Biodiversity and Conservation

, Volume 18, Issue 10, pp 2563–2583

Community based natural resource management in Zimbabwe: the experience of CAMPFIRE


    • WWF Southern Africa Regional Programme Office
Original Paper

DOI: 10.1007/s10531-009-9612-8

Cite this article as:
Taylor, R. Biodivers Conserv (2009) 18: 2563. doi:10.1007/s10531-009-9612-8


Communal Areas Management Programme for Indigenous Resources (CAMPFIRE) is a long-term programmatic approach to rural development that uses wildlife and other natural resources as a mechanism for promoting devolved rural institutions and improved governance and livelihoods. The cornerstone of CAMPFIRE is the right to manage, use, dispose of, and benefit from these resources. Between 1989 and 2006, CAMPFIRE income, mostly from high valued safari hunting, totalled nearly USD$ 30 million, of which 52% was allocated to sub-district wards and villages for community projects and household benefits. Whilst a number of assumptions underlying the success of CAMPFIRE as an innovative model for CBNRM have yet to be met, CAMPFIRE confirms the concept that devolving responsibility and accountability for natural resource management can be highly effective for the collective and participatory management of such resources. Elephant numbers in CAMPFIRE areas have increased and buffalo numbers are either stable or decreased slightly during the life of the programme. However, offtake quotas for these two species have increased with a concomitant decline in trophy quality. Although the amount of wildlife habitat diminished after 1980, following the commencement of CAMPFIRE the rate of habitat loss slowed down and in some specific instances was even reversed. More recently there has been increased pressure on habitats and other natural resources as a consequence of deteriorating socio-economic conditions in the country. Where devolution has been successful, promising results have been achieved and the recent acceptance and implementation of direct payments to communities is probably the most significant development since 2000. That this has happened can be attributed to CAMPFIRE enabling communities to maximize their roles within the existing set of rules, and by so doing, allowing these rules to be challenged. Donor (73%) and government (27%) investments into the programme amounted to $35 million during the period 1989 to 2003. Since 2003 however, donor funding has been reduced to <$600,000 over the past 5 years.


CAMPFIREZimbabweWildlife useEconomic benefitDevolutionProprietorshipGovernance

Introduction and background

As a consequence of historical land apportionment along inequitable and racial lines, the majority of Zimbabwe’s population occupies marginal agricultural land (Murphree and Cumming 1991; Jones and Murphree 2001). Until very recently, a dual economy based upon large-scale commercial agriculture on privately owned titled or freehold land (alienated land), and subsistence agriculture on communal land characterised land use. In the latter, uncertain livelihoods lead to opportunistic risk aversion strategies, whilst the recent colonial past shapes the wider socio-political context in which natural resource management occurs.

Communal Areas Management Programme for Indigenous Resources (CAMPFIRE) was designed by the then Department of National Parks & Wild Life Management (DNPWLM, now the Parks & Wildlife Management Authority, PWMA) in the mid 1980s (Martin 1986). It is a long-term programmatic approach to rural development that uses wildlife and other natural resources for promoting devolved rural institutions and improved governance and livelihoods (Child et al. 2003). The cornerstone of CAMPFIRE is the devolution of rights to manage, use, dispose of, and benefit from natural resources.

Wildlife conservation in Zimbabwe is legally mandated over some 50,000 km2 or 13% of the country, in the State protected Parks and Wild Life Estate, situated mostly in the lower altitude, poor rainfall areas around the periphery of the country (Fig. 1). An equivalent area of 40–50,000 km2 of communally occupied land is either adjacent to or near the Parks Estate where wildlife populations remain relatively abundant, especially where human population density is low (<10 persons/km2) and wildlife habitat (>50% of land area) is reasonably well maintained (Taylor 1999). It is in these less developed, more remote areas that CAMPFIRE was initially implemented in the late 1980s.
Fig. 1

Map of Zimbabwe showing CAMPFIRE districts

Environmental setting

Much of Zimbabwe is semi arid, with a low and variable rainfall making the country prone to drought. Comprised predominantly of broad-leafed deciduous plateau Miombo (Brachystegia spp.) and river valley Mopane (Colophospermum mopane) woodlands and savanna, land use varies from intensive crop production to extensive cattle and wildlife production along a rainfall-altitude gradient. This is reflected in the agro-ecological survey of the country (Vincent and Thomas 1960) which identifies Natural Regions IV and V as unsuited to rainfed agriculture, and best used for extensive rangeland production systems. This is further supported by economic research (Jansen et al. 1992) and the conversion of land from livestock to wildlife (Price Waterhouse 1994), which has demonstrated, economic distortions not withstanding, that wildlife utlisation is a highly competitive form of land use in these drier regions.

