4. FACTORS BEHIND CHANGES IN FISHING EFFORT

4.1 Introduction

Although common property theory (CPT) does not directly include assumptions about demographic growth, there is a very strong tradition in the management literature to do so. Pauly’s attempt (1994) to restore some of the main arguments developed by Thomas Malthus 200 years ago is certainly not the only example. Garret Hardin was also strongly influenced by Malthusian ideas in his famous article from 1968 where he “clearly connects the tragedy of the commons to over-population”, as Brox (1990:233) puts it. Our own findings on how fishing effort has developed in SADC freshwaters actually support the assumption that demographic growth also entails increased fishing effort.

Table 4.1 Demographic growth in the economically active population of 12 SADC countries (in thousands).

Source: FAO, 2000b

However, as shown in the previous chapter, population-driven changes in effort reflect far more complex social processes than simple birth and death rates. If one takes a look at the development in the economically active population between 1970 and 1990 in the 12 SADC countries presented in Table 4.1, one realizes that demographic growth^[9] only provides a partial explanation. While the economically active population increased by 67 percent between 1970 and 1990, according to Table 3.2 the number of fishermen increased by about 160 percent during approximately the same period. Furthermore, demographic growth cannot explain the dramatic shifts in effort that have been observed in the fisheries of Lake Kariba, where the number of fishers decreased by 75% in less than four years after 1963, and then increased by 150% in seven years during the 1980s. Lake Chilwa is another example where demography fails to explain growth in the number of fishers. Fishing stopped completely when Lake Chilwa dried up in 1994, but during the four consecutive years the number of fishers nevertheless grew by 55% compared to what it had been just before the drought. Neither can demography explain the exceptional stability in the number of gear owners and assistants after 1980 in Lake Malombe; situated as it is in an area where the demographic growth is one of the highest in Africa.

The first question that this chapter will address is therefore: which factors other than demographic growth are important for the understanding of population-driven growth in effort? The second question concerns factors that stimulate - or hinder - investment-driven growth in effort. Proponents of CPT assume that fishing effort grows as a result of increased investments. Based on the analysis of a prisoner’s dilemma situation, free access to the resource and the rent it represents, are considered to be the driving forces behind continually increasing investments in production material until the rent has been completely deleted. This becomes a vicious circle between investments, fishing effort and catches: increased investments lead to increased effort, which results in decreasing catch rates, which in turn leads to further increase in effort, and so on. But in Chapter 3, one of our main findings was that investment-driven growth has been very modest in the SADC freshwater fisheries over the last 50 years, irrespective of whether the investors have represented local fishermen or have been expatriate entrepreneurs. In the case of the former, investments have never lasted long enough to result in lasting increases in the investment patterns. Local fishers do not tend to accumulate fishing gear and other means of production; the work organization remains simple with few people involved; and production units tend to remain based on relations of family and kinship. Although fishing technologies have often changed, such changes are frequently necessitated by changes in the environment or by increases in the number of fishermen, rather than being an expression of the producers’ search for a greater share of the resource rent.

Of the four lakes directly included in this study, only Lake Malombe shows major investment-driven changes in effort initiated by the local population. With regards to foreign investors, we have seen many attempts to introduce more capital-intensive production systems. In fact, many Europeans (mainly from the Mediterranean countries) and Asians have tried to make a living in African freshwater fisheries throughout most of the twentieth century. In the last 20 or 30 years, an increasing number of African entrepreneurs have also tried. But these attempts have tended to fail, and in most lakes in the SADC region one may find idle remnants of catch and processing technology brought in by foreign entrepreneurs. In some cases, as in the Kapenta industry in Lake Kariba, the producers are still active even if the levels of investments - particularly on the Zambian side - seem to have fallen during recent years.

Hence, the extent to which we should continue to regard CPT as a useful analytical tool, depends on the answer to the third question that this chapter deals with: why are the CPT assumptions with regards to effort development not valid in SADC freshwaters - why do fishers in these areas not increase their fishing effort by increasing their investments? Let us first take a closer look at the causes behind the population-driven changes in effort.

4.2 Factors influencing population-driven growth in effort

Since fishing effort basically is an aggregate result of an infinite number of micro-decisions taken on a daily basis by the individual fishers or “fishers-to-be”, one may say that there are few aspects in peoples’ daily lives that do not influence their decisions on whether and how they go fishing. However, what we wish to highlight here are factors that can be considered to be of considerable importance for effort development in SADC freshwaters, and that national authorities therefore should take into account in the formulation of management policies. In broad terms, the data of our project indicate that in addition to demographic growth, at least four factors have considerable influence upon population-driven changes in the freshwater fisheries: the natural productivity in each lake; alternative livelihood opportunities; local access-regulating mechanisms; and increased capital requirements in the fishery.

