Marketing systems for fish from Lake Tana, Ethiopia:
|ARARI||Amhara Region Agricultural Research Institute|
|BoARD||Bureau of Agricultural and Rural Development|
|CIDA||Canadian International Development Agency|
|CSA||Central Statistical Authority|
|FAO||Food and Agricultural Organization of the United Nations|
|FPME||Fish Production and Marketing Enterprise|
|GDP||Gross domestic product|
|IFPRI||International Food Policy Research Institute|
|ILRI||International Livestock Research Institute|
|IPMS||Improving Productivity and Market Success of Ethiopian farmers|
|JTH||Jigerfa Tekle Haimanot (a collection point on Lake Tana)|
|MoARD||Ministry of Agriculture and Rural Development|
|PLW||Pilot learning woreda (of IPMS)|
|PRA||Participatory rural appraisal|
|UNDP||United Nations Development Programme|
|WOARD||Woreda Office of Agricultural and Rural Development|
Ethiopian birr (ETB)–unit of currency of Ethiopia; approximately ETB 8.7 = USD 1 in June 2006
kebele this is the sub-district administrative unit, akin to a parish tanqwa this is the traditional reed-boat used in Ethiopia, usually made from papyrus (a more sturdy bamboo version is used on some of the southern lakes)
woreda this is the sub-regional administrative unit, akin to a district Dates used in this report follow the Gregorian calendar. (The Ethiopian year starts in September. The year started in September 2006 was 1999 in Ethiopia.)
The team is very grateful to all those in governmental, non-government organizations (NGOs) and particularly in fishing communities and along the marketing chain, who generously gave of their time and their views. Improving Productivity and Market Success of Ethiopian farmers (IPMS) staff should also be thanked for their help in arranging the visit, with particular thanks to Kebede Assefa, the team’s driver and guide par excellence. The team is also grateful to Noah Kebede, of IPMS, who provided the maps included in this report. Dr Berhanu Gebremedhin, IPMS marketing expert and Dirk Hoekstra, IPMS project manager are gratefully acknowledged for providing feedback and comments on the draft report. The content is however the sole responsibility of the authors.
This report concerns the domestic trade in frozen fish sourced from Lake Tana in Ethiopia. It has been prepared for the Improving Productivity and Market Success (IPMS) of Ethiopian farmers project implemented by the International Livestock Research Institute (ILRI) on behalf of the Ministry of Agriculture and Rural Development (MoARD) of Ethiopia. The project seeks to contribute to the commercial transformation of subsistence agriculture through improved consideration of marketing opportunities and constraints. The field work was undertaken in June 2006 by a team comprising representatives of the Fisheries Research Unit of the Amhara Region Agricultural Research Institute (ARARI) on Lake Tana, Fogera Woreda Office for Agricultural and Rural Development (WOARD) and the WorldFish Center.
The Ethiopian market for frozen fish is small but growing rapidly. The market for frozen fish is heavily concentrated in Addis Ababa and urban areas in the production zones. Day-to-day and seasonal variations are extremely marked, reflecting uneven supply patterns and high demand on Christian fasting days and months. The main products are frozen fillets of tilapia, Nile perch, catfish and barbus. Under relatively conservative assumptions, aggregate demand for fish could be expected to grow by around 44% over the next 10 years. Historically, Addis Ababa has been supplied from the lakes in the south of the country. There is, however, a widely-held view that these lakes are now over-fished and market participants report growing difficulty in sourcing fish in the last 2–5 years, suggesting that demand is now growing faster than supply.
Lake Tana’s importance in this trade is growing but a number of constraints were noted, including:
amongst fishermen and traders, including the use of contracts, would enhance market development.
A number of market participants also indicate capital constraints, expressing interest in acquiring, e.g. large cold storage facilities and refrigerated transport. However, apparently low margins in the current trading system suggest that very careful analysis of the returns to such investments would be needed.
Fish entering the frozen fish marketing system from Lake Tana is not a significant source of livelihood in Fogera woreda currently. The recommendations, however, focus on improvements that would have wider livelihood benefits elsewhere on Lake Tana and in other fishing areas:
study focused on an improved understanding of the marketing system be carried out for dried catfish, with a view to identifying opportunities to improve/expand livelihoods.
This is a report of a study undertaken as part of the Improving Productivity and Market Success (IPMS) of Ethiopian farmers project. The project is a five-year project (2004–09) funded by the Canadian International Development Agency (CIDA) and implemented by the International Livestock Research Institute (ILRI) on behalf of the Ethiopian Ministry of Agriculture and Rural Development (MoARD). It has four main thrusts:
During the initial participatory rural appraisal (PRA), fish was identified as a commodity with market potential in one of IPMS’s eight pilot learning woredas (PLW), namely Fogera woreda on Lake Tana. ILRI invited WorldFish Center to provide inputs to IPMS in support of the fisheries-related work.
From a marketing perspective, two main fish products were noted, as well as local consumption of fresh fish in the fishing communities:
This study focuses on the frozen fish market chain for fish produced in Fogera woreda. In addition, in the margins of the main study, it was suggested that a preliminary assessment be made of the potential importance of the dried catfish trade.
The work was conducted in close collaboration with the Woreda Office for Agricultural and Rural Development and the Amhara Region Agricultural Research Institute (ARARI). The WorldFish Center economist sought to highlight key issues, help link WOARD and ARARI to other development partners, provide information and help build capacity on issues related to marketing systems analysis.
Following a short scoping visit in February 2006, the study involved an initial desk review of information available on fisheries and fish marketing in Ethiopia, followed by two weeks field work in June 2006 and subsequent analysis and write-up. The field work involved one day of preliminary discussions with market actors in Addis Ababa, followed by nine days on Lake Tana, visiting both fishing communities and marketing centres in the vicinity. The final few days were spent in Addis Ababa to follow-up with market actors in more detail. The study used the rapid market chain analysis methodology in which the various actors and their linkages in the chain are identified and their transactions analysed (Holtzman 2002).
The team comprised Sewmehon Tegegne Demissie (head of ARARI’s Fish Research Unit), Melaku Tadesse (officer in charge of fisheries, in Fogera WOARD) and Ann Gordon (WorldFish Center). Although working together for the most part, the initial meetings in Addis Ababa were conducted by Ann Gordon alone, whilst the market investigations in Addis Ababa at the end of the visit were shared among team members. The two Ethiopian team members also contributed a large body of information relating to fishing and fish marketing activities on Lake Tana.
Following Eritrea’s secession from Ethiopia in 1993 and the consequent loss of its coastline, Ethiopia has only inland freshwater capture fisheries. It has no significant aquaculture development. The inland capture fishery comprises: Rift Valley lakes (for example, lakes Chamo, Abaya and Ziway and the northern part of Lake Turkana) and Lake Tana, which although shallow, is the largest lake in Ethiopia; rivers; and small water bodies (reservoirs, natural ponds). There is fishing on all these water bodies, but commercial production (i.e. serving markets other than the local communities) is concentrated on the five lakes, with Chamo, Ziway and Tana particularly dominant (see Map 1).
Map 1. Important fish production areas in Ethiopia.
The main species are Nile tilapia, representing 60% of the catch, (Brueil 1995; Reyntjens and Wudneh 1998), Nile perch (favoured but increasingly scarce), barbus (two species) and catfish.
There are few reliable data on fisheries production in Ethiopia and the BoARD in Bahir Dar told the team that no data on landings had been collected for the last 5–6 years. Reyntjens and Wudneh (1998) estimated that total landings in 1996 were roughly 8500 t and rising rapidly. FAO (2003a) estimated that production of fish was 15,400 t in 2001.
There are seasonal variations in the availability of different types of fish, but as there are some differences between the lakes, traders can smooth out supply to some extent.
Fishing techniques are overwhelmingly artisan, with very few motorized boats (limited to a very small number on Lake Tana and on some of the southern lakes). The predominant boat is the reed (papyrus) tanqwa. It is even difficult to obtain certain materials for nets (lead rope and floats). Gill nets are the most common, but there is also some use of beach seines, cast nets and line-fishing (the latter for Nile perch) (Breuil 1995). Traps, scoop nets and baskets are also used, particularly in the rivers.