Resources used or managed

As originally envisaged, CAMPFIRE was to focus on the conservation and exploitation of four natural resources, wildlife, forests, grasslands and water. Because wildlife is able to provide direct and immediate tangible financial benefits, initial success of the programme was premised on the utilization of large mammal wildlife resources, mostly through high value trophy hunting safaris. Largely because of its strong links with wildlife management, CAMPFIRE has continued to focus more on the consumptive and non-consumptive use of large mammals rather than other natural resources. Subsequently CAMPFIRE diversified its natural resource management activities beyond wildlife utilisation to include non-consumptive eco-tourism ventures, timber and bamboo harvesting, honey and fruit production, fisheries, mopane caterpillars and the sale of non-renewable resources such as river sand for construction purposes.

Institutions responsible for managing resources

In Zimbabwe, PWMA is the legally mandated authority responsible for wildlife management in the country. The 1975 Parks and Wild Life Act decentralised state authority, and conferred privileges on owners or occupiers of titled land as custodians of wildlife, fish and plants (Government of Zimbabwe 1975). Land owners or occupiers were designated “appropriate authorities”, giving them de facto responsibility for wildlife and making them the beneficiaries of sound wildlife conservation and use. After 1980, similar rights were extended to communal farmers through an amendment to the Act in 1982, which delegated Appropriate Authority (AA) to Rural District Councils (RDCs). In practical terms AA represents the decentralisation of authority and control over wildlife to RDCs only and not lower level institutions (Murombedzi 2001).


At the time when CAMPFIRE was designed, donor aid was uncommon in Zimbabwe and the early establishment of the programme was characterised by a relatively low level of external funding (Child et al. 2003). DNPWLM envisaged initial Government funding through a public sector investment programme (PSIP) to initiate CAMPFIRE, with these funds ultimately contributing to the creation of a rural sector economy based on wildlife production. Furthermore a CAMPFIRE Agency under an appropriate Ministry was also planned, for which short-term donor funding would have been sought (Martin 1986).

Technical and other support was provided by a coalition of support agencies, initially the University of Zimbabwe’s Centre for Applied Social Sciences (CASS), Zimbabwe Trust (ZimTrust), the World Wide Fund for Nature (WWF) and subsequently others, notably the Ministry of Local Government, Rural and Urban Development (MLGRUD). Their inputs were coordinated through the establishment of the CAMPFIRE Collaborative Group (CCG)1 under the leadership of DNPWLM and later, the CAMPFIRE Association. This support was further enhanced, notably for CASS and ZimTrust through NRMP I funding (see below) whilst that of WWF came through an independently funded project.2 With the onset of NRMP II funding in 1994, the CCG was replaced by the CAMPFIRE Service Providers (CSPs).


NRMP I (1989–1994)

NRMP I was fully financed by USAID through an agreement with the GoZ (Government of Zimbabwe), which provided for sub-grants to CASS and ZimTrust. The project was “…designed as a pilot initiative to test the CAMPFIRE hypothesis on a limited scale before committing more substantial USAID resources…”3

The total grant was USD$7.6 million over 5 years and four districts in Matabeleland in northwest Zimbabwe, namely Binga, Bulalima-Mangwe, Hwange and Tsholotsho, were recipients of support. This included infrastructural development, capital equipment, and activities relating to wildlife management, institutional and community development including women, training and applied research.

Under NRMP I increased community involvement and greater rationalisation for the alternative use of marginal agricultural land was achieved, and the potential for wider impact was demonstrated through project activities in the four pilot areas. Although there were some problems related to financial and administrative procedures, NRMP I was generally viewed as successful and USAID support for CAMPFIRE continued under NRMP II.

