Lake Victoria is the world's second largest freshwater body and home to one of the most dramatic speciation of indigenous cichlids in the world. Bordered by Kenya, Tanzania and Uganda in East Africa, the Lake Victoria Basin (LVB) provides food, water and livelihoods for over 30 million people around its shores, with 10 million engaged in the fisheries sector. The Colonial Era introduction of the invasive Nile perch (Lates niloticus) in the 1950's, combined with the introduction of industrialized fish processing in the 1980's, transformed fishing throughout the Basin. The introduction and commercial harvesting of the Nile perch, subsequent human population growth, and the looming problems of climate change continue to compromise the health of this important fishery. This paper applies a global commodity chain framework with an ethnographic approach to the case of the export-oriented Nile perch from the Kenyan island of Mfangano in July 2007. Unless otherwise noted, all price and empirical data are based on interviews, market surveys, and participant observation conducted by the author in Kenya in June, July, and August of 2007. In 2007, prices paid to local Kenyan fishermen, boat owners, and agents represent 24% of the total value from the fishery (∼ 3% to each fisherman, 14% to boat owners and 4% to agents); prices paid to processors represent 36% of the value, and 39% of the value accrues to international seafood wholesalers and retailers.
The commodity chain for Nile perch (Lates niloticus) begins on hooks and lines of impoverished artisanal fishermen on islands like Mfangano in Lake Victoria, and ends on the dinner plates of affluent consumers from Madrid to Tel Aviv. While the Nile perch trade provides much needed foreign exchange to the three exporting nations of Kenya, Uganda, and Tanzania, the socioeconomic benefits of the Nile perch export trade have not accrued evenly across all actors involved. For each kilogram of fresh, exported Nile perch, 44% of the value accrues at the wholesaler and retail level in importing nations, 33% accrues at the processing level in the basin, 4% accrues to basin-based agents collecting fish from landing sites and transporting it to industrial processing facilities, 14% accrues to basin-based boat owners and about 3% accrues to each fisherman.
To help understand how value is distributed in the Nile perch trade, a global commodity chain (GCC) framework outlined by Gary Gereffi and Miguel Korzeniewicz is applied to the case of Nile perch from the Kenyan island of Mfangano in Lake Victoria (Gereffi and Korzeniewicz, 1994). While commodity chain analysis is useful to provide a snapshot of resource harvesting, processing and trading, it provides a static picture of a dynamic system, and should be considered within larger historic and current, political and economic transformations contributing to the distribution of value along the chain.
Commodity chains can be market-driven, buyer-driven, producer-driven, or in the case of the Lake Victoria Nile perch, international buyer and trader-driven. International traders and buyers exercise considerable control over the management and trade, from fisheries regulations to certifying the safety of processing plants that gut and fillet Nile perch for export (Thorpe and Bennett, 2004). According to Peter Gibbon, international trader-driven markets are characterized by: (1) relatively low value-to-weight ratios, with labor intensive direct raw material production and with low barriers of entry to begin harvesting; (2) globally dispersed and locally discontinuous supply patterns; (3) strong tendencies toward market saturation, partly due to partial substitution by “new” agricultural or manufactured products; and (4) a final or intermediate demand side which is either dispersed or concentrated, but segmented with respect to commodity variety (Gibbon, 2001). Nile perch, in comparison to higher-valued fish species, such as Bluefin tuna, has a low value-to-weight ratio. Extraction requires intensive fishing day and/or night by anyone able to afford a gill net and canoe. Once Nile perch is processed, fillets are transported globally and are largely unavailable locally (Nile perch “frames” are traded and consumed locally and used in animal feed and unknown quantities of sub-legal juvenile Nile perch are also traded locally). International markets for Nile perch face steep competition from other whitefish, such as whiting, catfish and tilapia, and are segmented by dietary standards and consumer preferences for fresh or frozen fillets. These global consumer preferences, exercised by the processes and practices of international traders and basin-based processors and fish agents, are one driver of Nile perch fishing around the basin.
This commodity chain is based on three months of ethnographic and market research conducted in Kenya, Uganda and Tanzania. 16 formal interviews were conducted with fisheries managers and research scientists with the Lake Victoria Fisheries Organization, Moi University, the USAID Commercial Aquaculture Program, and the Departments of Fisheries Management and Institutes of Fisheries Research in each basin country. Over 20 interviews were conducted with subsistence and commercial fishermen and fish farmers in the three basin countries. 8 interviews with fish processors, including artisanal processors and industrial processors were also conducted. Additionally, participant observation was conducted by fishing, processing fish and observing fish market transactions on Mfangano Island, Kenya and in Mwanza, Tanzania. Unless otherwise stated, all price and empirical data were based on interviews, market surveys, and observations conducted by the author in Kenya in June of 2007.
