish go to Mwanza, where they may be sold directly or re-shipped to points inland. Fried fish were reported to reach consumers in Tabora and Kigoma, but were not shipped any further afield because of the danger of spoilage. The income of Igombe traders varies widely, though none of them seemed to be exceptionally prosperous. A person can work full time or upon occasion, and can pocket anywhere from a few hundred to a few thousand shillings as proceeds from each consignment of processed fish, depending on the quantity involved, costs incurred (supplies of fresh fish and firewood, kiln rental if applicable, and transport), and final marketing arrangements.

3.3.5 Perceptions of Local Fish Traders

Fishmongers running enterprises of all ranks whether large or small seem to share a high regard for Nile perch. Nowadays it is abundant, in high demand especially amongst upcountry consumers, and profitable. As one trader at Semba observed, “People had to get used to sangara because it was new and different. But now we see how good it is. Nearly every part of that fish is business.” As in the case of fishermen, the conversation of traders did not dwell overmuch on concerns with changing species stock balances in the Lake brought on by the new fish. What was repeatedly stressed instead were the problems which impede the more effective pursuit of sangara as an object of commerce.

These problems are principally of two kinds and are interrelated. One bears on the firewood supplies needed for processing, and the other on movement of fish to markets. Fish processors are well aware that the upsurge in Nile perch has caused a huge increase in demand for firewood, which in some areas was already in short supply. Smoking is an old and standard method of preserving fish around the lakeshore, whereas frying is a recent innovation that seems to have developed in association with Nile perch. Whatever the method employed, the basic fact is that far more fish are being processed these days and that firewood resources are being consumed as never before. In the case of kiln smoking, it may be that Nile perch by nature require more firewood per smoking session than equivalent amounts of other fish. A woman fish smoker at Nassa, for instance, while not complaining about the proceeds she now receives from sangara, remarked that she used less fuel in the days when she specialised in other fish like Bagrus. “These sangara are ‘heavy’ fish,” she said, “and take longer to smoke properly. I have to use more kuni [firewood] with them.” What she may have been remembering also was that firewood was both more plentiful locally and far more cheap in the Nassa area several years ago than it is at present. Nowadays wood merchants truck it in from outside at the rate of TShs 7000/- a ton, a quantity sufficient to smoke perhaps two or three tons of fish, if carefully managed. The going rate for firewood in the other areas visited is about the same. A large traffic in imported firewood has long existed in the Mwanza area, and has swollen considerably with the coming of Nile perch, according to local supply agents there. At Shede Beach one supplier was interviewed who was selling firewood that originated all the way from Maisome Island, at the entrance to Emin Pasha Gulf in the extreme southwestern corner of the Lake. He does a good business, selling small bundles of three split pieces (weighing roughly 10kg per bundle) for as much as TShs 50/- each. In some localities of Tarime District in Mara Region, fuelwood is in such short supply that processors have resorted to using dried animal dung for fish smoking. Forestry officials in Mara reported that deforestation in this region has reached severe proportions directly as a result of artisanal fish smoking operations, and that restrictions on cutting are everywhere becoming more difficult to enforce as demand for fuelwood mounts.

To a great extent the lack of efficient transportation infrastructure and cold chain facilities contributes to problems of firewood shortages and deforestation. Opportunities for distribution of fresh fish, be they Nile perch or any other, are extremely limited. Fishmongers are therefore left with little choice but to employ such preservation methods as are locally known and feasible in order to extend for a week or two the shelf life of their products, thus enabling them to reach more consumers. Yet even preserved Lates is subject to inland market distribution limitations because of poor road conditions and slow and scarce transport services. Furthermore, as traders emphasise, just getting fish moved through the preliminary transfer routes from landing beaches to the principal lakeshore market and inland distribution centres is difficult enough. Considerable wastage can occur. Fresh fish for sale at the Mwanza Municipal Market may be 12 or more hours dead by the time they are put out for display; if not sold within a short while thereafter they must simply be thrown out. Mwanza fishmongers even have to throw out putrid and larvae-infested loads of processed fish on occasion. At Makoko near Musoma, the personnel who run the Catholic seminary's trawling and trading operations deliberately keep catch levels to a minimum because they know they would never be able to deal with larger quantities of fish, whether fresh or smoked, because of transport difficulties. Traders know full well that they could sell more fish providing they could deliver to the right places in the right time, but this combination is unfortunately extremely difficult to manage under circumstances now prevailing in Tanzania.