NRMP II (1994–2003)

USAID expanded its support to CAMPFIRE through NRMP II. The country level goal of NRMP II was to use NRM (Natural Resource Management) to develop economically sustainable communities on lands marginally suitable for agriculture. This goal was subsequently amended by USAID in 1998 after a mid term evaluation, as the Mission’s Strategic Objective One (SO1) namely “…strengthened NRM for the sustainable development of CAMPFIRE areas…”. Primary beneficiaries of CAMPFIRE have always been households at community (ward and village) level and were the intended ultimate beneficiaries of NRMP II. The CAMPFIRE Association (CA), RDCs, and CAMPFIRE Service Providers (replacing the former CCG) became the means for reaching these communities and as such, also became direct and immediate beneficiaries of the project.

A second related goal was Southern Africa Development Community (SADC) regional and in-country cooperation in the promotion of NRM activities, which would contribute also to the sustainable development of communities on lands marginally suitable for agriculture. The USAID contribution to NRMP II was USD$ 20.5 million of which USD$ 16 million was bilateral funding and USD$ 4.5 million regional funding. A 25% GoZ contribution in kind was estimated at USD$ 9.4 million (Child et al. 2003).


Initially directly and subsequently through WWF Norway, NORAD, the Norwegian Agency for Development Cooperation provided funding to WWF to support CAMPFIRE4 in two phases between 1994 and 2002. WWF’s support to CAMPFIRE was demand driven by both CA and DNPWLM and the nature of this support was clearly articulated in strategic plans.5 Specifically, WWF was charged with developing local level natural resource management techniques and capacity.

Phase I (1994–1998) included the development of natural resource management methodologies, using the concept and practice of Participatory Technology Development (PTD, Taylor and Bond 1999) and the development of training materials based on these methodologies. The funding provided for this phase amounted to USD$1,253,743.

Phase II (1999–2002) focused on the delivery of training nationally and locally, using the training materials developed in Phase I. Some USD$936,550 was made available for this work. Over a 9 year period, a total of USD$2,190,293 was provided to WWF for its CAMPFIRE work.

Other supporting funds of lesser amounts were also made available, including USD$113,000 for a specific component of WWF’s CAMPFIRE work, the development of quota setting methodologies, provided by the US Fish and Wildlife Service (USF & WS) through Safari Club International (SCI).

Since 2003, CAMPFIRE has received less than USD$600,000, approximately USD$100,000 annually from WWF, and the Ford and Kellogg Foundations.

Large, time-bound projects tend to focus on results or products, rather than the process needed in getting to the desired outcome. However, the allocation and application of funding resources should be strategically considered also, as frequently relatively small injections of support can be very cost effective. This combination of funding, as reflected above, has helped to achieve longer-term sustainability of programmes such as CAMPFIRE. Frequently, the time frame for most donor-supported projects is invariably too short for the needed behavioural and institutional responses central to the process of change.

Consequently, as a broadbrush overview, the substantial donor investments of more than USD$35 million over nearly 15 years can be considered relatively efficient when compared with the “return” on that investment as reflected in the gross CAMPFIRE earnings of USD$30 million over the past 20 years. A more detailed analysis of the allocation and application of this income as further described here outlines and questions its effectiveness and impact in an integrated conservation and development context.

Detailed description of programme activity

Originally designed to address those problems arising from communal ownership of natural resources, DNPWLM viewed CAMPFIRE as a programme for the long-term development, management and sustainable utilisation of natural resources, namely forests, grasslands water and wildlife in the communal areas of Zimbabwe. The programme would focus on the remote communal lands in natural regions III, IV and V around the periphery of the country (Martin 1986). Community participation would be voluntary, but custody and responsibility for NRM would be placed with participating communities. This was to be achieved through group ownership with defined rights of access to natural resources and appropriate institutions for the legitimate management, use and benefit of these resources.

In the process of implementation, three strongly inter-linked principles embedded in the original design have contributed significantly to the evolution of CAMPFIRE policy and practice (Jones and Murphree 2001):

Economic benefit

Firstly, economically competitive forms of land use have motivated the sustainable utilization of wildlife outside of formally protected areas. In the 1960s, wildlife policy moved from an earlier protectionist philosophy to one promoting the high economic and financial value of wildlife on alienated land in commercial agricultural areas, a key incentive for its sustainable management. The underlying assumption was that economics ultimately determine decisions regarding the allocation of land and the resources thereon. This may mean, however, wildlife being displaced by other more viable forms of land use. The early success of this utilitarian approach to wildlife provided compelling arguments for its wider application in the communal sector of the country, particularly after 1980. In the context of rural development and CAMPFIRE, placing wildlife in the realm of economics and land use, rather than conservation (Jones and Murphree 2001) provided an important opportunity to complement conventional and subsistence agricultural practice in the communal lands of the country.