4.2.1 Natural productivity and macroeconomic changes

One of the most significant characteristics of SADC freshwater ecosystems is the large environmental variation that is manifested in fluctuating water levels (see Chapter 5). Research findings from African production systems based on natural resources other than fish show similar fluctuations in the productivity of the ecosystems due to considerable climatic variation (Scoones, 1995). With regards to freshwater fisheries, the most obvious examples are shallow (but potentially very productive) lakes, such as Chilwa, Chiuta, Malombe and Mweru Wa Ntipa, where the water disappears from time to time. Lake Chilwa dried up completely both in 1969 and 1994. And for 20 years (1917-1937) the water connection between Lake Malawi and Lake Malombe was cut off, and Lake Malombe dried up on several occasions during this period (McCracken, 1987). In the first half of the 1990s, water levels in the Shire River reduced considerably again. Also in more permanent lakes such as Lake Mweru and Lake Kariba, there is considerable variation in water levels. Increased water levels generally entail improved inflows of nutrients, which normally bring an increase in the ecological productivity of the system. Such increases in productivity can be considerable and lead to booms in the fisheries of certain species.^[10]

Changes in the natural productivity of the lakes seem to influence the population-driven changes in fishing effort. Increased productivity tends to attract newcomers while reduced productivity clearly has the opposite effect. It is not difficult to imagine how the drying up of shallow lakes also leads to reductions in the number of producers. The opposite happens in periods when inflows (and water nutrients) are high, which leads to an increase in the recruitment of fishermen. This was the case during the filling up of Lake Chilwa after 1994 and in Lake Mweru Wa Ntipa in the 1980s. In Zambia, where access to the new fishery in Lake Kariba was relatively open for people from the Zambezi valley, recruitment rates were very high during the years of inundation. Immediately after the lake had filled up (in 1962-1963) there was a sudden reduction in productivity, which was followed by a dramatic reduction in the number of fishermen.

Fluctuations in the number of producers according to changes in ecological productivity are also reinforced by their high geographical mobility. For example, it is reported that some of the fishermen on Lake Chilwa go to Lake Malombe and Lake Malawi to work during periods when water levels fall. Also, the increase in effort in Mweru Wa Ntipa in the 1980s was a result of heavy immigration to the area (Skjønsberg, 1992). A similar kind of mobility appears in our surveys from Lake Kariba: many of the persons interviewed report that they had been fishing in other water bodies such as in the Kafue plains before they arrived at Lake Kariba.

Fluctuating productivity in the different water bodies affects peoples’ opportunities and thereby their decisions with regards to fish production. These decisions can therefore not be considered in isolation, but must be seen in the context of factors that change their opportunities in other livelihood sectors. In Chapter 3 it was emphasized that occupational mobility among SADC freshwater fishermen is considerable, and that this is mainly expressed through different combinations of fisheries and agricultural activities. Fishing is also combined with other occupations; in particular, wealthier fishermen tend to combine many sources of income. As has been shown for Lake Malombe and the South East arm of Lake Malawi (Hara and Jul-Larsen, 2003), most of the gear owners combine their activities in the fisheries with also being involved as shopkeepers, transporters, bar owners, flour mill managers, carpenters, charcoal producers, etc. Changes in the opportunities in parallel sectors are therefore as important for the development of fishing effort as changes in the natural productivity of the water bodies. Unlike what has been observed in other parts of Africa (e.g. Chauveau and Jul-Larsen, 2000), professional specialization in small-scale fisheries in the SADC region is not very common, and there are few mechanisms whereby professional rights or identities are ascribed to particular groups of people (as for instance in the form of a caste system which one finds among fishermen in Mali, ref. Quensière, 1994). Whereas professional stability in many parts of the world is reproduced through the control of knowledge, the great differences in terms of knowledge and experience between various groups of fishermen in SADC, and the relatively simple levels of techniques and technologies, tends to facilitate professional mobility instead.

In addition to many combinations of occupations, there are also many ways to organize them. The same person may undertake fishing and other activities simultaneously, or sequentially. Furthermore, the combination may be based on some sort of specialization between individuals within the same production unit, or it may be a mixture of different modes of organization. It therefore becomes difficult to identify aggregated patterns and to generalize about the extent to which developments in parallel sectors influence effort development in fisheries. One important exception seems to be the role of what is often referred to as the modern or formal sector; i.e. salary employment in major private or state-owned enterprises. Contrary to what is often believed, there are close connections between formal employment opportunities and informal income-generation opportunities, like fishing. Fluctuations in the macro-economic conditions therefore directly affect fishing effort to the extent that they affect job opportunities.

