Managing common pool resources in the Kafue Flats, Zambia: from common property to open access and privatisation

Abstract

Governing common pool resources in floodplains is a challenge due to high variability between seasons. Nevertheless, the case of the Kafue Flats in southern Zambia illustrates how local groups (Ila, Balundwe and Batwa) have developed common property institutions governing access to fisheries, wildlife and pasture. After the colonial and post-colonial periods these institutions were altered or eradicated by state control. State institutions have failed for the following reasons: complex economic and political processes and lack of knowledge have made state management ineffective; local rules have been eroded or severely altered by more powerful actors; immigrant groups (seasonal fishermen, commercial hunters, absentee herd owners) have increased their bargaining power as citizens; and neither local nor state institutions are enforced due to limited state capacity. This has led to open access situations and partial privatisation, both of which are major causes of unsustainable use of the commons.

Keywords:

• Floodplain
• fisheries
• wildlife
• pasture
• common property institutions
• state management
• institutional change
• power relations

1. INTRODUCTION

This paper explores how local institutions for managing pastures, fisheries and wildlife were operating during pre-colonial times and how they have changed during colonial and post-colonial times. The authors argue that local power structures have been influencing the management and institutional design of resource use that was based on local knowledge of this dynamic ecosystem, which changes annually from dry to flooded conditions. The main argument is that institutional change occurring since colonial times is at the heart of the problem of overuse of state-controlled common pool resources. Dismantling of local institutions by the state has changed some of the resources, such as fisheries and wildlife, to open access, while others, such as pastures, have been subjected to semi-privatisation.

We offer the following definitions to clarify the debate on the commons. Fisheries, wildlife, pastures and so on are common pool resources from which exclusion is difficult (but not impossible) and are subtractable (i.e. a unit used at one point in time is not available for other users; Becker & Ostrom, 1995). How these resources are owned and managed is then decided by the actors involved. The term ‘common property’ therefore refers to a regime and does not mean the same thing as ‘common pool resources’. Many common pool resources are managed in common property regimes (collectively owned by a group of people) but they can also be managed as state property by state actors, or as private property by an individual who obtains full control, or as open access, meaning that everybody uses the resource freely (McKean, 2000).

This study is based on the authors' research and on an analysis of the literature covering 40 years of study in the Kafue Flats, both well-founded scientific research and other works, published and unpublished, on the management of natural resources in the Kafue Flats wetland; in particular, recent research and literature reviewed by Chabwela and others (Chabwela, 1992a,b) and by Haller (Haller, 2007; Haller & Merten, 2006, 2008). Haller's work is part of a larger comparative project called the African Floodplain Wetlands Project (Haller, 2002, 2005, 2009).

The Kafue Flats is one of the most important wetlands in southern Africa and the most economically significant in Zambia, especially for the capital Lusaka, only 50 km away, and for smaller cities on the railway line. It covers an area of 6500 km² of floodplain grassland from the Itezhi-Tezhi Dam to the Kafue Gorge and forms a major component of the Kafue drainage system. The floodplain is bordered to the west by the dam and to the east by the gorge. Between these borders the floodplain is about 250 km long and 60 km wide (at its widest point; see Figure 1). The area is extremely flat – the river drops only 15 m in 400 km, forming large meanders, lagoons, oxbow lakes and ponds.

Figure 1: Map of Kafue Flats

The floodplain is inundated each year in the rainy season from November to April. This creates a resource-rich habitat for 77 species of fish, mainly bream (Tilapia spp., Oreochromis spp.), catfish (Clarias spp.) and sardines (Alestes spp.), and 127 species of mammals (Chabwela, 1992a; Ellenbroek, 1987). Apart from zebra (Equus burchelli), wildebeest (Connochaetes tauirinus), kudu (Tragelaphus strepsiceros) and hippo (Hippopotamus amphibius), the lechwe antelope (Kobus Leche kafuensis) is the most significant because it is endemic. The rich grasslands contain more than 17 species, among them the Echinochloa and Vossia cuspodata grasslands that are important for cattle and wildlife in the dry season (Chabwela, 1992a; Haller, 2007).