The general view seems to be that the lakes in the south are heavily exploited. For 8 important landing sites on lakes, FAO (2003b) estimated that offtake exceeds 60% of potential in 7 sites and 80% in 5 sites (including Awassa where offtake is estimated at 140% of potential).1 Only in Lake Tana is offtake estimated to be dramatically less than potential (15%). Breuil (1995) noted the small size of fish and declines in landings in lakes Awassa and Ziway. Anecdotal evidence also comes from fish traders in Addis Ababa, whose sources are largely but not exclusively the lakes in the south. The traders consistently claim that fish size is declining. One interviewee had noticed a marked difference in supply (difficulty in obtaining required volumes) in the last two years. This view, of heavy exploitation of resources in the southern lakes, is also endorsed by Dr Eshete (fisheries scientist at ARARI in Bahir Dar, personal communication).
1. These calculations are based on an estimate of maximum sustainable yield; thus offtake of 140% of poten-tial means that resources are being over-exploited and that yields at this level are unsustainable.
Domestic supply is only supplemented by imports to a very small extent. FAO (2003) indicated that 36 t of fish were imported in 2001. Information collected as part of this study suggested that this is likely to be mainly marine fish and high value fish (possibly including small amounts of Nile perch from other East African countries), destined for hotels or upmarket supermarkets.
Ethiopia’s population in 2004 was 70 million, with annual growth of 2.1%.2 Per capita gross national income was estimated at USD 110 in 2004. UNDP (2005) gave Ethiopia a very low human development index value (2003), with only 7 countries (out of 177) scoring lower. In 1999/2000, 44% of the population were estimated to live below the national poverty line, with the incidence of poverty higher in rural areas than in urban areas. Over the period 2000–2004, the annual average growth in GDP was 3.7%. The population is markedly rural. CSA (2001) indicated that less than 15% of the total population is urban (and, this must under-estimate the rural population share because it was based on a survey that included Addis Ababa but excluded 2 rural parts of the country). The population of Addis Ababa was 2.3 million in 1994 and is probably in excess of 3 million now. (Official estimates of around 3 million in 2006 are based on the population growth rate prior to the 1994 census. Many people think that Addis Ababa is now growing very rapidly and the official estimates may therefore underestimate its present size).
Ethiopians do not consume large quantities of fish, although there is no religious prohibition for the Christian and Moslem populations. Rather, this is a country with a strong tradition of livestock rearing and meat consumption. The Ethiopian Orthodox Church observes several fasting periods as well as fasting days every week, when meat is not consumed. Most Christians consider fish acceptable during those periods, though some strict followers will not eat any animal products.
These factors give rise to some particular characteristics of fish consumption in Ethiopia. Overall, per capita fish consumption is very low (perhaps as little as 200 g/year—based
2. The data presented in this section come from World Bank (2005) unless otherwise indicated. Note that some sources quote higher population growth of around 2.9%.
on the population and production figures provided above.)3 However, consumption is heavily biased towards quite limited geographical areas (production areas and Addis Ababa) and also heavily weighted towards fasting days (Wednesdays and Fridays) and fasting periods (55 days in March/April, 15 days in August, as well as other periods which may be less widely observed). Increasing scarcity (apparently reflecting both rising demand and supply constraints) has resulted in rising real prices for fish, so there is an increasing tendency for fish to be a luxury product consumed by higher income groups. Moreover, anecdotal evidence from interviews with traders and other observers suggests that higher income groups may represent a significant source of the increase in demand (reflecting wider exposure to different types of food and echoing the global shift in demand towards fish as a healthier source of animal protein), though population increase (particularly in growing Addis Ababa) and a modest general increase in incomes are also factors.
All three factors noted above will contribute to an increase in demand, i.e.:
per annum.5 This equates to a 20% increase in demand over 5 years and a 44% increase over 10 years.
The assumed shift in preferences would deepen the effect. Furthermore, if the population of Addis Ababa (a major focus for fish consumption) is growing faster than the rest of the population and incomes there increasing faster too, there would be a still stronger increase in demand.3 Note that this is based on the landed weight of whole fish, so actual consumption (after filleting or remov-ing the head etc.) is even less.
4 Assuming that the fish consuming population is experiencing income growth at least equal to the average increase in per capita GDP.
5To put this estimate in context, Delgado et al. (2003) estimated that over the period 1997–2020, aggre-gate fish consumption in sub-Saharan Africa will increase by 2.4% per annum. Their estimate takes account of factors that will contribute to increased demand but also takes into consideration the effect of rising prices (because supply will not keep pace with demand), which will in turn limit the potential for increased consump-tion. The crude estimate provided above for Ethiopia is simply the implied increase in demand. This takes no account of prices changes that will affect consumption. A survey of the relevant literature by Asche and Bjorn-dal (1999), cited in Delgado et al. (2003), found that own price elasticities of demand for fish are relatively high (–0.8 to –1.5), implying that a 1% increase in the price of fish would reduce consumption by 0.8–1.5%.
Where supply cannot match these increases in demand, the real price of fish will rise. The 1999/2000 household consumption survey found that calorie intake from fish, although still a negligible source of calories in people’s diets on average (0.0% after rounding), increased by 50% in rural areas over the four-year period to 1999/2000 and doubled in urban areas (CSA 2001). The country average was 0.3 calories per person per day and consumption in urban areas was twice this. Those reported increases in consumption during the second half of the 1990s suggest (but do not confirm) that fish remained affordable and hence that supply was able to keep pace with demand. The alternative explanation is that rising incomes and a shift in preferences (generally or in specific groups) meant that despite rising prices, people still chose to increase their consumption of fish—or were driven towards fish because of even greater price increases in substitute foods (such as meat).
Anecdotal evidence obtained through discussions with traders suggests that the tightening in supply may well be a more recent phenomenon (over the last 2–5 years) and that there were indeed significant increases in catch during the 1990s (when landings were still being recorded, at least to some extent). All traders indicated that they could sell larger quantities of fish if it were available—though clearly only at the right price.
As the discussion of production indicates, tilapia is the dominant species caught and consumed in Ethiopia, although this does not hold for all groups and for all areas. In fish production areas, fish consumption patterns reflect the local availability of fish (with some notable exceptions, such as catfish in Fogera woreda, see Section 4). Fish Production and Marketing Enterprise (FPME) retail staff indicated that the range of fish available has increased over the last twenty years: previously the focus was exclusively Nile perch and tilapia.
Relative pricing gives some indication of preferences and/or abundance: Nile perch (where available) is most expensive, followed by tilapia, catfish and barbus. Tilapia retains its value despite relative abundance, indicating strong consumer demand.
Outside of the production areas, where much of the local trade is in fresh whole fish, there is a relatively strong preference for fillets and most frozen fish is traded as fillets. Barbus is the exception and is sold whole, gutted and skinned.
There is negligible use of ice (as indicated by traders and observed with only some trucks using ice to limit thawing of frozen fish being transported overnight from production areas to Addis Ababa). Ice is used on neither fishing boats nor collector vessels.
At landings, fish is traded fresh for rapid consumption or resale. However, at trader collection points, freezers are used and the frozen fish is then sent to Addis Ababa and local urban markets (even those situated on lakes). Thus, the fish products on sale in Addis Ababa (and in urban centres near the production areas) are frozen—and very often filleted. (Or, more accurately, the fish is semi-frozen on arrival in Addis Ababa, following transport in normal non-insulated trucks).
In addition, and although not a focus for this study, it should be noted that there is a tradition of dried fish consumption (particularly catfish) in some parts of Ethiopia, notably Tigray. The product range of the formerly state-run FPME, which is the largest single trader of fish in Ethiopia, includes smoked fish and dried fish. Although there is presently a shortage of Nile perch (sourced from Lake Turkana), the FPME indicates that sun-dried Nile perch is usually its most important product line, with a good market in Addis Ababa.
The FPME, the former parastatal, is the largest single trader of fish products in Ethiopia, focusing overwhelmingly on the supply of frozen fish and dried Nile perch to domestic markets (with tiny quantities exported to Sudan and Saudi Arabia). FPME argues that its operations have been subject to competition from other traders for the last 12 years and that it does not operate under any privileged financial arrangements. However, the government is able to facilitate access to credit for FPME, which may confer some advantage over its competitors. Moreover, it is not clear what costs FPME incurred for existing assets; it seems likely that these represented a significant subsidy and advantage to FPME when fish marketing was liberalized. FPME has considerable plant and premises, including: freezing facilities and cold storage at production points, cold storage in Addis Ababa, refrigerated transport and other vehicles, and 17 retail outlets all with multiple freezers.