Gordon’s case study from Lake Mweru (Gordon, 2003) shows that it was mainly people who had lost their jobs in the mines of the Zambian and Zairian Copperbelt after the economic crisis that started around 1973-74, who came to introduce the new Chisense fishery. Similarly, the considerable growth in number of fishers on the Zambian side of Lake Kariba in the early 1980s is explained by the same phenomenon. People from all over the country (but particularly from the north) who had lost their jobs - first in the Copperbelt and later in the state sector in urban areas throughout the 1970s - immediately came to join the Lake Kariba fisheries when the Zimbabwean War of Independence ended in 1980. In a survey that included 250 “foreigners” who arrived after 1980, 83 percent reported that they came from jobs in the Copperbelt or in Lusaka (Jul-Larsen, 2002). Not all of them had jobs in the formal sector; many also worked in more informal enterprises which depended on the formal mining economy. Another example of the influence of macro-economic conditions on fishing effort is Lake Chilwa in Malawi, where the extraordinary increase in number of fishermen after the refilling of the lake in 1995 was partly explained by the return of labour migrants, who had been hit by the loss of opportunities in the South African mining economy.

There is no doubt that the overall economic crisis which has troubled most of the SADC countries for the last 20 to 30 years is one of the main reasons behind the very high increase in numbers of fishermen. However, since the crisis has lasted for many decades and has hit all the countries, it is difficult to assess the extent to which an economic recovery would reduce the number of producers. There are a few cases where employment opportunities in the formal sector have improved. They occurred a long time ago. One example concerns the dramatic reduction in numbers of fishers on the Zambian side of Lake Kariba in 1963. Although it must be considered a reflection of changes in the ecological productivity of the lake, as we described above, the reduction was probably strengthened by relatively good job opportunities in the formal sector during the years just after independence. The Gwembe Tonga already had good knowledge and experience from work migration to other parts of the country as well as to Zimbabwe, and thus easily gave up fishing when opportunities elsewhere improved.

Similarly, the lack of increase of fishermen in the Zimbabwean part of Lake Kariba as long as macro-economic conditions were good there, as described in Kolding’s case study (Kolding et al. 2002a), may be another indication of the “preventive” effect of alternative job opportunities on population-driven growth.^[11] Besides, interviews with hundreds of fishermen leave little uncertainty as to what fishers would choose if they could: jobs in the formal sector are far more attractive to them than their present situation as fishers.

4.2.2 Local access-regulating mechanisms

Neither the heavy increase in the number of producers, nor the high mobility between occupations in the fisheries and in other sectors, should be interpreted to indicate that people’s access to fisheries resources is “free” or “open”. However, in comparison with situations where mechanisms of ascribed rights, knowledge or other mechanisms directly limit people’s movement between occupations and access to resources, people in SADC freshwater fisheries compete for such rights and access on a broader basis. In these societies, competition for access to fisheries is generally high, and there is always a range of social institutions that regulate and sometimes limit people’s access to the resources. With regards to such institutions, it is important to reconsider the distinction between free access and common ownership that was established by the critique of classical CPT. The fact that people living around a lake have more or less equal rights to its waters does not mean that people from outside this local area can claim the same rights. In most villages or camps there are mechanisms that limit newcomers’ possibilities of settling in the community - a precondition for the allocation of rights to natural resources.

In some cases, rights to the commons are directly regulated through the residential rights; if one is allowed to live somewhere, one is also allowed to exploit the shared resources. However, given people’s high geographical mobility, it is often problematic to define the physical and social “boundaries” of a community and people’s “belonging” to this particular locality. Residential rights for outsiders or newcomers may also be associated with a particular activity (e.g. trade), which does not automatically give the individual shared rights to exploit resources. Furthermore, common rights based upon an ethnic or regional identity have tended to be in conflict with government policies - particularly after the countries achieved political independence. At the local level, one therefore often finds a range of unexpected mechanisms that aim to exclude residents from rights to common resources. It is for example puzzling that so few “foreigners” seem to have invested in the Lake Malombe fisheries. Despite their noticeable presence in the villages around the lake, we did not discover mechanisms that openly excluded them from participation. However, when we interviewed those few who had tried to invest in the fisheries, they told us how not only the other fishermen, but also the chiefs, would prevent the foreign boat-owner from dealing directly with his own crew whereby he was constrained in the management and decision-making of his unit. A foreigner is therefore forced to appoint a local representative to run the business on his behalf. Symbolically, this is expressed by a prohibition (a taboo) imposed by the local residents upon the foreigner against visiting the lakeshore when his boat returns from the lake. Not surprisingly, then, the few units belonging to foreigners were among those that experienced the biggest economic management problems. The deprivation of access of “foreigners” to resources is an important factor in the understanding of why the number of fishers has been relatively stable in Lake Malombe since 1980 (see Figure 3.5).

As mentioned above, the occupational mobility of fishers leads to a situation where regulations in the access to other natural resources, and to financial resources as well, may have a direct influence on effort development in fisheries. This was the case on the Zambian side of Lake Kariba in the 1990s. In order to illustrate how complex this process is and how many actors may be involved, we present this case in Box 4.1. It also explains why and how the number of fishermen, according to official frame survey data, declined by 50% during the 1990s (see Figure 3.4).