The Fisheries of Lake Victoria
The Colonial Era introduction of the invasive Nile perch in the 1950s and 1960s, combined with the introduction of industrialized fish processing in the 1980s, transformed fisheries throughout the LVB. Prior to the introduction of Nile perch in the mid-1950s, an artisanal multi-species fishery supplied local and regional markets with a diversity of fish species, including hundreds of small Haplochromine cichlids (Goldschmidt, 1996). With the proliferation of the Nile perch, hundreds of endemic haplochromines were predated into extinction, profoundly altering the lake's aquatic food web (Ogutu-Ohwayo, 1990; Witte et al., 1992). At the same time, fishermen began to catch individual Nile perch of over 100 kgs, a size of fish never before experienced by indigenous fish processors and traders. While some fishermen began referring to the fish as a “monster” and others a “savior,” many of the early catches rotted on shore due to the lack of local processing capacity or local and international demand (Abila and Jansen, 1997).
Realizing the opportunity to trade whitefish commodities from Lake Victoria in global seafood markets, regional and international investors and entrepreneurs built and began operating industrial processing plants for Nile perch around the lake. The Nile perch and simultaneous ecological transformations in the lake basin (including nutrient loading, the proliferation of the water hyacinth and subsequent eutrophication) have transformed this once diverse fishery into a tri-species fishery, dominated by the Nile perch, caught, processed and sold for export to developed nations. The other two species, omena (Rastrineobola argentea) and Nile tilapia (Oreochromis niloticus), are still processed and traded throughout the region. Presently around the world and in Lake Victoria, wild fish stocks are overexploited and vulnerable human populations increasingly lack access to reliable sources of protein (Pauly et al., 2005; Abila, 2000).
Lake Victoria is the most productive freshwater fishery in Africa. Annual fishery yields from the lake have fluctuated around 600,000 tons, valued in 2006 at $350–$400 million U.S. at the beach, with total export earnings estimated at $250 million US (LVFO, 2009). Around 80% of Nile perch harvests caught today is sold directly to industrial fish processing plants located around the basin. Once processed, they are destined for commercial export to Europe, Israel, the Middle East, the United States and Asia (LVFO, 2009). Despite increased catches from the 1980s-early 2000s, the development of the Nile Perch export industry, and investments in more technologically advanced gear and processing, the region has experienced significant decreases in food security as Lake Victoria fisheries have shifted from primarily a local artisanal fishery to a global industrialized fishery (Abila and Jansen, 1997). While there are a number of additional factors contributing to the transformation of this fishery, including significant population growth, eutrophication, deforestation and the introduction of other invasive species, the commodification of the Nile perch export fishery has profoundly transformed local economies, gender relations, and fish consumption in fishing-dependent communities around the lake.
Mfangano Island, Kenya
Mfangano Island, with a total population of over 16,000, is part of Kenya's Nyanza Province and is the farthest Kenyan island from the mainland, located about 10 kilometers from Kenya's western shore (IPUMS, 2001). Fishing, trade in citrus fruits, and mission aid form the backbone of Mfangano's economy. The transportation of people and goods on and around the island is mainly by boat, as there are no roads or electricity.
The island is comprised of Luo and Suba people, who arrived from Sudan and established artisanal fishing villages several hundred years ago (Herring, 1979). Artisanal fishing and processing continued to thrive on the island until the proliferation of the Nile perch in the late 1980's. Livelihoods on the island initially improved with the influx of cash from international trade in Nile perch. Fishermen were able to replace thatched rooves with corrugated iron and money from the fish trade was used to build primary schools. The Nile perch trade also produced disparities, marginalizing female artisanal fish processors and traders. Women were the primary processors and traders of native fish on the island. As the Nile perch consumed native fish species and fishermen increasingly targeted Nile perch, Mfangano's women lost access to the diversity of traditionally traded fish species, leaving them with few livelihood opportunities outside of drying and trading Rastrineobola argentea and subsistence farming. As argued by Kim Geheb et al. (2008), this transformation placed cash as the most contested commodity in the household, a commodity largely controlled by men who dominate the Nile perch trade.