These common pool resources have declined drastically (for a summary, see Haller, 2007). Reports from the colonial periods as well as local views indicate that fisheries, wildlife and pastures were relatively abundant and largely under-used (Mortimer, 1965; Smith & Dale, 1968; Chabwela, 1992b; Subramaniam, 1992; Haller, 2007). However, from the mid-1970s to 2004/05, fish catches have been reported to decline from 10 to six to three metric tonnes (Haller & Merten, 2008) and lechwe antelopes from 120 000 to 40 000 or 20 000, while pastures have been significantly reduced by some 25–50 per cent (Howard & Jeffrey, 1987; Chabwela 1992b; Kapungwe, 1993; Kampamba, 1998; Haller, 2007). Are these signs of a tragedy of the commons? To answer this question, the authors focus on historical data illustrating the institutional change that has taken place in the area, and argue that former common property regimes for managing common pool resources have changed to open access and, in rare cases, to privatisation. We further argue that changes in relative prices stemming from external changes have increased commercial interest in these common pool resources and led to institutional changes that are one of the most significant causes of the overuse of fisheries, wildlife and pasture in the area.

2. PRE-COLONIAL HISTORY AND ETHNOGRAPHY

The Kafue Flats wetland is inhabited by the Ila, the Balundwe (Plateau Tonga) and the indigenous peoples of the area, the Batwa fishermen (see Figure 1). The Batwa live on the levees along the Kafue River and have been settled in the area for over 100 years (Lehmann, 1977). In fact, the oldest villages at the riverbank go back up to 300 years (Haller, 2007). Today, the three ethnic groups are organised into 14 chiefdoms. The Ila are transhumant pastoralists and agriculturalists. Their major subsistence activities are cattle husbandry and cultivation of maize, but fishing and hunting have been important for subsistence in the past. The Ila households are largely polygamous, and kinship organisation is bilateral (Cutshall, 1980). In the past, the Ila and Balundwe lived in large villages to protect themselves from slave and cattle raids from the Lozi Kingdom in the north, and from wild animal attacks. After the Pax Britannica, by the 1930s, these large villages headed by ‘big men’ (i.e. leaders, mwami) broke up. Today, the people live in villages made up of scattered hamlets (Haller, 2007). Although the Ila were the victims of Lozi attacks, they were nevertheless known as fierce fighters. Conflicts were common between villages and between chiefdoms. However, there were collective access arrangements between such local groups with regard to mobile common pool resources such as wildlife and fisheries (Smith & Dale, 1968; Colson, 1970; Fielder, 1973; Cutshall, 1980; Haller, 2007).

In pre-colonial times the areas known as chiefdoms today had specific names linked to neighbourhood territories (chichi) that had clear boundaries and were subdivided into village territories. Villages were headed by big men who tried to attract people from other areas by distributing access to resources and giving them cattle or making cattle available for bride price. Although these big men were competing for followers, the concept of ‘chief’ did not exist. Traditional conflict resolution mechanisms and, at the same time, collective choice arrangements, were the so-called lubeta, meetings at which local big men resolved conflicts and discussed village affairs (Smith & Dale, 1968; Haller, 2007).

Power and access to resources were embedded in religious beliefs: all groups are traditionally monotheists who worship the supreme being Leza but also believe in the existence of spirits and especially ancestral spirits (mizhimo). Leaders argue that the spirits gave them the right to land, pasture and fish resources, and have a strong influence on the monitoring and sanctioning aspects of institutions governing these common pool resources (Haller, 2007). The ancestral spirits are very important for rainmaking and for foretelling sickness and other problems, and are worshipped at shrines (malende) (Colson, 1970; Haller, 2007).

3. LIVELIHOOD STRATEGIES AND INSTITUTIONS

This section describes the institutions that were developed in pre-colonial times, since it is important to understand how their rules and regulations worked and how robust they have been (see Ostrom's design principles; Ostrom, 1990). It should be noted that some of these institutions were partly operational during the colonial and post-colonial times.