In sociological and social anthropological literature on natural resources management in local communities, access-regulating mechanisms are often classified according to notions such as “community-based management systems”.^[12] Such general classifications can be useful, but as we will point out, they can also be problematic. One problem is to identify what kind of social institutions should be included in a system of fisheries management. The Kariba case illustrates how difficult this can be, and we are not convinced that the concerns and competition for arable land, tourist areas and theft control can be fruitfully integrated into something called a community-based management system for the fisheries. Another problem is that when we as outsiders construct something we refer to as a local management system (the population concerned do not refer to such arrangements as management systems), it may create an impression that “something exists which does not exist”. As a notion, “Community-based management systems” automatically leads us to imagine the existence of structures intended to solve long-term ecological or social problems. However, whether the access-regulating mechanisms we identified in Lake Malombe and Lake Kariba can be considered as mechanisms intended to regulate fishing effort in a long-term perspective, or simply as results of people’s use of institutional means in their daily struggle and competition for access to natural resources, is not easy to determine. Nevertheless, the answer is crucial if we want to know whether such arrangements can become effective parts of more long-term management institutions.

A related problem is that the notion of “community-based management” may lead to over-emphasis on the collaborative aspects of access-regulating mechanisms at the expense of a focus upon competition between individuals and groups. Elsewhere (Hviding and Jul-Larsen 1995:28-42), it has been underlined that community-based management systems always reflect people’s ability and wish to establish collective action towards the exploitation of the resources, at the same time as they reflect people’s inclination to compete and struggle internally for access to the same resources.^[13]

Resource management as a result of collective action is well documented in the management literature (for a review, see e.g. Acheson, 1981), which often describes how the management arrangements are reflected in practices based upon various kinds of local ecological knowledge. An example also emerges from this study when we emphasized the tendency to increase exploitation when the ecological productivity is high. As already mentioned in Chapter 1, this pattern of behaviour closely resembles the patterns of resource utilization observed in sectors other than fisheries. In line with the findings of Ellis and Swift (1988), which show that this “opportunistic” practice is an ecologically quite sound adaptation in pastures, Chapter 5 (examining the biological and ecological effects of increased fishing effort) shows that this is valid for fisheries in SADC freshwaters as well. People’s knowledge about the ecological rationale of “opportunistic fishing”, is evident in Zambian fishermen’s long-lasting and collective resentment against the authorities’ prohibition of certain fishing methods, such as ‘Kutumpula’. The history of fisheries management (see Chapter 3) is full of examples where formal management authorities have implemented and enforced prohibitions without any form of empirical investigation. In the case of Kutumpula, more recent research (Kolding et al., 2003b) has demonstrated that the assumptions behind the banning of this fishing method were groundless, and that many fishermen actually possessed the necessary knowledge to prove this (see also Malasha personal data). When fishermen demonstrate such extensive knowledge about the ecology of their fisheries, it is perhaps not so surprising that there is often resistance in the communities to the introduction of certain types of new gear, and that they develop local regulating mechanisms to avoid introduction of gear that they consider harmful for the collective benefit of the fishery. For example, in a situation where local Zimbabwean fishermen in Lake Kariba were in conflict over most things and where it was extremely difficult to establish any kind of collaboration, they managed to effectively keep beach seines away from their shores.

Generally speaking, we have found that local access-regulating mechanisms are far more effective in preventing or limiting population-driven rather than investment-driven growth of fishing effort - particularly if the investment-driven growth comes from within the community. This is quite clear if we look at how the access-regulating mechanisms in Lake Malombe have successfully limited newcomers’ investments in the fisheries, but how they have been unsuccessful in preventing the existing boat-owners increasing their investments in new and more gear. Also on the other lakes, it is the population-driven growth in fishing effort that seems particularly affected by local access-regulating mechanisms. This leads us to think that such mechanisms must often be interpreted as expressions of people’s struggle to control their access to the resources, rather than more long-term arrangements aiming to conserve the resource base.

4.2.3 Capital investments in the fisheries

Finally we observe that where investment-driven intensification has taken place through the introduction of more capital-intensive fishing methods, this seems to reduce the possibility for population-driven growth in effort. In Lake Malombe the shift from gillnets to a variety of much more capital-intensive seining methods, has substantially increased the entry fees into the fishery and thereby reduced the number of potential resource users. Despite a considerable increase in effort, the number of fishermen in the 1950s and 1960s was probably higher than it is today, and during the 1980s the number of fishermen remained fairly stable (see Figure 3.5). This is a type of effort development that can be observed in most places where the fisheries have undergone major technological and organizational changes. One example of increasing differentiation between fishermen and exclusion of the poorest when the fishery becomes technologically more advanced, can be found in the difficulties of participating that Norwegian small-scale coastal fishermen faced when the fishery became more capital- and technology-intensive from the 1930s onwards (Brox, 1990). However, since investment-driven growth is still not very common in SADC freshwater fisheries, this mechanism is of little importance as a limiting factor for increased participation in the fishery at the moment. In the next two sections, we will look in greater detail into the causes behind what many scholars and policy-makers have viewed as “lack of development”. As our findings show, this view needs much refinement.