In general, FPME buys fish via contractual arrangements with fishing co-operatives operating on lakes Tana, Ziway and Langano, and in Arbaminch (with its two lakes). Fish is gutted and filleted at the landing site and quickly frozen. The cold chain remains intact throughout transportation and retailing, which is unlike the operations of any other trader. Fish is sold through FPME’s own retail outlets, whose spread again reflects the general pattern of fish consumption concentrated in the capital and in production areas: 13 in Addis Ababa and one each in Debre Zeit, Gondar, Bahir Dar and Ziway. FPME also sells fish directly to some institutional buyers, including large hotels in Addis Ababa. It currently handles about 1200 t/year (weight of wet whole fish).
Although FPME was the dominant trader for many years and remains the single largest trader, it now handles only a small share of production (8% or less). Cumulatively, the volumes handled by the other smaller traders exceed those handled by FPME. However, FPME’s cold chain and the general standard of handling and hygiene (from landing to retail) set it apart from the other traders.
There are a number of other fish wholesalers operating in Addis Ababa, usually with links to a specific area of supply. Some of these operate all year round, whilst others have more seasonal operations (particularly during fasting periods when demand is high). Many have been operating for a long time—some interviewees had traded fish for 30 years.6 In the Piazza market area of Addis Ababa there are around 8–10 such traders and the team interviewed 2 others with fairly well-established operations in other parts of Addis Ababa.
Most of the traders in Piazza have small lock-ups, with 2–6 chest freezers. They trade roughly from 0600–1800 hours, but the morning is busiest. They receive fresh or frozen fish (the former from lakes close to Addis Ababa, such as Ziway, which is about 2.5 hours drive), sometimes transported with ice, usually arriving in Addis Ababa early morning. They have contractual arrangements with agents or fishing groups (i.e. co-operatives) in the production area and rent space in regular trucks (with fish being just one part of a larger load). Although some traders may buy fish that has been filleted at the landing, much of this fish is filleted on arrival in Addis Ababa for direct sale to consumers or smaller traders (for instance, ambulant traders operating in Mercato), or filleted and then frozen. These traders also sell to small hotels and supermarkets.
The two larger traders interviewed by the team did not have retail outlets but had larger stores and conducted their trade by telephone, selling to larger buyers (including large hotels)—to whom the fish is delivered or who collect with their own transport. These two traders have significantly more freezer capacity (three tonnes or more) but do not have refrigerated transport.
Unsurprisingly some traders and the chef of a large hotel in Addis Ababa expressed considerable concern about the poor quality of the fish available in Addis Ababa—saying that it was often a reason to reject a shipment or to ‘blacklist’ certain agents sending fish from the production areas. The principal indicator of poor quality was the smell. The key issue here seems to be the absence of a cold chain for all fish traded except that handled by FPME.
6. As far as the authors could ascertain, FPME operated as a parastatal until 12 years ago, when it stopped receiving direct financial support from government. Other traders were able to operate, though clearly FPME had a significant advantage.
On the basis of the work described here, it is very difficult to arrive at an accurate estimate of market size in Addis Ababa—principally because there is clearly so much day-to-day and seasonal variation, handled by many different agents. Moreover, the team did not visit other production areas, which are important supply sources for the capital. Nevertheless it is possible to at least have a rough idea of its magnitude.
Addis Ababa has 13 of FPME’s 18 outlets, representing annual sales of roughly 800 t (weight of whole wet fish). The two largest hotels in Addis Ababa probably use around 50 t (filleted weight) per annum (or roughly 125 t of whole wet fish). There are at least 10 specialized (non-FPME) frozen fish retail outlets, with daily capacity in the range 0.5–1 t (filleted weight). If one assumes that average daily sales are 0.5 t, with trade 6 days per week, then these outlets represent sales of around 1560 t/annum (filleted weight) or roughly 4000 t (weight of whole wet fish). These figures, that do not include all sources of fish in Addis Ababa, amount to aggregate sales of around 5000 t/annum (weight of whole wet fish), or roughly 1/3 of estimated national supply (2001).
This estimate is clearly very crude but not inconsistent with observations around Lake Tana, where more than half the visibly traded (frozen) fish seems to go to local urban markets (Gondar and Bahir Dar) and the remainder to Addis Ababa. In addition, a significant proportion of the catch is consumed within fishing communities or traded as fresh fish in the immediate vicinity. Thus, it would seem plausible to suggest that for Lake Tana perhaps 20–25% of the catch is destined for Addis Ababa.
Historically, Ethiopia’s economy was heavily controlled by the government with parastatals and government-supported local-level co-operatives. In the early 1990s there was progressive liberalization of markets such that government institutions no longer play a significant role in direct marketing of rural produce, with this role now left to private traders. However, as in many other African countries, there is growing recognition that these liberalized markets are not delivering the development required to lift rural areas out of poverty and largely subsistence activity. Recent debate has focused on how to address problems of market fragmentation, weak demand, high transaction costs and weak market actors.
New initiatives and policy directions in Ethiopia include:
These general shifts in government thinking and emphasis constitute the policy context within which the fisheries sub-sector operates. Fish marketing issues have not been singled out for special treatment within this broader policy debate.
Notably, FPME is one example of a parastatal that continues to operate, albeit now subject to competition from other traders and apparently without any significant public subsidy. There is periodic talk of privatizing FPME but no public decision on this to date. Whilst there may be elements of unfair advantage enjoyed by FPME over other market actors (for instance, easier access to credit and substantial assets, in the form of premises, plant, and vehicles, previously acquired from government or donor projects), it may actually operate at a disadvantage in some other respects (it may be less easy to reduce staff numbers or even dispose of some assets, thereby reducing its flexibility and adding to running costs).7
Many co-operatives remain—some of them little more than producer groups not currently engaged in any co-operative enterprise, some still heavily-supported by government or by projects, but others (such as Tana Haik I Fishing Co-operative) appear to be well-managed as a producer co-operative enterprise. The latter has sought to incorporate smaller producer groups operating at kebele-level, in a bid to improve efficiency and realize economies of scale.
The kebele is the lowest-level government-determined administrative unit. A kebele comprises, on average, between 4–5 villages. Elected kebele-level positions are financially supported by government and these provide an important link to woreda-level local government. Within each kebele there may be sub-groups of producers, e.g. fishers (sometimes called co-operatives). These groups may incorporate producers from more than one kebele.
7. In addition, FPME argues that it takes care to source fish from reputable suppliers, observing recommended fishing practices (notably relating to mesh size). It alleges that its competitors are not so vigilant in this regard.
Central government ministries are represented at regional-level and at woreda-level. Thus, in Bahir Dar (the regional capital), the MoARD is represented by the regional Bureau of Agricultural and Rural Development (BoARD) and in Wereta (the administrative capital of Fogera woreda) by the Woreda Office for Agricultural and Rural Development (WOARD).
The BoARD has officers responsible for fisheries (statistics and planning) and the WOARD in Fogera also has an officer responsible for fisheries data collection and extension.
A National Fisheries Proclamation was ratified by Parliament in 2003. It provides broad guidelines relating to resource conservation, food safety and aquaculture. This document puts considerable emphasis on regulation, permits and the role of the fishery inspector. It is intended that the regional administrations should then use this as the broad framework within which their own proclamations are developed.
Although not all of the regional proclamations have been finalized, the proclamation for Amhara Region was developed in 2003. It covers the same areas as the national policy, but has an additional objective relating to the creation of employment opportunities in fishing communities. It also states that information, including research findings, should be made available to the fishing communities. As with the National Proclamation it relies heavily on regulatory measures (‘command and control’) and the role of the fishery inspector. There is no mention of co-management, though one of the stated objectives (‘…to prevent and control over-exploitation of the fisheries resource’) would seem to leave open this option.
At both national and regional level, the proclamations reflect concern that fish products should conform with prescribed standards, although the latter had yet to be developed when the proclamations were drafted. At national-level, guidelines have now been developed and were with the Council of Ministers when field work for this study took place in June 2006. Pending official endorsement, these are not yet in the public domain.8 It is noted that although such standards exist in some other sectors, they are not widely used or enforced (Woldu 2005).
In the past, the fisheries sub-sector benefited from a number of donor projects.
8. It is understood that the draft Fish Quality Assurance Regulation meets EU export requirements. Based on the Kenyan model (which supports a significant export trade in Nile perch from Lake Victoria), one of its require-ments is the establishment of a Fisheries Authority.