HIV/AIDS rates on Mfangano Island are some of the highest in the world at 38% of the total population, compared to around 8% for Kenya overall (WHO, 2003; WHO, 2005). This local trend is consistent with the global literature reporting higher HIV/AIDS rates in fishing communities (Allison and Seeley, 2004). The high HIV/AIDS rates are attributed to the mobile nature of fishing, exposing fishermen to new people and new diseases. The lack of livelihood opportunities for women also contributes directly to HIV/AIDS infection rates, as women fish traders may be expected to engage in relationships with the fishermen who supply their catch to ensure access to future catches.
From Mfangano to Madrid
The commodity trade in Nile perch began in the early 1980s, bolstered by the development agenda of the Washington Consensus, promoting export-oriented growth and the removal of trade barriers. The ecological supply chain for Nile perch began with the Colonial Era introduction of the species in the 1950s continued on with the proliferation of the perch throughout the next thirty years and is now a permanent facet of Lake Victoria's aquatic life. The social side of the Nile perch supply chain can be defined as the set of interdependent agents (fishermen, agents, processors, and distributors) that work together, either knowingly or unknowingly, to bring Nile perch to consumers (Thorpe and Bennet, 2004). The percentage of total value accrued at each level of the commodity chain is obtained by: dividing the price paid at each level of the chain minus the price paid at the level immediately below, by the total price paid. For example, to assess the value to processors in 2007, the price paid by processors to agents per kilogram of whole fish ($1.90) is subtracted from the price paid by EU wholesalers for a kilogram of fillets ($4.60) and divided by the price paid by EU consumers ($8.23) for a total of 33% of total value. In 2007, prices paid to local fishermen, boat owners, and agents represent 24% of the total value from the fishery, prices paid to basin-based processors represent 33% of the value, and 44% of the value accrues to international seafood wholesalers and retailers (Table 1).
Please note: (1) 1 kilogram of filleted Nile perch represents around 2.4 kilograms of whole fish. The prices given are for filleted fish from consumers down to processors and for whole fish from agents to fishermen. All prices were collected by the author in Kenya in July of 2007, unless otherwise noted. (2) The price of $8.23 is for a kilogram of fresh fillets. (3) Each fishing boat targeting Nile perch is operated by at least two fishermen who receive 6% of the total value, each fisherman receiving only 3% of the total value.
Like most global commodity chains, the bulk of the economic value from the Nile perch trade is earned at the processing, wholesale and retail levels in importing nations (Gundmundsson et al., 2006). Approximately 80% of all Nile perch caught in the lake is exported, while the other 20% is consumed by animals as feed or by local people as sub-legal juveniles, or as “fish frames” (LVFO, 2009). “Fish frames” consist of the head and small pieces of flesh remaining on the Nile perch skeleton after the industrial filleting process. While the local trade in juvenile perch and Nile perch frames for human and animal consumption is economically and nutritionally important, for the purposes of this study only the export-oriented portion of the commodity chain will be addressed further. A graphical representation of the commodity chain is given in Figure 1.
As of June 2007, basin-based fishermen, traders, and processors receive around 56% of the value of Nile perch through harvesting, bulk collection, and transport (see Table 1). While 24% of this value is distributed locally to fishermen, boat owners, and agents, because each fisherman only receives 3% of the value, the majority of the value to locals in the basin accrues to a small number of boat owners and processors and not the thousands of those doing the actual fish harvesting. International wholesalers and retailers of Nile perch receive 39% of the total commodity value by marketing, transporting, and repackaging the primary produced perch fillets. A breakdown of the value added at each link in the commodity chain as of June of 2007 is given in Table 1.
Fishermen (local actors)
Fishermen catch Nile perch in Lake Victoria with longlines, gillnets, and beach seine nets off islands like Mfangano Island in Kenya. The fleet of vessels on Lake Victoria ranges from small paddled canoes with sails and outboard motors to larger motorized canoes ranging from 8–45 feet in length, some equipped with outboard motors from 5–55 horsepower. Many boat owners are not fishermen themselves, but rather pay fishermen a percentage of the total catch. In July 2007, boat owners on Mfangano were paying each fishermen targeting Nile perch 10–15% of the total catch, which is approximately 3% of the total value per exported kilogram. While most fishermen target Nile perch, fishermen with small boats catch mostly lower valued omena and Nile tilapia for local and regional markets, fishermen with larger boats, capable of fishing at greater depths and distances from the shore, target the higher valued Nile perch.