3.3.6 Perceptions of Local Consumers

The strong commercial success achieved by Nile perch in a relatively short period of time attests to a rapid turnabout of consumer attitudes towards the fish. Around the vicinity of the Lake, initial reaction was apparently unfavourable in the extreme. Sangara was too oily, it smelled, and it caused vomiting and “running stomachs” when eaten. A few of those market customers and local residents interviewed still harbour such views. But the clear tendency now is to think of Nile perch as a most useful fish -- cheap, plentiful, and even, after all, rather palatable. This does not mean that it has become the top food preference of the majority of fish consumers. Those who dwell close to the lakeshore have always had the advantage of being able to pick and choose from a variety of species and have by no means abandoned their old favourites. Indeed, the testimony of fishmongers in the large municipal markets of Mwanza, Musoma, and Tarime indicates that tilapia continue to be the fish of choice for most local consumers, and that other fish such as Bagrus are also generally more highly esteemed for their taste than Lates. Nevertheless, the newcomer has without doubt become a popular item.

Traders with contacts in places like Shinyanga, Tabora, and Dodoma, report that upcountry consumer reaction to Nile perch was very positive almost from the outset. In the words of one of the committee members of the Mwanza African Fish Dealers Co-operative Society, “The people who live in those places far from the Lake were never able to have much fish to eat until sangara came. Fish is the best food; meat and chicken are more expensive. And sangara is one of the best fish because it is easy to get and it sells at a low price. Everyone these days knows there would be serious problems getting enough fish if it weren't for sangara. That is why it is called ‘mkombozi’.”

3.3.7 Industrial Operations

The frozen Nile perch fillet business which has grown so vigorously in Kenya is not at all evident in the Tanzania sector of the Lake. As indicated earlier, geographical remoteness and infrastructural problems have been obstacles to the establishment of industrial processing. The facilities of the parastatal Nyanza Fishing and Processing Company could conceivably be utilised but reportedly are in some disrepair and have not been adapted to deal with Nile perch after the collapse of the Haplochromis fishmeal venture. In the private sector, however, some novel and interesting developments are taking place. A Nile perch canning and packaging operation is being organised by the Mwanza company whose pair trawling operations were reviewed as a case example earlier. This appears to be a promising enterprise.

Thus far the Company has been running pilot operations out of Nyegezi. Trial canning is being done at the Fisheries Training Institute, using a small plant recently rehabilitated for the purpose. Much larger facilities are planned for construction on a site adjacent to the Nyanza Fishing and Processing Company in the Pasiansi area, at a cost of some $1.6 million. These will include an ice plant, cold room, cleaning and processing space, canning equipment, and a smoking kiln. Most of the factory's machinery and equipment will be supplied from abroad. Packing tins (150g. size) will be supplied by the Metal Box Co. Ltd. in Dar. Projected production levels are ambitious. It is hoped that up to two tons of Nile perch will be canned per day. Additional fish will be processed and packaged as frozen and smoked fillets. Supplies of fresh fish will come from the Company's pair trawling operations, though these will have to perform far more effectively than they do at present in order to meet requirements. Alternative supplies of fish from other trawling operations or from artisanal operators may have to be organised.

Canning would seem to be a particularly attractive method of fish preservation given the difficult conditions of transport now prevailing. Products packaged in this form have a prolonged shelf life and can be moved over great distances to domestic markets and export destinations both in neighbouring countries and overseas. Trials thus far have shown that Nile perch cans very well indeed. The fish have been packed in both tomato sauce and vegetable oil mediums, and sampling confirmed that they retain a delicate flavour and firm but flaky texture, rather like that of tinned tuna. Domestic consumer taste surveys have not been undertaken as yet but it is known that there is a market for tinned fish, as steadily increasing amounts have been imported over recent years. The European export market for canned Nile perch has not been fully appraised, but canning experiments in Norway have shown good results.


The basic points brought out by review of the statistical and other reports pertaining to the Tanzania sector of Lake Victoria fisheries, and of the direct field observations taken during the mission, can be summed up as follows:

Previous Page Top of Page Next Page