NGOs (e.g. the Interchurch Foundation Ethiopia and the Interchurch Organization for Development Co-operation) have been active throughout Ethiopia. Their most visible impacts are seen in the southern part of Lake Tana, where their activities led to the introduction of motorized vessels and improvements in the physical landing/marketing infrastructure, now used by Tana Haik I Fishing Co-operative and FPME. Wageningen Agricultural University was involved in fisheries research and the European Commission (EC) was the main donor. The 1st phase of the EC-funded Lake Fisheries Development Project started in 1982 (focusing on improved fishing technologies, resource surveys on lakes Ziway and Abaya, and upgrading FPME’s processing and marketing capacities). The 2nd phase started in 1992 and changed emphasis to reflect the shift to a market economy. It focused on the Rift Valley lakes and Lake Tana and helped put in place routine data collection and analysis systems9 (Reyntjens et al. 1998).
The only major fisheries development project now on the horizon is supported by the African Development Bank. Although the planning is still at an early stage, it proposes a USD 1.4 million project to improve fish exploitation, focusing on the preparation of a fisheries development plan and activities to support its implementation. It is anticipated that this will commence in the next fiscal year (Head of Livestock Division, MoARD, Addis Ababa, personal communication).
9. These systems were subsequently abandoned around 2000 (BoARD, Bahir Dar).
Located in Amhara Region, Lake Tana is Ethiopia’s largest lake and the source of the Blue Nile. It is a shallow lake, not exceeding 14 metres in depth. With recent improvements in the road system, Bahir Dar, the main town on the lake in the extreme south, can be reached by tarmac road from Addis Ababa in about 9 hours.10 From Bahir Dar, there is a tarmac road to Gondar, running north to the east of the lake, 3–7 kilometres inland, making this side of the lake relatively accessible. (From this road there is also one tarmac spur to the lake, from Meksenyt). The rest of the lake is served by dirt roads (many of which are not passable by vehicles during the rainy season), a ferry from Bahir Dar serving the western and northern shore villages, and informal boat transport (see Map 2).
Bahir Dar has a population of about 110 thousand and is currently expanding rapidly, with much new construction. It has good services and is both a regional (administrative) capital and a commercial centre—drawing in migrants from the surrounding rural areas. It attracts both Ethiopian and foreign tourists because of its attractive location on the lake and because the area has many famous monasteries. It has many hotels and restaurants (including new development) and is an important market for goods produced locally and items shipped in from Addis Ababa.
The only other significant town in the vicinity of Lake Tana is Gondar—an important transport and tourist centre, roughly 40 kilometres north of the lake and 3 hours drive from Bahir Dar. Gorgora is a small town on the northern shore, with no tarmac road and no mains electricity. The settlements in the west are even less accessible.
Fogera woreda is located 55 kilometres to the north of Bahir Dar on the eastern shore of the lake. The woreda has two small towns: Wereta and Alem Ber. Wereta is the woreda capital. Located on the main road, it has good services (transport, electricity, telephone) and government offices. To its west, the woreda comprises a plain that floods from the lake during the rains, making it an important and extensive wetland area. To the east, the land gradually rises to 2400 metres (650 metres higher than Lake Tana). The woreda is drained (into Lake Tana) by two main rivers, one on its northern border with the next woreda (Rib River) and the other in the south (the River Goumera). The population of the woreda was 234 thousand (IPMS 2004).
10. Only the Nile Gorge section of the road is not tarmac.
Map 2. Lake Tana and surrounding area.
Away from the main road, there is no mains electricity and virtually no telephone network (although mobile phones work in some areas and settlements may have one land-line telephone in a government office). The dirt roads are poor and many areas can only be reached on-foot or by donkey for several months in the year, following the rains. Ownership of vehicles is very limited in these areas. Some villages have diesel-powered mills but generally very little use is made of power generators. Lake transport is also very slow and uncertain (transfer from Fogera woreda to Bahir Dar could take 5 hours, not including the time needed to walk to the lake shore, negotiate a ride to a point where it might be possible to find transport to Bahir Dar, and waiting time). The two villages near the lake are Nabaga and Wagatera.
At woreda-level, extension officers are non-specialists (often with a diploma in agriculture) and positions covering different sub-sectors are rotated every few years.11
In Fogera, the person covering fisheries is responsible for: data collection, liaison with the fisher groups at kebele-level, making sure that fishing regulations are observed, and trying to improve market prospects for fishers (including gathering information on products and technologies, trying to introduce buyers to the fishers’ groups or otherwise raise trader awareness of the products available). Such a role in marketing aspects is very progressive and (more broadly) rather unusual in agricultural (fisheries) extension (partly as a result of the IPMS project focus on redefining the role of the extension staff). Unfortunately staff do not receive training in this aspect of their job. Nevertheless, some of these officers are very knowledgeable about the market and how it functions. These officers may also have some responsibilities for other sub-sectors too. (For instance, the officer in charge of fisheries in Fogera covers apiculture too).
Although the extension services in Fogera are constrained in their capacity to reach out to farmers (in terms of staff numbers, staff capacities and resources for field work), within that broad context, fisheries extension is relatively well-covered in the woreda. There is only one officer for the sub-sector, but he does have access to transport to make field and market visits (though this may be atypical and relate to the IPMS project). Moreover, that part of the fisheries sub-sector that is legal and commercial in Fogera is relatively small— relating only to a few lake fishermen and the Nabaga seasonal ponds (involving many fishers but in a concentrated area).
However, there are clearly capacity constraints relating to data collection on Lake Tana fishing activity more broadly and negligible experience with participatory methods to
11. Ashagri, former officer in charge of fisheries, WOARD, Fogera woreda (personal communication).
improve fishing practices (notably relating to co-management and reducing the use of illegal gear). At present the dual roles of friend (extension agent and market contact facilitator) and policeman (enforcer of regulations) reside rather uncomfortably with the same woreda-level fisheries officer. The officers concerned also receive little or no training in their market development facilitating role—which is clearly an important area where there is a need to build stronger capacities.
The annual catch from Lake Tana is estimated at around 1000 t. The last year for which the BoARD collected landings data was 1996; recorded landings were in the range 470– 1470 t in the preceding 4 years. Depending on the season, this catch comprises varying volumes and proportions of tilapia, barbus and catfish.
March–August: highest catch levels from Lake Tana November–May: harvest period from seasonal ponds
January–July (and especially Mar/April/May)—tilapia important June–September (especially June/July)—catfish important June–September (especially August)—barbus important
Closely associated with the lake fishery are: (a) the natural seasonal ponds (mostly on the eastern shore) that are linked to the lake and flood during the rains—producing tilapia or catfish, and (b) the rivers which are important spawning grounds for lake barbus (particularly three rivers that drain into the lake, the Goumera, Magetchi and Derma).
The vast majority of fishermen use tanqwa—traditional reed boats (which numbered nearly 400 in 1998, according to Tegegne (2003), and numbers are likely to have increased since). For the whole lake, there are fewer than 25 motorized fishing boats, most of which land their catch in Bahir Dar (either directly or via collector boats). Traditionally, the lake was only fished by the indigenous tribe, but recently, with improved marketing prospects, fishing has been taken up on a part-time basis by the long-established agricultural migrant community (i.e. migrants who came to the area more than 100 years ago and now represent the dominant group).
Inaccessibility is a significant constraint to marketing and hence acts as a brake on fishing activity. Thus, the resources in the south of the lake within reach of the Bahir Dar market, are allegedly heavily exploited. The eastern part of the lake is now the focus of attention—with potential to sell fish in Bahir Dar (via motorized collector vessels) or to carry fish on-foot or by donkey to the main road, where traders have become established with the completion of the tarmac road and the arrival of mains electricity. There is also fishing activity in other parts of the lake but most of this is subsistence in nature, or relatively small-scale because the prospects for marketing are so limited. However, some of these areas have recently started producing dried fish.
The fishermen are organized to some extent within their kebeles. This local administration has a defined structure, with officers responsible for different sectors (health, education etc.). Leaders are elected but then paid by government. The kebele serves in part as a link between the woreda-level administration and village affairs. Community-level decisions are taken by the kebele administrators in consultation with community members. Within this structure, a fishers’ group was identified in Wagatera and in Nabaga. Those interested in lake fishing were encouraged to become members of the Tana Haik I Fishing Co-operative based in Bahir Dar (see below). Those fishermen either rent boats from the Co-operative or are waiting until more boats become available.
Traditionally the role of women in capture fisheries in Ethiopia is mainly confined to net-making (and post-harvest activities described in the following section).