3.1 Cattle husbandry and pasture institutions

For the Ila and Balundwe, the main activity and source of identity has always been cattle herding−and according to Fielder the Ila were the richest cattle-owning group in this part of Africa (Fielder, 1973), with about 13 head of cattle per adult male person. Cattle numbers have always been distributed unequally, but as cattle were not predominantly private property but the communal property of a larger resident group (extended families and final and affinal relatives), these differences could be mitigated. Apart from their symbolic and high political value, cattle also had economic value. Milk and milk products, such as sour milk, were very important for the diet of the Ila, much more than beef, which was consumed only on special occasions such as funerals. Cattle were a sign of wealth and power and part of the local social security network, while fish and game supplied the necessary protein.

The transhumance system (kuwila and kubola) consisted of coordinated movements between the villages in the woodlands adjacent to the Flats and the cattle camps (lutanga) in the Flats. Access to pastures had been regulated by giving access to cattle camps distributed by a local control group in the pastoral area claiming spiritual ownership as overseers. The case of the territory of Mbeza is illustrative. Access rules evolved on the basis of conflict over coordinating movements to the pastures, where one cattle camp for all the villages of the area had once been set up. The problem was solved by the local control group allocating other cattle camp sites for the other groups in the area, but asking one head of cattle for a long-term hereditable usufruct right to pasture (the matanga rule) for coordinating and supervision work (Haller, 2007).

3.2 Fisheries and fishing institutions

Fishing was important for the Ila and Balundwe for food security and protein intake since they seldom consumed beef, but for the Batwa it was a basic activity. They fished in lagoons, oxbow lakes, ponds and in the main Kafue river and its tributaries. Today fishing activities are still coordinated according to flooding patterns. This seasonality means that there are many different institutions related to fisheries. There are two main ones. First, one needs to be a member of a village community – the notion of a local leader's spiritual ownership is important, as he is the person who distributes access rights to river and tributary sections and ponds. Second, the local communities established clear territorial rules according to timing (flooding periods or dry seasons) and gear (gender specific and season specific, baskets for women, spears for men, traps, no nets).

There are other important monitoring and sanctioning rules and regulations related to season, local knowledge of breeding characteristics of fish, and distributional justice. During the first inundations, ikuo (basket and torchlight fishing) takes place at the village level. During high floods, fisheries are open access with two restrictions: firstly, people are not allowed to camp and make fires along the higher grounds; and, secondly, no fishing is allowed in the special areas known to the Batwa people where bream make nests in the sand on the river bed. When the water is receding, weirs are constructed and traps are set by men in the tributaries (private property). In the early dry season, fishing is stopped completely and the area is monitored by a ritual master (utamba) and his helpers. A time is then set by the spiritual owner of an area in order to call members of village, territory and neighbouring groups based on reciprocal invitations to fish collectively. To prevent people from fishing before the collective fishing days, the utamba monitors the area. Free-riding is punished by fines, and the belief that the ancestors will cause the free rider to be attacked by crocodiles discourages the practice. Final collective fishing events are organised in the pastoral areas at the level of cattle camp users. The Batwa themselves had similar rules but also had a special fishing technique unique to the area (shadow fishing with spears in the reeds along the Kafue River; Haller, 2007; Haller & Merten, 2008). Table 1 summarises how the traditional fishing institutions are adapted to flooding patterns.

Table 1: Fishery regimes according to season among Ila, Balundwe and Batwa

Season	Where	Regime	Rules	Gear and methods	Costs of economic dependability
Rains (December–February)	Tributaries, ponds and river sections	Closure communal collective (people from chichi and others)	Breeding areas at the river	Baskets (women), spears (men), ikuo	Low
			Fishing in village and chichi reciprocity		Higher
Floods (February–April)	In all inundated areas	Open access	No rules	Boats, spears, Tonga baskets (women)	Highest (low for specifically known places, end of season)
			Some exceptions: no fires, no fishing in breeding areas
Water receding (May–July)	River, tributaries and ponds	Closure private communal collective	No fishing in restricted areas	Some weirs (buyeelo) still allowed	Higher to low
			When water stops flowing, invitation by ritual master	Lwuando controlled
			Reciprocity	Ila baskets and spears
Dry (September–November)	River, tributaries and ponds	Communal to closure	Reciprocity	Lwuando collective fishing in ponds	Low
		Batwa still fish at Kafue River	Ritual master closes fishing when water is low	Ila spear fish standing in shallow water
				Batwa spear from boats
Source: Haller (2007).