Devolution and decentralisation6

This wildlife policy shift, formalized in the 1975 Parks and Wild Life Act, and amended in 1982, decentralised state authority and conferred certain privileges on occupiers of land. Such devolution, coupled with alternative economic opportunities and incentives for rural development, was intended, inter alia, to better serve wildlife conservation, given the inadequate government resources to do so. It also recognised that land occupiers are the primary determinants of habitat and wildlife status. Further and importantly, CAMPFIRE was viewed as a means of improving rural resource governance through fiscal devolution (Child et al. 2003).

This devolved responsibility was initially granted to 12 RDCs in 1989 and 1990, recognising that long-term success depended on further devolution to sub-district levels, even to a community level institution. The absence of any legal persona below the level of RDC, however, obliged DNPWLM to decentralise administrative authority and legal rights to wildlife to RDCs, but on condition that rights and benefits were to be further devolved to what were termed “producer communities”.

Collective proprietorship

Whilst the transfer of proprietorial rights, together with accompanying financial incentives, was highly successful on commercial farmland, similar replication in communal lands faced numerous legal and institutional impediments. What was required was a communal property regime behaving as a proprietorship unit over land and resources. Such a regime or unit should comprise a defined group collectively managing and exploiting common property resources within a defined jurisdiction (Jones and Murphree 2001).

In the event, Ward7 level producer communities emerged through the establishment of Ward Wildlife Management Committees (WWMCs) or Ward Wildlife Committees (WWCs).8 These village-elected committees were formally constituted with a membership comprising a Chairperson, Secretary and Treasurer and others, with or without a specific portfolio or responsibility. The Chairperson represented his/her Ward on the District Wildlife (or Natural Resources) Committee, a sub-committee of the district council. Although in effect, these new committees were sub-committees of the local government units (Murombedzi 2001), one of a number of constraints or difficulties faced was the perception of such committees as “parallel” institutions to WADCOs (Ward Development Committees) and VIDCOs (Village Development Committees), and thus potentially competitive, or even subversive.

Participation and implementation

In 1989 two RDCs in the Zambezi valley, Guruve and Nyaminyami, were granted AA status and commenced earning revenue through the marketing of trophy hunting quotas to an international safari hunting clientele. This was rapidly followed by a further seven districts wanting to join the programme and requesting AA status. By 1992, 12 RDCs had acquired Appropriate Authority and by 1996 there were 19 CAMPFIRE Districts (Table 1). The CAMPFIRE Association was formed in 1991 to lobby for, and promote the role of communal land wildlife producers. This gave CAMPFIRE an important level of political legitimacy and an ability to play a proactive advocacy role, locally and internationally. Its membership, however, has remained the RDCs and not the true wildlife producer communities at a sub-district level.
Table 1

Numbers of participating wildlife producing communities















No. RDCs














No. wards














No. house holds














Source: CAMPFIRE monitoring and evaluation database, WWF SARPO Harare

Following major donor inputs after 1996 for building or strengthening institutional capacity and NRM micro-project development, CAMPFIRE had grown to include 37 RDCs with AA status by 2001. Many of the latter, however, were not traditional wildlife producing districts.

Of these, 19 or 51% could be considered as fully participating, i.e. producer districts generating revenues for communities through sustainable natural resource management activities and receiving benefits in terms of funded projects, training and membership of the Association (Table 2). Over 70% of wards (271) and villages (1,217) in these 19 districts could be considered also as fully participating producer communities.
Table 2

Number of fully participating communities in CAMPFIRE

No. districts

Full participation


No. wards

Full participation


No. villages

Full participation











Source: CAMPFIRE Association

The CAMPFIRE model

In practice sport hunting and ecotourism have provided the primary economic and financial basis for the implementation of CAMPFIRE over the period 1989–2001. Although there is considerable biophysical and socio-economic variability between RDCs with Appropriate Authority, Bond (2001) describes a general model for the income or revenue earned from the use of wildlife and the subsequent allocation of this revenue (Fig. 2). Consumptive (sport hunting) and/or non-consumptive (ecotourism) rights to wildlife and wild land are leased to private sector operators by the RDC. The conditions of such leases, such as financial structure, duration and location are determined primarily by the RDC which negotiates these conditions with the safari operator.
Fig. 2