In Fogera woreda, fishing activity on the lake comprises:
catch of the Co-operative’s motorized boats is about 25–30 t of fish per vessel. The six fishermen from Fogera woreda are supposed to sell their catch to the Tana Haik collector boat that makes a daily trip from Bahir Dar to an island just south of Fogera (Rema Medhane Alem). There are two or three other collector vessels that sometimes collect fish from the same area (representing the FPME, the Zege Fish-for-All group and Georgis, a Bahir Dar trader—see discussion of constraints below).
The woreda has two large seasonal ponds that are connected to the lake and flood during the rainy season. As the waters recede, these become extremely productive ponds for catfish which are caught in the period from February–May with gill nets and spears by large numbers of local farmers turned seasonal fishermen (perhaps around 600). The catch is dried (women help with this aspect) and sold to traders. This fishery is easily the most important in the woreda in terms of revenues and employment. (In the woreda to the south, Hamusit, there is a similar natural pond that produces wild tilapia, which is sold fresh.)
It should be noted that until three years ago, the Fogera ponds were not fished at all. Catfish is not eaten by the local population and the fish were apparently left to die. However, a combination of factors (including improvements in the roads to Sudan) has led to greater trader interest in dried catfish. Starting in 2004, a trader leased the two ponds from the kebele12 (he provided funds for the community to construct a church) gaining exclusive fixed price purchase rights to the fish, which was dried by the local community. The following year, the lease was renewed, in return for a payment of ETB 5000, and in 2006, the payment increased to ETB 19 thousand (roughly USD 2200).
The rivers, particularly Goumera River to the south, are important spawning grounds for lake barbus, but also a source of tilapia and catfish. Around 400 people are involved in the fishery on a part-time or seasonal basis, with fishing (using traps) particularly pronounced during the official closed season for barbus (August–October).13 Although widely traded and consumed in Ethiopia, barbus is the least preferred of the main fish species available in Ethiopia, so the catch (which is sold fresh) attracts relatively low prices from traders and local hotels and restaurants. However, the volumes are significant, so fishermen find it worthwhile despite the poor prices. Fishermen complain
that their efforts to realize better prices are hindered by the illegal nature of the fishing and the poor bargaining position this confers. Certainly lake barbus sold to traders in Bahir Dar commands higher prices than that caught in the rivers. However, the low prices received for river barbus almost certainly also reflect the very high volumes caught during the peak season and the limited size of the local rural markets.
The spawning season, when there is most fishing activity, coincides with the lean season, so the fishery represents a seasonally important source of sustenance and income for some households. Fisher–farmers interviewed during the course of this study, who concentrate on this activity during a four-month period, estimate that it represents about one-seventh of their annual income.
The team did not investigate this fishery further, because of its illegality. However, it should be emphasized that the fishery is important to large numbers of households and particularly so during the most difficult part of the year (i.e. the lean season).
There is no firm information on the landings from these different fisheries at the present time. Melaku Tadesse (co-author and officer responsible for fisheries within the Fogera WOARD) estimates their order of importance, in terms of employment and marketing revenues, to be:
In common with small-scale fisheries worldwide, many fishermen in Ethiopia lack the resources to buy fishing gear. It is an irony of fisheries that although a ‘commons’ (and hence often an important source of livelihood for the poor, who lack land or other resources), most fishing requires some financial investment. The result is a heavy concentration on gear that is low-cost (tanqwa) or requires only labour and local materials (traps). Traditional tanqwa, once made all around Lake Tana, or the papyrus, are now purchased in Bahir Dar, where the papyrus is still available. Depending on the weather (humidity), a tanqwa lasts 15–60 days and costs USD 2–4.
Fishermen reduce the cost of nets by making their own, but still many fishermen do not have their own nets. This limits fishing activity or makes fishermen reliant on gear loaned by traders, locking them into sales agreements, usually on poorer terms.
Tana Haik I Fishing Co-operative rents out motorized boats to its members on a full cost recovery basis. (The Co-operative receives no outside support.) It too lacks capital to purchase sufficient boats to meet the demand from fishermen (41 or roughly one-third of its membership are still waiting for boats).
There are two sources of credit in Fogera woreda:
Moreover, interviewees in Enfranz indicated that there were no input–credit–sales arrangements between fishermen and traders in operation there.
Motors for boats, whether in-board or outboard, are not easily available in Ethiopia. It is also difficult to obtain spare parts. The General Manager of Tana Haik I Fishing Co-operative described how he recently purchased five second-hand engines, in need of repair, from another co-operative on a lake in southern Ethiopia. It seems that earlier development projects were the source of some of the motorized boats (8 of Tana Haik I’s 13 boats were ‘inherited’ from an earlier project in which it participated). However, the lack of availability of motors must also reflect lack of effective demand (i.e. demand backed by purchasing power). If there were higher demand, presumably an importer would seize this trading opportunity. Presumably too, the co-operative from which the Tana Haik I second-hand motors were purchased would have been less willing to see them idle for want of repair and to sell them on.
Floats and lead rope used with nets are also difficult to obtain in Ethiopia.
Current instability in the arrangements for fishing and collection suggests that margins were very low at the time of field work in June 2006—apparently brought to near-zero with recent increases in fuel costs, increased numbers of motorized fishing vessels and a seasonal reduction in catch. A number of changes were taking place: (a) fishermen were being offered lower prices, (b) established trading partnerships were breaking up and new ones emerging, (c) traders were less willing to send collector vessels to those parts of the eastern lake where the motorized fishing boats operate, claiming that the costs were prohibitive given the small catch size, and (d) fishermen were responding by reducing the number of days they fish per week.
Some of this may have been triggered by the need to offload the additional catch of a new entrant onto a marketing system that lacks capacity (at least seasonally) to rapidly absorb a larger volume of fish. (The Zege Fish-for-All fishing group, with five motorized vessels which were provided as a grant to needy families, started operations two years ago but production has been gradually increasing.) Certainly, the motorized vessel fishers seek to maximize fishing time and minimize costs, by crews changing in situ without the boats returning to base, by the use of collector vessels that collect from several boats and by the use of second-hand motors. (Tana Haik I Fishing Co-operative has 12 motorized fishing boats, served by one collector vessel.)
The cost advantage of the fishermen working on the lakes in the south appears to stem in part from the shorter distance to Addis Ababa (at least for Lake Ziway which is 2.5 hours drive from the capital, but less so for the other lakes) and also from the fact that there is much better road access to some of those lakes, than there is to Lake Tana. Lake Tana appears to be disadvantaged by the fact that there are very few points where there is year-round access to the lake shore.
The relationship between these costs and marketing is further explored in Section 4.
The major marketing constraints faced by the fishermen are: (a) physical access to landing points, collector boat collection points, and road-side traders, (b) prices that are insufficiently remunerative to fishers (partially reflecting the high costs of landing the catch), and (c) loss of quality because of limited options for conservation and time/distance from trading points. These will be explored in the next section, but are mentioned here because they clearly act as a brake on fishing activity (evidenced by rapid increase in fishing effort when these constraints are removed).14
14. The team interviewed traders and fishermen in the small town of Enfranz. Enfranz is located on the tarmac road, to the north of Fogera woreda, at a point just two hours walk from the lake. With gradual improvements in the tarmac road and the advent of mains electricity, Enfranz has become an important fish trading centre in the last four years. There are presently nine traders located there, each with freezer capacity, sending fish principally to Bahir Dar and Addis Ababa. Fish are head-loaded or carried by donkey from the lake to the traders. Until 3–4 years ago, only the traditional indigenous minority community fished, but now the larger lakeside community is involved (more than 100 fishermen), using tanqwas, with numbers still rising. All these fishermen reportedly have their own gear (tanqwas and nets). One collector boat operates in this part of the lake but the fishermen prefer to sell to the road-side traders who offer higher prices. Each trader has ‘scouts’ who will lead fishermen to them, but otherwise there is not much evidence of price competition or other strategies to assure supply. At least one trader has operations in other road-side locations east of the lake too (Yifag, Hamusit, Woreta).
This section focuses on lake fish that enters the frozen fish marketing system, destined principally for Bahir Dar and Addis Ababa. It does not explore the local village and woreda-level trade, because this trade takes place relatively unhindered (being close to the landing) and any scope it offers for growth can be met relatively easily. Nor does it explore the river barbus trade, principally because of the illegal nature of the fishing activity. The section concludes with some summary observations on the recent emergence of a significant trade in dried catfish. Indicative snapshot information on prices and selected costs is attached at Annex 1. The questions used to guide interviews with market participants are attached at Annex 2. Box 1 below summarizes the trading options and marketing systems for different fishers.