3.3 Hunting and hunting institutions

Traditional tribal hunters used traps such as snares and pitfalls, and sticks, spears and dogs. More recently, shotguns have been introduced in the area and are used to hunt large animals. The most important animal to be hunted was the local endemic lechwe antelope for meat, for skins for women's dress and for paying tributes to the Lozi Kingdom (Haller, 2007). Within a chichi, wildlife was common property and all community members were allowed to hunt, but hunting in another chichi without permission was dangerous – violators could be beaten or killed. However, the Batwa, Ila and neighbouring Balundwe faced the problem that game such as the lechwe lived in migrating herds, which did not stay within the chichi boundaries. To secure a good hunt, the inhabitants of each territory collaborated in collective hunting activities called chila, coordinated by a local leader as spiritual owner. Ila, Balundwe and local Batwa people would be invited by such a leader to meet in the Flats in a specific area, where they would encircle the large lechwe herds and kill 2000−4000 animals. Hunting was done with spears and dogs on land and canoes in water (Haller, 2007). However, it should be noted that hunting was forbidden for a period before and after the collective events and the area was supervised by local leaders.

An analysis of the robustness of these pre-colonial institutions shows that most of Ostrom's (1990) design principles can be found in the institutions that have developed on the Kafue Flats to manage fisheries, wildlife and pasture as common property. All areas have clear boundaries but, in order to adapt to the mobility of the resource and the flooding patterns, there are reciprocal arrangements. There are also clear monitoring and sanctioning devices and conflict resolution mechanisms, to name the most important principles, and there are coordinating bodies embedded in religious belief systems (see analysis in Haller, 2007; for the Kafue Flats case in comparison with other floodplain cases, see the African Floodplain Wetlands Project in Haller, 2009).

4. ECONOMIC AND INSTITUTIONAL CHANGE

The pre-colonial institutional system described above that embodied local knowledge and local economic and political interests has been drastically changed by political and economic changes in the colonial and post-colonial eras in what is today Zambia. The country came under control of the British, who passed it to the British South Africa Company (BSAC) at the end of the nineteenth century. The BSAC controlled the area until 1924, pacifying and governing the area by installing and using local chiefs as a basic method of indirect rule. Land was distributed according to the needs of settlers and the mining interests of the BSAC. On the so-called tribal land, installed chiefs were now given more power than the previous big men had had in the local system, including the power to collect taxes and to conduct criminal and civil courts. The colonial state took the rights to and control of common pool resources such as fisheries and wildlife out of the hands of local entities. The BSAC, and later the British colonial administration, allowed other ethnic groups (Lozi and Bemba) to move into the fisheries for commercial reasons to provide protein to the urban and mining areas. This happened first at the main river, as the Batwa were too few in numbers to mount any form of protest. At this time the Ila and Balundwe were not much affected by the commercial fishing activities and the use of nets. In the late 1950s, when catches began to decline, the government introduced technical and administrative measures and started to regulate fishing activities with ordinances and laws, demanding licences and specifying mesh sizes and closing times (Mortimer, 1965). But these changes included neither local rules nor local knowledge (Haller, 2007; Haller & Merten, 2008).

Local hunting activities had for a long time been viewed by the colonial government as a problem, and local people as poachers. Hunting ordinances had been developed since the 1920s, restricting local access, further banning local technologies and requiring the use of guns. Chila hunting was considered barbaric and destructive and was blamed by white scouts for the decline in lechwe numbers. However, other authors (see Haller, 2007) have argued this could not be the case, and that the introduction of guns might have had a lot to do with this crisis.