4.3 Factors impeding investment-driven growth

Since much of the literature on management of African fisheries more or less assumes such a growth, the findings in Chapter 3 of very limited investment-driven growth in SADC freshwater fisheries may appear surprising. However, confronted with the huge literature on economic development in sub-Saharan Africa it can hardly be considered an astonishing conclusion. Economic growth in Africa has been very slow for many decades, particularly in the rural areas where freshwater fisheries are located. Much of the research on African economic development has been focused on identifying constraints to this phenomenon and it is within this literature we have to seek an answer to why the investment-driven growth has remained so limited, despite the increased demand for fish which is being registered in all the sub-Saharan countries.

Broadly speaking, one may say that the literature focusing on the problems in economic development contains two main currents of thought - those who put most importance on the impact of external power groups and those who focus on internal cultural or structural constraints. In his attempt to reveal the bottlenecks for vertical growth in the traditional Norwegian cod fishery, Brox refers to two theoretical models. One is the classical peasant model represented by Wolf (1966) and Shanin (1971) which mainly focuses on the power relations and the exploitation of the peasants by the wider economic systems in which they operate. The other theory is related to Chayanov’s theory, from the beginning of last century, of peasant economic behaviour, which in more recent times has been taken up in James Scott’s model of a “moral economy” (1976). Here the constraints are connected to cultural attitudes and values in the peasant community. Brox finds that these two explanatory models to a large extent explain why fishermen in northern Norway for a very long time did not seek to accumulate or invest in new gear. The combination of middlemen who extracted and controlled the resource rent, and an economic rationale among the producers aimed at reaching specific production targets rather than general expansion, led to a situation where there were few incentives among fishermen to invest in new gear.

Though peasant theory to some extent explains “lack” of development in northern Norwegian fisheries a hundred years ago, it seems less relevant for an explanation of the situation in SADC freshwater fisheries. Peasant theories based upon the existence of clear politico-economic hierarchies and economic exploitation seem of limited value in the study of rural Africa where these hierarchies are often multiple and combine lots of different interests. Our own empirical material supports this. We have not observed any direct economic exploitation of fishermen in the sense that the values they create were being systematically appropriated by other groups of actors, e.g. traders. It is more common that traders or middlemen have very little control over the harvesters and in most cases it seems as if the fishing population is more in control of their daily lives than the fish traders are. This is reflected in a phenomenon appearing in several of our case studies (Hara and Jul-Larsen, 2003; Overå, 2003): trade is often used as a stepping stone for accessing fisheries, while the opposite rarely happens. Although they try, traders have big problems gaining control over the production process by extending credit to fishermen. Instead of securing a stable supply of fish, the indebted fishermen tend to disappear and traders who attempt this strategy most often lose their capital.

The debate about whether people in rural Africa primarily act as ‘economic agents’ or whether they act on the basis of target-oriented principles of a moral economy, never seems to fade out completely. Last time the debate revived was in the 1980s when Göran Hyden introduced the concepts of “the economy of affection” and “the uncaptured peasantry” (1983) to explain why the agricultural policies in Tanzania were so unsuccessful. In our view, this debate is not particularly fruitful. Although the fact that investment-driven change (when it occurs) mostly involves “foreigners” may indicate some relevance for the notion of a local “moral economy”, it does not explain why virtually all investment attempts (that actually do take place) tend to be temporary and are followed by reduction in investments and eventually a collapse of the enterprise. The Kapenta fishery in Lake Kariba, for example, illustrates the great constraints on technological and economic development in the southern Africa context, despite clear ambitions towards investments and expansion among the entrepreneurs (Turid Bøe, pers. com.; Overå, 2000). These observations indicate that the problem is far more complicated than the question of individuals’ personal behaviour, attitudes and aims. In this context, it seems more fruitful to investigate in further detail the institutional landscape - i.e. not only the rules but also the underlying values and norms - under which the fisheries (including production as well as processing and distribution) operate. Therefore, let us look at some of the factors that seem to have been preconditions in the few instances when investment-driven growth actually has happened in the lakes we have studied.

4.3.1 Market development

First of all, the existence of an urban and/or export market for fish has been a precondition for the possibility of maximization of the resource rent through capital investments in SADC freshwater fisheries. In Malawi, for example, the population density of the Shire Highlands has been very high since early colonial times. When plantations were established, and an export market developed in Zimbabwe (then Southern Rhodesia), it was the emergence of these new markets, as well as the growing domestic urban market, that made capital investments in the fishery interesting, both for African traders and for colonial entrepreneurs (Chirwa, 1996). In the case of Lake Mweru, it was the copper mining industry in Zambia (then Northern Rhodesia) and the Democratic Republic of the Congo (then Belgian Congo) that created new markets (Gordon, 2003). Where African traders had previously traded fish to the urban areas on bicycles on a small scale, investors with capital could operate on a much larger scale because they were given the opportunity to supply the copper mines on a contract basis. It is thus not only the existence of an outlet for fish that matters; predictable profits from the market are key for investors in their decision to invest in the fisheries.