Box 1: Overview of marketing systems for Fogera woreda lake fishers
Motorized boat fishers (6 fishermen, 1 boat)/Fogera woreda
→ sell to Tana Haik I Fishing Co-operative (via collector boat)
Traders send fish to traders in Addis Ababa negotiating directly with the latter—there are no extra points in the marketing system between the two towns.
All fishermen retain fish for home consumption and informal local sales.
Fish are caught using tanqwa and motorized vessels. No use is made of ice on-board. Most of the catch is caught overnight/early morning and transferred to Bahir Dar via collector vessels that purchase fish around 0700–0800 hours each day, from points just south of Fogera woreda, returning to Bahir Dar in time to land the fish at around midday. It seems that occasionally one trader purchases fish from the lakeshore (principally from tanqwa fishermen), taking it to Hamusit by car (where he has freezers).
The following collector vessels operate in this part of the lake:
operating from Nabaga complain about the distance they must go every day to sell their catch (depending on conditions this takes 1.5–2 hours each way).
As far as the team is aware, the only other collector vessel operating on Lake Tana belongs to Stefanos, a well-known trader active in the northern half of the lake.
The collector vessels land their catch at Bahir Dar (with the only exception being an allegedly temporary arrangement that came into effect in June 2006, whereby the Zege association will land its catch at Zege—a small village, located to the northwest of Bahir Dar, less than one hour by road).
The fish landed by Tana Haik I Fishing Co-operative is filleted and frozen as soon as it is landed, with most of the catch generally frozen by mid-afternoon, if not earlier. A superficial assessment of handling conditions, gained through observation on several occasions, suggests that hygiene standards are quite good: fish is immediately transferred from the boat to the adjacent packing shed; the shed is continually washed out; the women packers wear protective clothing; appropriate packaging is used; and the fish is frozen as soon as it has been filleted and packaged. Around 65% of the Co-operative’s annual catch of roughly 300 t (weight of whole wet fish) is usually sold to FPME (which has cold storage facilities for 15 t at the Co-operative landing),15 so although FPME’s own vessel landings were not observed, the operation should be similar. The Co-operative itself also owns chest freezers, housed in its buildings at the landing.16
The Zege Fish-for-All association does not have freezing capacity, so it sells its catch to Georgis or FPME (whoever is offering the higher price). Georgis has chest freezers in Bahir Dar and again, on landing, fish is filleted and frozen. Whilst fishermen may opt to sell their catch to the Co-operative’s collector vessel at Rema Medhane Alem (from where it reliably collects four days/week), the Co-operative does not buy fresh fish at its Bahir Dar landing (apparently to assure the quality of the fish it handles).
There are two other ways in which fish enters the Bahir Dar marketing system. Although not investigated as part of this study, some fish is landed by local tanqwa—for home consumption and local sale. More significantly, note the emergent importance of Enfranz as a landing point, serving the Bahir Dar and Addis Ababa trade (see discussion in the previous section).
The main supplier of fish from Lake Tana to Addis Ababa is the FPME. About 30% of the fish handled by FPME nationally is sourced in Lake Tana. All of its barbus and most of its tilapia and catfish comes from Lake Tana. This represents about 300–400 t/annum (weight of whole wet fish). The Tana Haik I Fishing Co-operative is FPME’s main source of supply on Lake Tana.
Usually, FPME buys all the Co-operative’s catch three days per week. The frozen fillets are then held in cold storage in Bahir Dar, or sent to its headquarters in Addis Ababa (depending on the market situation), or sold through their market outlets in Bahir Dar
(2) or Gondar (1). Fish sent to Addis Ababa (90% of the fish purchased by the Bahir Dar
branch) will be subsequently sold through its 13 retail outlets there (and in Debre Zeit), or to institutional buyers. The Ethiopian market has been growing steadily, but the volume of catch handled by FPME has declined and market share has fallen considerably (to now roughly 8% of fish entering the marketing system nationally).17
FPME has refrigerated transport used to take fish from Bahir Dar to Addis Ababa (and other routes)—a round-trip operation that takes an astonishing 4–5 days (travelling empty from Addis Ababa, loading in Bahir Dar, returning and unloading), although a car can cover this route in 10 hours (one way).
Despite its superior facilities, the FPME has lost market share to smaller-scale more flexible operators, albeit unable to operate a cold chain. Although exposed to competition, this does not seem to have translated into effective strategies to win back market share and reduce costs (possibly reflecting a feeling within the organization that the government would not let it fail).18 Its retail outlets operate from unmarked premises, with no attempt at advertising. Retail pricing seems to be inflexible, irrespective of market conditions and inability to turn over stock in locations where there are other retailers selling at lower prices. Its transport operations appear to be high-cost (see above). It is not even able to meet the orders of large institutional buyers, which would normally represent prized ‘bread-and-butter’ customers (regular high volume buyers, associated with significantly lower transaction costs for the vendor, once the initial contracts have been set-up). There is no indication that the organization has adapted its operations to the changing marketing environment or has introduced new products or reviewed and made changes in its retail outlet network.
During June 2006, FPME could not reach agreement with the Tana Haik I Fishing Co-operative on prices, leading to a (most likely temporary) break-down in their long-established trading relationship. The FPME seemed confident of its position, in the knowledge that at that point it had plentiful stocks of frozen fish. It failed to reach agreement with the Zege Fish-for-All association too, its other main source of Lake Tana fish. This may, of course, be posturing to limit upward pressure on prices, but given that few fishers can land fish of the freshness required by FPME, and Lake Tana is its most important supply source, this would seem to be a risky strategy—particularly as FPME continues to lose market share. The manager in Bahir Dar indicated that if need be FPME would operate its own fishing vessels on Lake Tana.
The rest of the fish traded from Lake Tana to Addis Ababa is a much more ad hoc trade, though cumulatively the volumes are quite significant.
The team identified the following traders:
• one trader currently sending around 4 t (frozen filleted weight)/week to Addis Ababa, including recent purchases of fish from Tana Haik I Fishing Co-operative (so these volumes are atypically high); he has chest freezers in Bahir Dar (at the landing) and cold storage in Addis Ababa; however, he sends frozen fish fillets overnight from Bahir Dar by normal truck (with no ice); he sells to 15 hotels and 5 supermarkets in Addis Ababa; he has no retail outlet;
•one trader operating in the southeastern part of the lake, with freezer capacity in three road-side locations (including Woreta in Fogera woreda), currently sending about 0.5 t/week of frozen fillets to Addis Ababa;
• traders located in Enfranz on the eastern side of the lake, but further north; one trader is particularly dominant (5 of the 9 operations there are actually his agents); it is difficult to estimate volumes but their combined chest freezer capacity probably exceeds 5 t and can be filled in 1–5 days, suggesting weekly turnover of at least 7 t; frozen fish fillets are sent from here to Addis Ababa, Bahir Dar and to Gondar (though the latter is a less important destination) using normal trucks with no ice; the Addis Ababa share of this trade must be at least 50%.
Ignoring the first of these traders (whose operation has essentially taken over the trade usually passed through FPME), it is clear that the smaller traders are cumulatively responsible for at least as much frozen fish traded from Lake Tana to Addis Ababa as FPME. A very crude estimate of total weekly volumes (i.e. all three sources identified above) suggests about 8 t of frozen fillets (roughly 20 t of whole wet fish),19 with considerable seasonal variation in this. If those volumes were maintained for, say, two-thirds of the year, this would amount to an annual trade of 680 t fish (weight of whole wet fish). This suggests that, although important to FPME, Lake Tana fish still remains relatively insignificant in overall market volumes in the capital, and indeed this was the impression gained from discussions with individual traders in Addis Ababa.
On the market in Addis Ababa, there are certain perceptions amongst traders about Lake Tana fish:
19. The filleting percentage obtained by FPME is 45, 34 and 41% for catfish, tilapia and barbus, respectively.
Whilst some of the Enfranz traders have regular customers (traders/retail outlets, hotels, restaurants) to whom they sell in Addis Ababa, much of this trade takes place on a day-to-day basis, with trader decisions reflecting day-to-day market conditions.