During the colonial period, it became clear that only the pastoral land was to remain in local hands, and that access to it was to be controlled by the local chiefs. Interestingly, however, the local notion of territorial rights−meaning who is allowed to get access to which pastures−was still respected among the ethnic groups. It should be noted that one major development was the alienation of land in the eastern part of the Kafue Flats, large pieces of which were allocated to settlers and to commercial farming, particularly for sugar cane production (Haller, 2007).

5. INSTITUTIONAL CHANGES AFTER INDEPENDENCE: DEVELOPMENT OF OPEN ACCESS TO FISHERIES AND WILDLIFE

Shortly before independence Northern Rhodesia was under considerable political pressure from local nationalists led by Kenneth Kaunda and other leaders from the African National Congress represented by an Ila activist, Harry Nkumbula. Flouting the colonial laws against, for example, fishing in excess and poaching was regarded as a symbol of resistance against the colonial power. However, after Kaunda came to power in 1964, and as Zambia gained independence, the government inherited the colonial policies, which they even strengthened by investing considerable financial resources and personnel because Kaunda had become extremely interested in conservation issues (Gibson, 1999). As a result, fisheries and wildlife were adequately managed in terms of financial resources invested in monitoring and sanctioning – but not without attracting criticism (Gibson, 1999). Two national parks (Blue Lagoon and Lochinvar) and a Game Management Area were established in the Kafue Flats to protect and manage wildlife (Chabwela, 1992a; Haller, 2007).

These were important developments in the area, but at the same time the government launched a campaign to increase agricultural production with a kind of green revolution programme aiming at increasing maize production with new seeds and fertiliser (Haller, 2007). Furthermore, government strategy to boost industries increased the demand for energy. This necessitated the building in the 1970s of two dams on the Kafue River, the Kafue Gorge Dam for hydropower generation and the Itezhi-Tezhi Dam to regulate the water flow to the Kafue Gorge Dam downstream (Balasubrahmanyam & Abou-Zeid, 1982; Chabwela & Siwela, 1986). These dams changed the flooding pattern considerably, making the dry season wetter and the wet season dryer, especially in the very dry years, and the area much more accessible from the outside (Chooye & Drijver, 1995; Haller, 2007).

From the mid-1960s the economy and urbanisation of Zambia depended largely on high prices for copper. However, after the decline in copper prices in 1975 the country faced an economic crisis, and debts increased (Anderson et al., 2000). Staff of departments of fisheries, wildlife, mining and agriculture lost their jobs and had to look for alternative sources of income, and the problem worsened with the introduction of Structural Adjustment Programmes (Anderson et al., 2000) from the early 1990s, during the second republic. While salaries and prices for copper and agrarian products were low, prices for cattle, fish and game were rising considerably. These price changes significantly affected the common pool resources in the Kafue Flats because fisheries, wildlife and pasture became highly valued and commercialised. In addition, it is important to note that, in response to these changes, resource access and resource use required more relevant institutional designs, bargaining power and ideologies to use the common pool resources effectively (Haller, 2007).

These developments are not unique to Zambia and the Kafue Flats. Since the 1980s many African countries have faced similar economic problems due to changes in prices for minerals and cash crops as well as increasing debts and structural adjustment programmes (Woodhouse et al., 2000). A comparison of the national economy of five African countries and related price developments in Mali, Cameroon, Tanzania, Zambia and Botswana supports this finding, with the exception of Botswana (Haller, 2005, 2009).

5.1 Changes in the fisheries

Fisheries in the Kafue Flats became one of the most important informal sectors of the country, attracting massive waves of seasonal immigration of commercial fishermen and fish traders (Muyanga & Chipungu, 1982; Subramaniam, 1992; Haller & Merten, 2008). Today one can find seasonal fishing camps of 900 households or more in the Kafue Flats, consisting mostly of young or middle-aged men as well as women from urban areas who are all engaged in fish trading. These camps have problems of poor sanitation and garbage disposal, leading to health problems. They suffer from inadequate security and alcohol abuse, and observers agree that law and order has broken down here (Haller, 2007; Haller & Merten, 2008).