4.3.2 Infrastructure

A physical infrastructure that makes it possible to transport fish from a lake to the market is also of vital importance. Some of the lakes in this region (such as Lake Bangweulu) are located in areas that are sparsely populated and remote from the markets. The distance between the lake and the market is not always very long in kilometres, but it can be very far in terms of hours and days spent on a trading trip, punctures, petrol expenses, bad roads, unsurpassable rivers during the rainy season and other hardships. Without doubt, the remoteness of the Zambezi Valley and lack of roads leading to Lake Kariba in the 1960s, the collapse of the little infrastructure there was during the Zimbabwean war of liberation, and deteriorating maintenance of roads as the Zambian economy declined in the 1980s and 1990s, has meant that poor infrastructure has limited investments in the fisheries of Lake Kariba (Overå, 2003). An important strategy for investors in the area in the 1990s is thus to invest in roads, transport and ice plants. Since such strategies demand large amounts of capital and political mobilization, this has primarily been a concern for the Kapenta entrepreneurs and tourist operators. However, on a smaller scale, traders who attempt to establish stable supply contracts with particular fishing camps, have in some cases managed to mobilize the labour of local residents for the communal maintenance of roads in order to ensure access to the fishing camps during the rainy season (Ibid.). Another example of the importance of infrastructure, is the building of a new road to connect Lake Mweru with the expanding urban markets in the Copperbelt in the 1930s, which was a precondition for the investment boom by expatriate entrepreneurs in the fisheries (Gordon, 2003).

4.3.3 Production-distribution linkages

Investments in fisheries often happen through vertical integration of production and distribution in order to reduce the uncertainties of fish supply (for the distributor) and market access (for the producer) (Platteau and Abraham, 1987). There are three well-known strategies of linking production and distribution:

• the boat owner invests in transportation and establishes his/her distribution network in order to control more of the market chain,

• large market actors invest in fishing equipment in order to secure and increase their own supply of fish,

• the fisher/boat owner and trader enter into a contract where the former gets a steady market outlet (and often receives equipment on credit) and the latter receives fish in return (often on credit).

This last strategy is the most common when it comes to small-scale fisheries.

As we saw in Chapter 3, the expatriates who invested in fishing vessels in Lake Malawi employed the first strategy by also investing in lorries to ease their access to the markets and increase their profits. The same is the case with the Kapenta operators when they seek to establish their lines of distribution to overcome the insecurities of the present marketing system. The investments in Lake Mweru happened through the second strategy whereby expatriate businessmen invested in new fishing technology in order to increase their supply of fish for the Copperbelt markets. The Africans who invested in fishing boats in Lake Malombe, on the other hand, relied on the large numbers of small-scale traders to sell their fish. The boat owners thus got access to the market, but their relationship with the traders did not involve extensive and profitable credit-supply contracts as such (the third strategy): to invest in fishing equipment they depended on remittances from labour migrants rather than on credit from traders (Hara and Jul-Larsen, 2003). Interestingly, the cases where investors in the fishery managed to extend their control to include the whole market chain (colonial Mweru and Malawi) seem to be the result of particular privileges given to expatriate entrepreneurs by the colonial governments (e.g. allocation of land, supply contracts and fishing concessions). When they lost these privileges (as in Lake Malawi the 1950s and in Lake Kariba in the 1960s) the expatriate investors experienced the same problems as the Africans in integrating production and distribution.

The inability of producers to link themselves up to the market in order to increase their share of the profits, as well as the inability of market actors to gain control over production seems to be one of the most important direct reasons for why investment-driven growth in effort in most SADC freshwater fisheries has been so limited. The difficulty of integrating production and distribution through relations of trust and/or control is an institutional problem that seems almost insurmountable in these fisheries contrary to what happened, for instance, in the West African canoe fisheries (Overå, 1998; Chauveau and Jul-Larsen, 2000). Hence, in Lake Mweru the processing companies do not invest on the production side, and in Lake Kariba the Kapenta operators have not been able to establish a marketing system for their produce that is independent of small-scale traders and fluctuating prices. Likewise, when they invest in marketing of inshore fish, they make sure they do not waste their capital on investments on the production side (Overå, 2003).

4.3.4 Capital

Obviously capital is a major “component” in processes of investment-driven growth. As we shall discuss later, lack of capital is not primarily a practical bottleneck; it is a central institutional problem. However, in the SADC countries, the macro-economic situation has for many years been so difficult that the potential for accumulation of capital through the fish trade has been very limited. One could almost talk about an “involuted” fish market, where a growing number of small-scale fishermen produce for a growing number of small-scale traders who market the fish to increasingly poor customers^[14]. The expansion in the Chisense fishery in Lake Mweru in the 1980s could be described as such a process of involution: the increase in effort in the “new” fishery was not a result of more efficient and capital-intensive technology; it was rather the result of absorption of large numbers of unemployed women and men into both production and distribution of a commodity (Chisense) that suited impoverished consumers in a declining urban economy. Therefore Gordon (2002) characterizes this as population-driven rather than investment-driven growth in fishing effort. In such a system, there is not much capital in circulation that could have been invested in modern fishing equipment: potential investors are caught in a sort of poverty trap.