The main suppliers of the Bahir Dar market are:
This catch is sold principally to local restaurants and hotels (including those in the small towns north of Bahir Dar), with Bahir Dar and Gondar the largest local markets. Bahir Dar and Gondar also have specialized retail outlets (both have at least four). Fish sold to retail outlets and to hotels/restaurants is usually filleted and frozen.
The urban markets in Gondar and Bahir Dar are both growing. Both are important and expanding centres for tourism and Gondar has recently seen the opening of a number of new retail outlets. The volume entering these markets from Lake Tana exceeds that sent to Addis Ababa, by at least a factor of 2 (not counting fish caught for home consumption or traded in the fishing village).
These inter-related points have been mentioned above so are just summarized here:
Improved organization amongst fishermen and traders, including the development of contracts, might lead to improved and smoother market development.
A number of market participants also indicate capital constraints, expressing interest in acquiring, e.g. large cold storage facilities and refrigerated transport. However, apparently low margins in the current trading system suggest that very careful analysis of the returns to such investments would be needed.
Women have a role in post-harvest handling of fish entering the frozen fish marketing system. In Bahir Dar, they are responsible for filleting and packing of fish traded by Tana Haik I Fishing Co-operative and FPME. When fishers sell their catch to road-side traders or traders buying fish in the fishing communities, women sometimes assist with filleting. In Enfranz, one of the road-side traders is a woman. A lot of the fish retail outlets visited in Bahir Dar, Gondar and Addis Ababa were staffed by women. Women are also involved in the preparation of fish meals—whether in the home, or in hotels and restaurants.
Although per capita consumption of fish is very low in Ethiopia, there is steady growth in demand reflecting population increase, rising incomes and a shift in preferences. The main areas of consumption are Addis Ababa and the populations and towns close to the main production areas. Fish consumption is strongly linked to the fasting traditions of the Ethiopian Orthodox church: most people consider that fish can be eaten on days when meat is not allowed (Wednesdays, Fridays and during the fasting months). Under relatively conservative assumptions, aggregate demand for fish could be expected to grow by around 44% over ten years, but there are some indications that demand is growing even faster than this.
Historically, Addis Ababa has been supplied from the lakes in the south of the country. There is, however, a widely-held view that these lakes are now over-fished and market participants report growing difficulty in sourcing fish over the last 2–5 years, suggesting that demand is now growing faster than supply.
Lake Tana is considered a rich source of fish, which generally does not yet show signs of over-fishing, but it is a relatively high cost source. With tightening in the market and rising prices, fish traders have started to take more interest in Lake Tana and a number of new small-scale operations have developed. High costs stem from: high landed costs because there are relatively few places where fish can be landed (where there is both access and amenities); the costs associated with shipping the long distance to Addis Ababa; and spoilage associated with the time lapse from capture to freezing and during transit to Addis Ababa. Thus, although the fishery is known for its larger size fish, it has a reputation for high prices and poor quality.
Current instability in the production and marketing system suggests that the system itself is relatively competitive, with affordability in the end-market (in particular, Addis Ababa) exerting pressure on prices.20 Certainly there are many buyers and many sellers (one condition for a competitive market) and although the plant owned by FPME should confer
20. FPME is the single largest supplier of frozen fish to the Addis Ababa market, though cumulatively the volumes handled by other traders exceed that handled by FPME. There is clearly relatively high concentration in this market (the four largest companies accounting for more than 40% of market share) though, as noted above, current instability suggests that the system behaves quite competitively.
advantage, in practice it appears to be a high cost operator, that is losing market share and slow to respond to changing market conditions.
The team was asked to briefly assess the importance of this activity. In Fogera woreda, the fishing and processing of dried catfish is significantly more important than either of the other two main fisheries: river barbus (which is an illegal but significant fishery) and the Lake Tana fishery. The seasonal dried catfish trade (ponds are fished February– May) employs as many as 600 fishers and their households (including women who are involved in the preparation and drying of fish). From a livelihoods perspective it is very important in that: it involves large numbers of people including women; it is a dry season activity; and can be undertaken in less accessible (and hence less advantaged) communities because the product has a long shelf-life. The trade, moreover, seems to be growing in importance:
visit, i.e. that further investigation of this fishery and associated trade may help identify important entry points for significant development of livelihoods in Fogera.
In this section, a number of recommendations are made on areas of intervention with scope to improve the opportunities generated by this commodity system, for the benefit of fishers, post-harvest workers (including women), traders and consumers. However, note that the system is relatively unimportant for the communities in Fogera woreda (six fishermen and perhaps a few additional tanqwa fishermen sometimes able to sell their catch to traders) and is likely to remain so pending any major improvements in accessibility. Even if some of the suggestions below improve the profitability of fishing activity in Fogera woreda, other parts of the lake are likely to retain a cost advantage. These recommendations would, however, benefit other fishing communities on Lake Tana and elsewhere in Ethiopia. (As noted above, both the seasonal ponds and the illegal river barbus fishery make a much more significant contribution to livelihoods in the woreda than the commercial lake fishery.)
No separate recommendation is made regarding opportunities to improve women’s livelihoods because this ‘lens’ should be part of the way in which each of the following recommendations is explored. Post-harvest activities are more likely to involve women than fisheries capture (at least where the latter involves boats), but nonetheless the way in which women are likely to be involved (or could be helped to benefit) should be given explicit attention in work focusing on the following areas.
Since Ethiopia does not have strong fishing and fish marketing traditions and FPME’s operations were always based on frozen fish, there seems to be very little interest in or knowledge of the use of ice and insulated storage to improve quality (on-board, upon landing, in transit, and at the retail outlet). Subject to the recommendations of a fish handling expert, it is suggested that trials be conducted with fishermen and traders to test the feasibility (technical and financial) and acceptability of the use of ice and insulated storage. There may also be scope for improvements via the use of low-cost evaporative cooling.
There are numerous benefits stemming from collective action in fisheries. These include: improved access to credit and inputs, scope for collective marketing and potential to exert more market power, shared access to other services too costly to supply/buy-in individually and improved opportunities for effective resource management. Tana Haik I Fishing Co-operative in Bahir Dar seems to be a very professionally-run producer co-operative, operating in the pure commercial sense of a members’ co-operative. It nonetheless struggles to identify alternatives when its main buyer switches sources. It has difficulty accessing credit and it has not been able to acquire sufficient motorized boats to meet the demand of its members. There could be considerable benefits from working with the Co-operative and other fisher groups, in a participatory way, to strengthen their capacity to identify solutions to the problems they face. Indicative areas of intervention could include:
developed for Lake Tana would also generate wider benefits for other fisheries and fisheries-dependent populations in Ethiopia.
Market information systems, particularly for agricultural commodities (broad sense) are not easy to implement effectively. Too often, the information is not current or does not include sufficient detail on product characteristics. This is particularly true of fish. In addition to the normal supply, demand and location factors that influence prices, fish prices vary depending on quality, size of fish (different size fish may have different per kg prices), how it has been prepared (filleted, gutted, headed, skinned etc.) and time of sale (fresh fish usually trades at a discount later in the day).
Yet there is certainly scope to improve information in the frozen fish marketing system. The Ethiopian market for fish is subject to a number of co-ordination problems associated with very marked shifts in demand (fasting periods and fasting days), multiple sources of fish each with different seasonal patterns of supply (affecting catch size and species), constraints on storage capacity (frozen fish), no wholesale market for fish, and many different actors at both production and marketing levels. Improved information could increase the net worth of the sub-sector via efficiency gains, increased prices and volumes, or reduced spoilage. During the field work in June 2006, there were simultaneous reports of insufficient supply in Addis Ababa and difficulty in marketing fish from Lake Tana.
It is suggested that the scope for improved information (type of information and modus operandi) be explored in close consultation with actors in the production and marketing system—keeping in view the need for systems that are relatively low cost, not overly complicated (and hence prone to failure), and preferably financed from within the industry.
The shifts in the market may present opportunities for niche marketing of higher value products including fish pickles and or lightly smoked fish. Through discussion with hoteliers, supermarkets and traders, potential products could be identified for test production and marketing. It is very important that such trials be conducted with the private sector, whose experience of the market will be critical. This is particularly important with product development, because it is quite difficult to persuade people to (rapidly) change their eating habits and this is not an area of work in which the public sector has a strong track record. Indeed, there have many ill-fated projects focused on the development of new products. In countries with a more developed food industry, such research would be considered too ‘near-market’ for public sector involvement.