The CAMPFIRE model for income generation and allocation (after Bond 2001)

The number and types of leases vary according to the abundance of wildlife resources, the quality and diversity of scenic and/or cultural landscapes and market appeal. Most of the primary wildlife producing districts, however, have chosen to lease internationally marketed sport hunting rights to private sector partners because this has been the highest valued use to date (Cumming 1989; Bond 1994; Taylor 1994a). The decentralisation of AA has encouraged both the application of market-based mechanisms by RDCs and greater efficiency of resource use by safari operators (Child 1995; Bond 1999). Lessees pay all their fees to the RDC, but the level of involvement of sub-district community representatives in the lease allocation process, whilst variable between districts, generally has been minimal (Bond 2001; Jones and Murphree 2001).

The gross wildlife revenue earned is allocated to district council levies, district wildlife management activities and to wildlife producer communities as represented by wards. Whilst the breadth and depth of wildlife management activities varies between districts, most have a small core team of personnel who undertake law enforcement; problem animal management and wildlife monitoring (e.g. see Taylor 1994b). It is that revenue allocated to communities through Ward Wildlife Management Committees which is intended to provide the financial incentive for households to participate in the collective management of wildlife (Bond 2001). Wards choose to allocate revenue to management (salaries for resource monitors, allowances for committee members, fence repairs and maintenance), projects (grinding mills, schools, clinics, agricultural services) and household dividends (subsidised tillage, drought relief &/or uncommonly cash).



Revenue earned at district level from wildlife 1989–2006

Between 1989 and 2006 the revenue earned by Rural District Councils with Appropriate Authority totalled nearly US$30 million. Some 90% of this revenue was earned from the lease of sport hunting rights to commercial safari operators. The remaining revenue (Fig. 3) came from the lease of tourism rights (2%), ivory sales and sale of hides (6%) and other minor resources such as crocodile and ostrich eggs and firewood (2%).
Fig. 3

Income (USD$) earned from wildlife by CAMPFIRE 1989–2006

Annual income at the commencement of the programme in 1989 was US$350,000 when only two Appropriate Authority RDCs were in place and operational. This had increased to over US$2 million in 2001 by which time there were 12–16 wildlife producing RDCs with Appropriate Authority Table 1). In 1999, income exceeded $2,750,000 following a CITES-approved one-off sale of elephant ivory (Fig. 3). In 2006, income was over $US2 million, at which time only 15 (out of 23) wildlife producing RDCs were providing financial data, or indeed earning income.

Allocation of revenue earned from wildlife 1989–2006

Bond (2001) defines five categories for the allocation of revenue earned from wildlife by RDCs. Councils with Appropriate Authority are not legally obliged to devolve revenue to sub-district levels but are encouraged to do so through a set of guidelines. Originally developed by DNPWLM in 1991 (Anon 2003), the guidelines for CAMPFIRE have been the subject of on-going debate (Jones and Murphree 2001). Recently revised, the guidelines have now been endorsed by the CAMPFIRE association in its financial management manual (Anon 2003).

These seek to ensure that producer communities are the primary beneficiaries of the revenue earned and make the following recommendations. At least 55% of gross wildlife revenue should be devolved to ward level; up to 30% can be retained for wildlife management purposes at RDC level; and no more than 15% retained as a council levy. Apart from the council levy, the allocation of revenue over the past 18 years has been less than satisfactory in terms of the revenue guidelines. Even wildlife rich and well-endowed districts have been unwilling to devolve the recommended 50% of income earned to wards and households (Taylor 1994a). The declines in distribution to sub-district levels can be attributed largely to the declining economic condition of the country, associated exchange rate distortions and the unwillingness of cash-strapped RDCs to disburse scarce financial resources. Following a meeting called by PWMA to remind RDCs of their obligations, the proportion increased again in 2005 and 2006. This followed agreement in 2005 whereby safari operators pay the proportion due to the wards directly into ward level bank accounts.