For the most part, fishing has been severely destructive. Most fishing methods are illegal according to state law and can also be said to violate indigenous institutions. People fish with draw nets with an extremely small mesh size, or even with mosquito nets or agricultural shade-cloth. They use large draw nets set without boats or drive fish into small meshed nets by beating the surface of the water. They fish during the closed season between December and February, the time when the fish breed. The dismantling of local institutions and the introduction of government institutions has not been well monitored, and the financial limitations mean the regulations are not enforced.

If Ostrom's design principles, such as monitoring and sanctioning, are used as a reference point for robustness, the state institutions must be said to be very weak. The Kafue Flats fisheries have not been effectively regulated, since the country's weak economy has caused a shortage of staff and equipment. The fisheries have become an open access resource suffering heavy overuse. Furthermore, young immigrant fishermen make a good profit out of the local people's low bargaining power. The overriding argument is often that these migrant fishermen consider themselves to be citizens of the country and they think local people in the area have no right to tell them how to fish. The problem is that state institutions, which should enforce the rules, are absent (Haller & Merten, 2008). In addition, the collective fishing institutions, which were once important to the communities, are no longer adhered to, especially by young local people, who fish out the ponds and river sections before collective fishing is announced, because they want money for investing in grocery stores (Haller, 2007).

On top of this, the Ila and Balundwe, who have lost cattle due to a disease in the past, are now engaged in women's basket fishing, while some women in turn move to the fishing camps to obtain fish for sex from willing migrant fishermen (Haller, 2007; Merten & Haller, 2007) – a strategy that probably contributes to the high HIV prevalence (which in the absence of data from the area is estimated to be similar to the Zambian average of 14.3 per cent; Central Statistical Office, 2007; see also Merten & Haller, 2007).

These uncontrolled fishing strategies are signs of increased vulnerability and food shortages, especially among poorer families and households who suffer increasingly from the loss of access to common pool resources that are taken freely by more powerful users (Merten & Haller, 2008). This also reflects the general trend in Zambia, where the Human Development Index dropped from 4.9 in 1985 to 4.1 in 2002.1

5.2 Changes in wildlife management

The situation of wildlife management is very similar. Hunting is relatively easy for urban people, who can get a licence with less travel and financial difficulty than rural people (Chabwela, 1992a; Haller, 2007). According to local informants, commercial hunters bribe local scouts in order to be able to shoot more animals than the licence allows. The harsh restructuring conditions since the time of structural adjustments and the cuts in the state budget have meant that less money is available for wildlife conservation, and the underpaid game scouts are weaker in resisting corruption (Haller, 2007). This problem has been prevalent since the 1980s and attempts are being made to give some benefits from wildlife back to the local people. The Administrative Management Design for Game Management Areas programme was set up in 1983 to introduce a kind of participative approach to wildlife management. However, for several reasons the programme was not successful. In particular, its gains were quite insignificant compared with the losses of wildlife and the limited access to pastoral areas and fishing grounds. Furthermore, the programme lacked local involvement and its organisers failed to understand incentive structures (collective versus individual household incentives; see Gibson, 1999; Haller, 2007).

After the defeat of the socialist Kaunda government, the Chiluba regime semi-privatised the management of wildlife and conservation activities by restructuring the wildlife Department into the Zambian Wildlife Authority, which was now responsible for finding donors for conservation activities. The main aim was to involve local communities in Community Resource Boards. In the Kafue Flats these were initiated with the help of the World Wildlife Fund (WWF) Wetlands Project and the Partners for Wetlands Project. However, one of the main problems was the strong involvement of local chiefs as the representatives of local communities, which was a misunderstanding of local power structures. Most chiefs in the area were challenged by other potential leaders, following the logic of the big men structure of Ila politics. Revenues from protected areas such as Lochinvar were perceived by many local stakeholders as going to the chiefs, and the chiefs were accused of corruption. All of these participatory measures thus did not help, because immigrant hunters did not follow the law and because local people were not interested in the conservation of animals that they considered no longer their animals but government animals taken away by more powerful urban hunters. This again created an open access situation (Haller & Merten, 2006; Haller, 2007).