Within Fogera woreda this is easily the most important source of fisheries-related livelihood. It generates the most aggregate income and involves large numbers of people, including women. The production of a dried product is also an appropriate adaptation to less advantaged, seasonally inaccessible communities. Preliminary observations and enquiries indicate that the dried fish trade is currently experiencing rapid growth in Ethiopia.
It is recommended that IPMS follow-up with a market study for dried fish, to explore in more detail the livelihood potential it offers. This should be done during the period February–May when the seasonal ponds are being fished.
This fishery is clearly important to large numbers of people in Fogera woreda and has particular significance to the most vulnerable households during the lean season. However, the illegal nature of this fishery has made it difficult for government officers to work with fishers to explore potential solutions to this dilemma.
It is suggested that this be revisited—initially in discussion with government and research officers—to examine the scientific parameters on which the ban is based. A closer examination is then needed of the fishing activity to yield information on timing, numbers involved, size and nature of fish catch, and its value. As the present fishery yields such low value fish, within the context of a participatory investigation, it may be possible to identify win–win solutions whereby both net worth and fish resources can be improved. Of course, there may not be such a fortuitous solution but certainly more information is needed before reaching that conclusion.
Implicit in these recommendations is an emphasis on participatory modes of working and building capacity in government institutions for improved understanding of markets. The potential identified by this study can only be realized if the interventions are developed and tested in close collaboration with fishers and traders. In taking this work forward, IPMS should place a strong emphasis on building capacity to work effectively with communities and the private sector. Improving the understanding of markets and developing government capacity to play a facilitating and enabling role in market development are also important areas of emphasis.
ADB (African Development Bank). 2004. Federal Democratic Republic of Ethiopia: Grant proposal—Fisheries Resources Development Plan Study.
Asche F and Bjorndal T. 1999. Demand elasticities for fish: A review. Globefish Special Series 9. FAO (Food and Agriculture Organization of the United Nations), Rome, Italy.
Breuil C. 1995. Review of the Fisheries and Aquaculture Sector: Ethiopia. FAO (Food and Agriculture Organization of the United Nations), Rome, Italy.
CSA (Central Statistical Authority). 2001. Report on the 1999/2000 household income, consumption and expenditure survey. Statistical Bulletin 258. CSA, Addis Ababa, Ethiopia.
Council of the Amhara National Regional State in the Federal Democratic Republic of Ethiopia. 2003. The Amhara National Regional State Fisheries Development, Prevention and Utilisation Proclamation, 92/2003. Federal Negarit Gazeta, Addis Ababa, Ethiopia.
Delgado CL and Courbois S. 1997. Changing fish trade and demand patterns in developing countries and their significance for policy research. Markets and Structural Studies Division Discussion Paper 18. IFPRI (International Food Policy Research Institute), Washington, DC, USA.
Delgado CL, Wada N, Rosegrant MW, Meijer S and Ahmed M. 2003. Fish to 2020: Supply and demand in changing global markets. IFPRI (International Food Policy Research Institute), Washington, DC, USA, and WorldFish Center, Penang, Malaysia.
FDRE (Federal Democratic Republic of Ethiopia). 2003. Fisheries Development and Utilisation Proclamation, 315/2003, Federal Negarit Gazeta, Addis Ababa, Ethiopia.
FAO (Food and Agriculture Organization of the United Nations). 2003a. FAO Fishery Country Profile: Ethiopia. FAO, Rome, Italy.
FAO (Food and Agriculture Organization of the United Nations). 2003b. Information on fisheries management in the Federal Democratic Republic of Ethiopia. FAO, Rome, Italy.
Holtzman JS. 2002. Rapid appraisals of commodity subsectors. In: A guide to developing agricultural markets and agroenterprises. World Bank, Washington, DC, USA.
IPMS (Improving Productivity and Market Success of Ethiopian farmers). 2004. Fogera woreda Pilot Learning Site: Diagnosis and program design. Unpublished draft report. IPMS, Addis Ababa, Ethiopia.
Eleni Gabre-Madhin. (ed). 2005. Unleashing markets for agricultural growth in Ethiopia: Summary proceedings of the MoARD–IFPRI workshop held 18–20 May 2005, in Addis Ababa, Ethiopia.
MoARD (Ministry of Agriculture and Rural Development), Addis Ababa, Ethiopia–IFPRI, (International Food Policy Research Institute), Washington, DC, USA.
Ministry of Finance and Economic Development, Federal Democratic Republic of Ethiopia. 2002. Ethiopia: Sustainable development and poverty reduction programme. Addis Ababa, Ethiopia.
Reyntjens D and Wudneh T. 1998. Fisheries development in Ethiopia. Project News. Bulletin 11(1). Lake Fisheries Development Project, Addis Ababa, Ethiopia.
Tegegne SD. 2003. Socio-economic study on Lake Tana fishery. MSC thesis (unpublished).
UNDP (United Nations Development Programme). 2005. Human Development Report 2005. UNDP, New York, USA.
Woldu A. 2005. Regional perspectives on market development in Ethiopia: Agricultural Marketing System in the Amhara Region. In: Eleni Gabre-Madhin (ed), Unleashing markets for agricultural growth in Ethiopia: Summary proceedings of the MoARD–IFPRI workshop held 18–20 May 2005, in Addis Ababa, Ethiopia. MoARD (Ministry of Agriculture and Rural Development), Addis Ababa, Ethiopia–IFPRI, (International Food Policy Research Institute), Washington, DC, USA.
World Bank. 2005. World Development Report 2006. World Bank, Washington, DC, USA.
Annex 1. Fish prices: ETB/kg whole fish, June 2006 (unless otherwise indicated)
|Producer prices||Tilapia 3.55||Catfish 2.65||Barbus 1.55||Source FPME||Comment Bahir Dar landing, February 2006|
|3.25||2.20||1.10||Fisherman||Collector boats—near Fogera|
|2.50||2.00||1.25||Fisherman||Collector boats—near Fogera|
|3.00||1.20||1.10||Zege FFA||Collector boats—near Fogera|
|3.50 4.00 2.75 3.25 3.74 3.40||1.50 3.00 1.50 4.05||1.20 2.87 0.50||Zege FFA Co-op Co-op FPME Trader Trader Fisherman||Zege landing Bahir Dar landing, up till June 2006 Bahir Dar landing, new price buyer proposed Bahir Dar landing, agreed price Northern lake—fisherman sells filleted fish Enfranz on road—fisherman must fillet fish Approximate—illegal river barbus|
|Retail/Bahir Dar||21.00||13.00||FPME||February, fillets|
|15.00||Interviewee||Reported price fillets|
|Retail/Gondar||17.00 21.85 15–20||16.20 12.00||10.50 9.00||Interviewee FPME Trader1||Reported price fillets, March Fillets except barbus (whole, gutted, skinned) Fillets (except barbus as above)|
|13–25||Trader1||Fillets—price range over year|
|24.00 22.00 34.00||Trader3 Trader4 FPME||Fillets, maximum price over year Fillets, maximum price over year Fillets, maximum price over year|
|7.00||FPME||Minimum price over year|
|16–21||FPME||Fillets—price range over year|
|27.00||Hotel||Paid by hotel, delivered, fillets|
|Dried fish trade|
|0.90||Fisherman||Nabaga, head-off for drying|
|1.20||Fisherman||Nabaga, head-off for drying|
|10.00||Trader1||Enfranz, buying dried fish|
|20.00||Trader1||Selling dried fish in N Ethiopia|
|Filleting 0.20 Collection on lake/kg 0.43 Collection on lake/kg 0.86||Trader FPME FPME||Cost of filleting/kg whole fish at landing Costs of fuel/skipper—assum-ing 1000 kg fish Costs of fuel/skipper—assum-ing 500 kg fish|
Annex 2. Guidelines for market research in Addis Ababa, June 2006
Visit the four fish trading areas: Mercato + Piazza + Abinet + Bole Road/Tele areaNote who you talk to (not necessarily name—just trader1, trader2 etc.—and which area).
Talk to some of the Enterprise shops too, if possible; ask supermarkets source of their fish.
Talk to lots of people (traders—ambulant and fixed, customers, transporters); keep asking the same questions. Also observe.
Ask price questions at end—do not ask the same trader about purchase and sale prices.
Use your judgement on whether or not to take notes—but if you don’t, then you must make notes immediately after the interview.
1. Status and trends in the market (NB—don’t confuse seasonality/long-term trends)
2. Marketing system
3. Information about these four trading areas
4. Information about the role/presence of fish from Lake Tana