Another constraint on access to capital in the SADC countries is of course that banks and other credit institutions that provide loans to small-scale producers and traders hardly exist. This is one of the reasons why investments in fisheries have depended upon input of external capital, as in the case of Lake Malombe where most of the capital came from labour migration. In this context it is interesting to note that whereas macro-economic recession often leads to population-driven growth in fishing effort, it constrains most possibilities for investment-driven growth. For example, the reduction in investments in the Lake Malombe fisheries in the 1990s is at least partly explained by the decline in labour migration to South Africa and Zimbabwe, after the crisis in the mining economy there started in the 1980s and continued throughout the 1990s.

4.3.5 The local institutional landscape

There are many practical constraints like the ones just mentioned, that limit investments in SADC freshwater fisheries. Most of them are already mentioned in numerous pieces of research and commissioned reports. However, even if factors such as infrastructure, the organization of production and distribution, and access to capital are central to whether investment-driven growth will take place or not, this only poses the questions at another level: why do these practical problems continue to constrain investments when we know that the demand for fish has increased substantially, that most individuals and groups involved seem eager to overcome them and that great amounts of foreign aid have been spent to solve them? Why does investment-driven growth not take place in the SADC freshwaters when such growth arguably has taken place in other small-scale fisheries in the South like in West Africa and in Asia, despite many of exactly the same practical problems?

The analytical perspective on institutional development and underlying much of the co-management thinking provides some useful understandings of these issues. Jean-Philippe Platteau stands out as one of the important contributors to the school of new institutional economics. One of Platteau’s concerns has been to explain why small-scale fisheries in developing countries seem to do so well despite the emergence of modern production systems, which according to conventional economic theory are far more economically effective (Platteau 1989a and b). In order to explain this, he starts by observing that the market systems in most fisheries in developing countries do not function as conventional economic theory assumes. The ‘dysfunctions’ are caused by market imperfections (i.e. information about availability and prices of fish does not circulate freely and is not evenly distributed among the actors), which may be observed everywhere, but particularly in the new markets in developing countries (and SADC is no exception). In order to get access to the information required, the actors in the market have to spend time and other resources. These costs are what the economists call “transaction costs”. Platteau argues that a fishery based upon “modern” market institutions implies considerable transaction costs, which may often be higher than those of a fisheries based upon more “traditional” institutions^[15]. In order to control labour, get access to fish, administer credits and security systems, the actors in the market will try to reduce transaction costs by making use of various types of social control mechanisms. In an imperfect market, labour and credit must be controlled by using alternative rules of recruitment, and access to fish depends on various types of interdependent control between sellers and buyers, otherwise the actors will lose and the system will collapse. “Traditional” institutions based on various types of personalized trust such as patronage, and permanent buyer-seller relations are often more cost-effective than modern institutions in terms of overcoming all these transaction costs. Platteau’s analysis shows that, where the market imperfections are considerable, traditional forms of organization may simply be more profitable for the actors than the modern ones. However, the institutions have their limitations, especially in terms of the number of workers and the amount of credit that can flow through such systems. Complicated owner/worker relations thus provide limited room for accumulation and expansion, and this constrains the ability and profitability of investments. Production technology must therefore remain on a capital-extensive level.

This type of analytical approach seems particularly fruitful in explaining why a great number of industrial attempts have failed in SADC freshwaters. In both Lake Malawi and Lake Tanganyika, one of the problems of the industrial sector has been competition from artisanal small-scale fisheries. In Lake Mweru the direct cause for the disappearance of the industrial vessels was the collapse of Mpumbu, but this does not explain why the fleet was not set to target other species. In Kariba, the semi-industrial Kapenta fishery is not competing with the inshore fishery. Nevertheless, the new institutionalist approach is useful in understanding the difficulties of expansion that this fishery has due to the highly imperfect markets (Overå, 2003).

However, the same approach is less fruitful when it comes to understanding why the investment-driven growth from within the small-scale fisheries - contrary to what we observe in West Africa and Asia - seem to be so slow in SADC freshwaters. Part of the problem is related to the maintenance of the dichotomy between modern and traditional institutions, which makes the approach unsuited to analysing developments and changes within the traditional sector itself. Furthermore, it tends to consider traditional institutions, such as kinship relations, tribal or clan relations and patron/client relations, as phenomena that can easily be defined according to the set of rules which guide them. In the ‘real’ world this is not the case. Institutionalized rules for rights and responsibilities in a rural African community, and the underlying values and norms legitimizing these rules, are not clear and commonly shared among community members, and this complicates the analysis of the role of institutions. Studies of institutions in rural Africa in fields other than fisheries indicate that considering the question of investment-driven growth simply as a matter of “traditional” vs. “modern” institutions is far too simple. In-depth historical and ethnographic research (Berry, 1985, 1993, 2001; Peters, 1994; Fairhead and Leach, 1996; Guyer, 1997) has documented the extent to which local rules and regulations - crucial for people’s access to vital resources - are unclear and ambiguous: sometimes they even contradict each other.