Overall, 52% has been disbursed to community level over the past 18 years (Fig. 4). Some 14% of income ($US3 million) remained unallocated between 1989 and 2001, and increasing to $US9.1 million (31%) since, and matched by a dramatic decline in allocations to district level wildlife management activities. Unallocated revenue is generally assumed to have been committed to activities not related to wildlife and CAMPFIRE (Bond 2001).
Fig. 4

Allocation of income (US$) earned from wildlife by CAMPFIRE 1989–2006

Importantly, whilst there has been diversification beyond wildlife into ecotourism and other NR products, a number of RDCs have treated such income from these activities as General Revenue and not CAMPFIRE income. A study commissioned by CA (PWC 2001) suggests that these income generating activities may have provided as much as 30% more revenue than is reflected in the CAMPFIRE accounts (Child et al. 2003). Furthermore, most if not all of this income, fails to reach communities.

Conservation and natural resources

Wildlife populations

Following the commencement of CAMPFIRE in 1989, and as further RDCs were granted appropriate authority, increasingly more of the key wildlife districts were surveyed for large mammals (Taylor and Mackie 1997). Censuses have either been part of larger country-wide surveys or of specific CAMPFIRE areas. Consequently not all districts have been surveyed on a regular basis over the 14 year period 1988–2001. Table 3 summarises elephant and buffalo counts for eight of the 12 primary wildlife producing districts. The number of annual counts implemented in any one district varies from 4 to 12. Nevertheless, the results are adequate for the purpose of indicating broad trends (Fig. 5).
Fig. 5

Elephant and buffalo numbers for CAMPFIRE areas 1988–2001. Closed squares indicate elephant and open squares, buffalo

The estimated mean number of elephants was 6,840 (0.43 animals/km2), ranging from a minimum of 4,181 in 1989 to a maximum of 12,707 elephants in 2001. Densities are approximately half or less those found in State protected areas, but consistent with the maintenance of woodlands and healthy habitats for other wildlife populations (Taylor and Cumming 1993; Cumming et al. 1997a, b). These results also conform to increasing national and regional elephant herds, which have grown at an average 4–5% per annum over the past decade or more (Price Waterhouse 1996; Blanc et al. 2005).

Due to sampling difficulties, estimates of buffalo numbers tend to be highly variable (Taylor and Mackie 1997), nor are buffalo present in all wildlife districts. The mean number estimated was 12,042 buffalo (0.96 animals/km2). This is low for the southern savannas and numbers may have been influenced by a regional rainfall deficit, coupled to periodic droughts over the past decade or more (Foggin and Taylor 1996). As a general observation elephant numbers have increased in CAMPFIRE areas and buffalo numbers are either stable or have declined slightly over the past 14 years.

Quota offtakes and trophy quality

Child et al. (2003) show that the value of big game trophies has been maintained in CAMPFIRE areas, suggesting that trophy hunting and monitoring systems are effective. A review of quotas, offtake, trophy quality and “catch effort” across four key species, elephant, buffalo, lion and leopard, however, indicate that whilst national quotas and actual offtakes for elephant and buffalo have been increasing between 1992 and 2002, trophy quality for these two species has been declining (Grobbelaar and Masulani 2003). For lion and leopard, offtakes have either declined or are stable, whilst trophy quality is stable for leopard and increasing for lion. There is also a strong correlation between increasing quotas, declining trophy quality and increased “catch effort”.

There has been a noticeable shift in DNPWLM quota setting policy in the latter half of the 1990s. Whereas quotas were set to maximise returns prior to 1996, thereafter there was a switch in emphasis to more sustainable trophy quality, which resulted in a reduction of most quotas. Although attribution of causality is difficult, this shift may be linked to the growing acceptance and adoption of a participatory approach to quota setting and monitoring (Taylor 2001), successfully initiated in CAMPFIRE areas. This methodology emphasises the adaptive management of quotas in response to indices of animal abundance, trophy quality, community monitoring, illegal offtakes and safari operator “catch effort”.

Maintenance of wild land and wildlife habitat

1. As an initial assessment of the likely extent of wild land and habitat within CAMPFIRE areas, Taylor (1999) used wildlife producing wards9 as a proxy for this land, recognising that such wards comprised a mosaic of wild and settled land. On average wildlife producer wards made up 36% of the total number of wards in CAMPFIRE districts, with their land area of 39,580 km2 constituting 55% of the total area under the programme.