5.3 The paradox of open access and privatisation in pastoral commons

Institutional changes also led to the overuse of specific areas of the pastoral commons. However, the situation here is puzzling. The official figures show an increase in the number of cattle up to 1990, but research on the ground has shown that the number of cattle owned by households decreased dramatically due to the East Coast Fever (Theilerosis) epidemic from the end of the 1980s (Haller, 2007). In general, over 50 per cent of herds were lost to this epidemic; for example, people in a chichi in Mbeza now have only 60 per cent of the herds they had before the epidemic. One would therefore expect less pressure on pasture (Haller, 2007).

One of the outstanding problems facing pastures is the irregular flooding pattern caused by the dams and the invasion of foreign species of shrubs (e.g. Mimosa pigra), which have decreased the size of valuable pastures by 25−50 per cent (Haller, 2007). However, although the data indicate that the number of cattle and the size of the pasture have decreased to a similar degree, there are signs of local overuse that has to be linked to institutional change. A new national inheritance law from 1989 (the Succession Intestate Act; see Malungo, 2001) for the protection of the property of widows and children, and a local organisation called Law and Development Association (LADA) supporting this law (Haller, 2007; Merten & Haller, 2008), pushed for the privatisation of cattle among the Ila, which created a demand for private cattle camps. This in turn has reduced the mobility of cattle and confined animals to small unit areas for grazing. Such a change creates profits for some local leaders who are trying to claim more rent payments for cattle camps and have shown interest in getting leasehold titles for pieces of grazing land, based on the new Land Act of 1995 (Haller, 2007). However, there is a counter-trend to this sort of semi-privatisation involving rich people from urban areas: such absentee herd owners often try to get access to pastures by involving local people, who have free access rights to pasture as members of the community. In this way, sometimes up to 900 heads of cattle graze for some months on a limited area, which considerably reduces the carrying capacity of the pastures.

6. CONCLUSION

This paper has shown that when the colonial and the post-colonial state took control of the commons, local institutions were increasingly marginalised, which had a negative effect on the management of these resources. While the pre-colonial land tenure system included the regulation and management of pasture, fisheries and wildlife as a whole, since colonial times they have been managed by separate government departments. Unfortunately, colonial and post-colonial state institutions have not performed well in terms of monitoring and sanctioning. They thus do not satisfy Ostrom's design principles for robust institutions, reflecting mostly what economists call transaction costs (Ostrom, 1990), in the way the pre-colonial institutions did (Haller, 2007). Common pool resources in the Kafue Flats are therefore held in open access regimes or made into private property, which leads to overuse.

This consequence of an institutional change is similar to what Ensminger (1992) describes as changes in relative prices that stem from external changes in the economy, formal institutions, technology and demography and that influence local changes. In the case of the Kafue Flats, such changes, especially the economic crisis and the loss of state institutions, have led to high prices for common pool resources and local institutional changes. Due to the paradox of the state being both present and absent, and the ideology of citizenship (see Haller & Merten, 2008), immigrant hunters and fishermen have more bargaining power to change the previous common property regime to open access, from which they derive more profit. In addition, local powerful actors (men versus women, young men versus community) opt for open access. In pastoral areas, high prices for cattle lead to two trends that seem to co-exist: on the one hand, some leaders try to privatise in order to profit from high cattle prices; and, on the other, absentee herd owners try to use pastures in open access, with negative consequences for the sustainable use of the common pool resources. What is needed to achieve sustainability in the Kafue Flats is the participatory development of new institutional settings.

This special issue was produced with the support of the European Union's Sixth Framework programme through the Cross-Sectoral Commons Governance in Southern Africa Project No. 043982. This work does not reflect the Commission's views and in no way anticipates its future policy in this area.

Notes

¹Based on United Nations Development Programme data, compiled by the National Centre of Competence in Research (NCCR) North–South, Bern, Switzerland.