With the proliferation of the Nile perch in the 1970s and 1980s, fishing effort increased rapidly on the lake, with the number of nets increasing by over 100% from 1970–1981 (Reynolds and Greboval, 1988). Recent increases in the number of fishermen, fishing vessels and industrialized fishing gear in Lake Victoria have been equally steep. From 2000–2005, the number of fishermen in Lake Victoria increased by 52%, from 129,305 to 196,426. The number of fishing vessels also increased during this time by 63%, from 42,493 to 69,160, while the number of fishing crafts using outboard motors increased by over 200% from 4,108 to about 12,000. The total number of gillnets used in the lake increased by 88%, from 650,653 in 2000 to 1,222,307 in 2006, while the use of longline hooks increased by 61% from 3,496,247 to 9,044,550 hooks (LVFO, 2009). Industrialization of the fishery and associated ecological and socio-economic implications have further transformed the aquatic ecosystem and the fishing way of life in Lake Victoria (Abila and Jansen, 1997).
Fishermen themselves tend to be primary school drop outs and in many cases orphans, who turned to fishing as a last resort to support themselves.
Fish agents (local actors)
Each morning fish agents fill up their canoes with ice supplied by fish processing plants, if available, and head out to collect fish from landing sites around the lake. Agents have an established daily route and often longstanding relationships with both fishermen and other agents. Agents collecting fish from the eleven fish landing sites on Mfangano Island take the fish to collection sites near Mbita Point, where large ice-filled trucks wait for several days to be filled with enough fish to justify a trip to urban processing plants. The fish agent interviewed for this study happily mentioned that he was making a good living from his work. While the economic value from this portion of the commodity chain accrues locally, it is important to note that the two agents operating on Mfangano and the surrounding islands in the export-oriented fishery replaced unknown numbers of artisanal fish traders who generated their livelihoods from trading Nile perch and other species prior to the export-oriented trade.
Agents tend to be much better educated, live in urban areas and have a higher socio-economic status than fishermen. The agent interviewed on Mfangano Island has a college degree, and lives near the collection site on the mainland. In June of 2007, fish agents paid fishermen the equivalent of $1.60 per kilogram of Nile perch and sold it to processing plants for around $1.90 per kilogram.
Fish processors (regional and transnational actors)
With the proliferation of the perch, foreign and basin-based investors and entrepreneurs realized the opportunity to trade Nile perch from Lake Victoria on international markets. They built and began operating 34 industrial fish processing facilities in the LVB (12 in Kenya, 10 in Uganda and 12 in Tanzania), capable of processing up to 25 tons of Nile Perch fillets per day for commercial export (LVFO, 2009; Ntiba et al., 2001). In the early 1990s an investor required just over $330,000 US to establish a filleting plant comprised of a ten-ton cold storage facility, a blast freezer with a batch capacity of six tons, a two-ton ice plant, and six five-ton insulated trucks (Ogunja et al., 1992) It was estimated that these processing plants could break even if they produced at one-third of their maximum processing capacity.
Initially, processors invested in both western-style processing plants and industrial fish trawling vessels to exploit fish far offshore (Mbuga et al., 2000). Due in part to Kenya's smaller share of the lake, Kenyan processors invested more heavily in trawling vessels than either Uganda or Tanzania, and six individual trawling vessels were operating from Kenya by the early 1990s. These vessels fished heavily in Ugandan and Tanzanian waters and were a significant source of conflict between the three nations. By 1995, trawling was officially banned in the lake due to negative ecological impacts associated with this gear type (overfishing, habitat damage and bycatch). However, despite the ban trawling vessels remained in operation on the lake until 1999 (Mbuga et al., 2000).
In order to maintain control of fish supplies without trawling vessels, fish processing companies hired agents to collect fish from rural fish landing sites and eventually deliver them to processing plants, located near urban centers, such as Mwanza in Tanzania, Kisumu in Kenya, and Entebbe in Uganda. The agents “sponsor” fishermen on behalf of the processing plants by supplying gear (boats, nets, motors, and bait) in return for guaranteed catches at fixed prices. While this is favorable for processors, fishermen, despite the formation of fishing cooperatives and efforts to provide storage facilities at landing sites, most boat owners are unable to negotiate fish prices with agents and must take the price offered.