For 12 primary wildlife districts the amount of wild land varied from <500 to over 5,000 km2 with an average size of 3,300 km2 (Fig. 6). Of these, three districts had wild land in excess of 90% of the district area, six had 50–70% wild land and in three districts only, <35% of the district constituted wild land.
Fig. 6

Size frequency distribution of wild land in CAMPFIRE areas based on producer wards

The availability of wild land is negatively correlated with human population density (P < 0.01), with the maintenance of wild land (>50% of area) more likely under lower rather than higher population densities (<10 persons/km2; Taylor 1995, 1999). At a coarse scale of resolution, these results suggest that wild land has been maintained in a relatively intact state, albeit with variable levels of habitat fragmentation at different sites. This also has implications for potential household earnings from wildlife with those areas sparsely populated and relatively high wildlife densities standing to benefit more (Bond 2001; Murombedzi 2001).

2. Using a combination of aerial reconnaissance, aerial photography, resource inventories and field visits, habitat maintenance and productivity was assessed for three WWF Project areas by Conybeare (1998). In three wards of three districts in the Zambezi valley, loss of wild land and habitat over 8 years between 1989 and 1997 was minimal (Table 3), amounting overall to no more than 2% of a total 1,650 km2. The major threat was identified as population growth and demand for more agricultural land.
Table 3

Wildlife habitat loss in WWF project areas between 1989 and 1997 as reflected by change in size of ward wildlife area


Wildlife area 1989

Settled area 1989

Settled area 1997

Wildlife area loss










Guruve RDC









Wildlife corridor

North Gokwe RDC










Nyaminyami RDC









Source: Conybeare (1998)

3. Using a combination of aerial photography and remotely sensed imagery, Dunham et al. (2003) have recently examined more critically, the area and quality of wildlife habitat in selected CAMPFIRE areas. Quality was defined from the perspective of elephant cow herds which avoid not only (1) cleared patches of natural habitat, but also probably (2) a zone of habitat around the periphery of the cleared patch. Such patches can be categorised as “poor quality habitat”. A third category of habitat (3) may also be of poor quality even though sufficiently distant from settlement or cultivation to be unaffected by “edge effects”, if this patch is smaller than the minimum size of the home range of an individual or group. Finally, there are those patches (4) of good quality habitat currently unaffected by human activity comprising natural vegetation (at the time of the study).

These ecological consequences are dependent on the proportion of the area in question that is settled and/or cultivated and on the spatial distribution of such settlement and cultivation. For example, even if only a small proportion is settled/cultivated but widely scattered, a high proportion of the area falls into category (2) above.

Not withstanding the problems associated with using and comparing different methods of mapping (1:25000 black and white aerial photographs and LANDSAT satellite images), it was established for three Zambezi Valley districts that the percentage of habitat conversion through settlement and/or cultivation had increased markedly between 1981 and 1999. The percentage of poor quality habitat in both categories (2) and (3) while variable between years also increased but marginally so compared to the baseline year. The percentage of good quality habitat declined by half in Binga District and almost totally so in North Gokwe10 (Table 4), but not in the North Gokwe Wildlife Corridor (see Table 3). Only Nyaminyami District retained much of its original natural habitat. Natural habitat in both Binga and Gokwe was <50% coverage in the baseline year compared to Nyaminyami which was >50%. Importantly, the greatest negative changes occurred between 1981 and 1993, mostly before the implementation of CAMPFIRE. These changes were mostly associated with tsetse fly clearance (Child 1995) and associated aid-sponsored rural development programmes following Independence in hitherto unsettled areas of the Zambezi valley (Derman 1990).
Table 4

Summary statistics for three CAMPFIRE districts for which settlement and cultivation were mapped from aerial photography and remotely sensed imagery

Habitat quality defined from the perspective of elephant cow herds

District and year


Binga North

Gokwe North










Habitat quality

Defined as:


1. Habitat destroyed











2. Poor quality habitat

Natural vegetation

Inside 2 km buffer zone around settlement & cultivation










3. Poor quality habitat

Natural vegetation patch outside buffer zone but <57 km2










4. Good quality

Natural vegetation










Values are percentages

Source: Dunham et al. (2003)

Child et al. (2003) point out that no simple uniform or systematic approach to measuring wildlife areas over all CAMPFIRE districts has been undertaken and that this is a serious omission. Nevertheless, primary wildlife producing districts have been assessed, albeit at coarse levels of resolution and using imperfect methods. Dunham et al. (2003) also suggest that rather than focusing on human-wildlife interactions from the perspective of the problems that animals cause to people, further work is required on the effects that people have on the spatial distribution of wildlife, especially those species that contribute to CAMPFIRE revenues, notably elephant and buffalo.