Our research supports these findings. Looking at how access to fish is regulated among potential buyers, we find that many different rules may be applied at the same place at the same time. In Lake Kariba a wife will normally have certain rights to the fish of a fisherman, which she will then dry and sell on the market in Lusaka. But other members in the fishing community may also claim fish from him with some sort of legitimacy. They may be relatives or they may have helped the fisherman when he established himself, or they may have drawn his nets when he was sick. In addition, a number of traders may also have rights in his catch. Some may be his relatives, another may have been particularly helpful at certain occasions like in a marriage arrangement or when one of his children got sick and needed medical attention. Others again may have helped him with small loans of goods or money, which gives them some rights in return. At the same time, no rules exist as to how much of the catch each of the entitled people can claim, so this is generally open for negotiation and discussion and will depend on a lot of factors. A fisher/fish buyer relationship is thus often far less defined and includes many more actors than many scholars - including the neo-institutionalists - tend to believe. The relationship is based upon a wide range of often conflicting rules and values, and this makes agreements and contracts very unpredictable. Although this situation may secure at least some access to fish for a large number of people, it is not well suited for those who want ensure increased and stable fish supplies. If the fisher in the example above wants to increase his production and needs credit, the traders will definitely be reluctant to give it to him: their extension of credit will not automatically guarantee their access to his fish. The traders will still have to compete with all the others who are entitled to some of the catch, and whom the fisher cannot ignore.

Another important aspect of the ambiguous institutional landscape is the relationship between owner and crew. Even if owners continuously try to build stable labour relations upon existing mechanisms of social control like kinship relations, they are often unsuccessful. The crews will often claim that they are being exploited in ways that are not acceptable. As a result, they launch activities that jeopardize the basis of their own source of employment and income. A good example is the way in which most of the crewmembers on the Kapenta rigs in Lake Kariba are probably involved in thefts of fish from the owners. They do this with reference to how bad the working conditions are. Many people in the fishing communities support their views and their actions and it is therefore limited what owners can do in order to secure their catches. Also in Lake Malombe, all owners continuously complain about their lack of control over crews in exactly the same way as Crispo Gwedela did to the “Commission of Enquiry into the Fishing Industry” almost 50 years earlier^[16]. Their testimonies are a long story of attempted thefts and destruction of fishing gear and other material (Hara and Jul-Larsen, 2003). The Malombe community’s sanctions against the presence on the lakeshore of “foreign” owners when their boats landed the catch, is yet another example of the ambiguity within the local institutional landscape. Workers shifted from owner to owner, and they also developed a system whereby owners (especially those who were not indigenous to the area and those who were women) were excluded from selling the catch. Owners’ lack of control made the cost of labour prohibitive. From the point of view of the owner the problems they faced qualify as theft and sabotage, but in the view of the fishermen and other community members, this was not necessarily the case. This institutional problem made it risky for investors to rely fully on fishing, and they therefore tended to diversify their investments in several sectors outside the fisheries rather than in the accumulation of fishing gear. In both of these cases, the conflicts between owners and workers represent something much deeper than resistance by individuals to accepting ‘the rules of the game’. The problem is that the rules themselves are unclear and ambiguous, and this makes it possible for all parties to find legitimacy for their actions. It is not difficult to imagine how such a situation affects the owners’ interest and willingness to increase their investments. Similar studies in the agricultural sector in Africa have led the anthropologist Sara Berry (1993) to conclude that institutional ambiguities and contradictions related to the access and use of labour, probably more than those related to access to land or financial resources, represent the most important constraint for economic growth in rural Africa.

Only by introducing this type of perspective can we understand why the Kariba inshore fishery, the Bangweulu swamp and the Chilwa fisheries have developed so little in technological and organizational ways. As long as the local institutional landscape does not improve so that new types of mutual trust and reliable behaviour can emerge among the actors, it is very difficult to see how investment-driven growth in these fisheries can take place. But the constructivist institutional perspective has important bearings on how resource management can be conceived. In this type of institutional landscape, the introduction of any externally driven measures of management will have few chances of success. On the contrary, one may easily foresee that such measures may lead to a further proliferation in ambiguities in the local institutional landscape. Internally driven changes may have a better chance in this respect, but as we demonstrated in the case of Kariba, it is probable that internally driven measures have their basis in the interests of specific groups and not in those of the community as a whole. We shall return to these issues in Chapter 6, but first we shall look at the consequences of increased effort on the regeneration of the fish stocks.