Fish processing plants around the lake, due to concern over the safety of fish, must be Hazard Analysis and Critical Control Point (HACCP) certified to export fish to Europe (Ponte, 2007). In 1997, concern over Salmonella found in imports of Nile perch to Spain led to a temporary import ban on Nile perch to the European Union. The ban was lifted in July 1998, but reinstated in April 1999 in response to the concern that fishermen were catching fish using pesticides. Following a dramatic restructuring of processing facilities around the lake and creation of mandatory fish inspection for pesticides and heavy metals, the ban was lifted in October of 2000 (Henson and Mitullah, 2004). As a result of the ban, the three basin countries began developing trade with nations in the Far East and Middle East, whose food safety standards were not as stringent as Europe's at that time (Australia, Japan, Hong Kong, Singapore, Malaysia and China) and in the Middle East (mainly Israel and the United Arab Emirates) (Abila, 2006). Kenya was the last country to upgrade its processing facilities to HACCP standards (in 2002) and as a result the proportion of its exports to the European Union continue to lag behind both Tanzania's and Uganda's.
In 2007 over 25 processing plants were in operation, each with a processing capacity up to 25 tons of fish per day (Ponte, 2007). In June 2007 many processing plants in the basin were operating at less than 50% of total capacity, due to a lack of supply and overcapitalization in processing equipment. This excess fish processing capacity is a driver of high exploitation rates, despite fisheries management interventions and is a primary rationale for the expansion of industrial aquaculture in the basin.
Fish retailers (transnational and multinational actors)
Major global retailers import fresh and frozen Nile perch fillets into Europe, the Far East, and the Middle East. These companies add value to the fillets by packaging and seasoning the perch to suit individual tastes in each country. For example, Nile perch imported to Israel is packaged and sold with the skin on, to suit local kosher standards, while fillets preferred in Europe are sold plain or marinated.
In June of 2007, European importers paid basin-based processors $4.60 per kilogram of frozen Nile perch and sold it to consumers in the European Union for $8.23 per kilogram. According to basin based fish processors and fishery managers, the biggest threat to maintaining the value of the Nile perch export trade is the availability of lower-priced whitefish substitutes, such as farmed tilapia and catfish from Asia.
Large retailers maintain control over harvesting, processing, and transport in the three basin countries through the regulatory requirements of global certifying bodies, such as the HACCP. International buyers maintain control of the trade through their suite of preferences for the final product. The need to meet these international standards and preferences dictates processing practices and defines the type and quality of fish available for consumption by disparate types of consumers. While local managers and scientists may operate with the best of intentions, they are carrying out the mandates of global traders and buyers on top of this global commodity chain.
Through the creation of an export-oriented Nile perch fishery, fishing has been largely transformed from a subsistence activity to an income generating one. By adding value to Nile perch through industrial processing, transport, and repackaging, the commodification has transferred monetary and nutritional benefits of the fishery from fishing dependent communities in the LVB to shareholders and consumers in the developed world.
The export oriented Nile perch fishery, driven by global demand and the subsequent influx of industrial fishing technology, transformed the fishing way of life in the LVB. The sustained commitment of importing nations and basin-based managers to maintaining steady fish supplies to Europe is reflected in the definition of fishery sustainability in operation in the basin (as maintenance of Nile perch exports) and targeted European aid to the fishery sector in the LVB.
Over the past twenty-five years, the global trade in Nile perch has compromised ecological and human well being around the lake in favor of profit for global capital and gastronomic pleasure for consumers. In part due to eutrophication, habitat created by invasive water weeds, and reduced predation due to falling Nile perch stocks, cichlid species once thought to be extinct are returning. The ability of the cichlid to adapt and repopulate itself is testament to the resilience and adaptive capacity of the LVB's aquatic ecosystem. Restoration of the cichlid offers an alternative path for the LVB, a chance to understand the fishery transformation perpetuated by the Nile perch trade and chart a new course. There is growing pressure to promote fisheries that feed people who are hungry first, empower fishing people who are marginalized, and allow ecological adaptation to the many stresses of climate change, pollution, and overfishing. Through a better understanding of the social and biological complexity within this social ecological system, an opportunity will arise to wake up from “Darwin's Nightmare” and co-author a conscious tale with a different ending.
This research would not have been possible without the generous financial support from the University of Michigan Center for Afro-American and African Studies, the Doris Duke Conservation Foundation, the University of Michigan International Institute, Rackham Graduate School, and the School of Natural Resources and the Environment. I am deeply in debt to the scientists and fishery managers in the basin who answered my many questions frankly and enthusiastically. I remain forever grateful to the women and men of the fishing communities visited throughout this research, particularly of Mfangano Island, Kenya and Mwanza, Tanzania – thank you for sharing your stories, struggles, and hopes.