Fiscal devolution and institutional development

From the inception of CAMPFIRE to the mid-90s, the amount and proportion of revenues devolved to producer communities increased rapidly, providing the primary impetus for wildlife conservation and for improvements in community institutional development and governance. Subsequently, the rate of devolution levelled off and after 2000, the process reversed itself (Child et al. 2003). By 2001, only 38% of revenue was being returned to producer communities with 20% being used for CAMPFIRE management and over 40% retained by RDCs for general purposes, compared to the guideline upper limit of 15%.

Nevertheless, in 2003 the concept and level of devolution in many districts was still strong. This is confirmed by Zimbabwe Trust (2003), which reported the strong correlation between fiscal devolution and institutional development. Through NRMP II investments, such as the CAMPFIRE Development Fund (CDF) and the establishment of community Trusts, these principles are being adopted in most projects as the norm rather than as previously contentious issues. There still remains, however, a high level of taxation imposed on producer communities by RDCs through their various levies. More recently, and in response to these adverse and imposed conditions, some wards and village collectives, notably in Chiredzi, Chipinge, Guruve11 and Nyanga12 Districts, negotiate directly with safari operators and other private sector partners, direct payments of hunting and ecotourism revenues. Some RDCs, especially their technical staff, tend to support such innovation, even if only implicitly, recognising their own limitations and inability to overcome this problem.

Child et al. (2003) comment that one of the more notable achievements of CAMPFIRE has been the strength of institutional development at the producer community level. The use of wildlife dividends appears to be decided democratically. People retain and sometimes can, but uncommonly, use their right to have household cash benefits and many projects are well implemented. Finances are reasonably well managed in a transparent and peer reviewed manner, thus preventing widespread or large-scale misuse.

In terms of governance, more than 100 democratically elected and constituted village and ward CAMPFIRE committees exist in 23 districts. These structures provide for a high level of community participation and decision-making with a transparent flow of information relating to key issues, planning and projects. These committees have been equipped with basic organizational skills including holding meetings, minute taking, book-keeping, and the fundamentals of project and financial management. In the primary wildlife producing districts, the community leadership and locally employed NR monitors are able to organize and implement a number of wildlife management skills as a result of workshops and training courses. These include counting wildlife, setting quotas, monitoring hunting, marketing wildlife and undertaking problem animal mitigation measures. Fire management has been implemented in the four districts of Chipinge, Chiredzi, Gokwe North and Guruve. Illegal activity is also monitored and penalties imposed on offenders. However, the basis for such achievement is inextricably linked to the incentive to do so which in turn, is directly related to the strength of the associated benefit.

Performance indicators

CAMPFIRE (and its equivalents elsewhere in southern Africa), confirms the concept that the devolution of responsibility and accountability for natural resource management can be highly effective for the collective and participatory management of such resources. Such devolution also leads to improved local institutions and governance. However, it is pertinent to examine also the assumptions or external factors underlying the success or otherwise of CAMPFIRE. Based on the “Monitoring and Evaluation Plan for CAMPFIRE” (Wright 1997), four areas of assumption and their indicators were identified as necessary for the successful implementation of CAMPFIRE. These are government policy, markets for natural resources, climate and macro-economic performance (Table 5). A review of these assumptions indicates that at least three are presently unfavourable for CAMPFIRE (WWF SARPO 2003a, b).
Table 5

Assumptions and indicators for CAMPFIRE (Modified from WWF SARPO 2003a, b)


Area of indicator



1. Continued government commitment to devolve CBNRM to local levels

Changes in policies for wildlife, forestry and other NRs

DNPWLM/PWMA have personnel and other resources for their CAMPFIRE Projects

Wildlife policy & legislation

Forestry policy & legislation

Land policy & legislation


budget (USD$)

No substantive revision of NR legislation or defined legal framework for devolved NRM in communal areas; Policy remains fragmented and largely centrally controlled

DNPWLM/PWMA budgets reduced from c. $200/km2 to c. $10/km2 over past 20 years

2. Favourable market environment for CAMPFIRE products &/or services exist &/or maintained

Demand for CAMPFIRE products and services maintained/expanded

The total number of days of sport hunting per annum

The gross value of sport hunting per annum

The trophy fees paid for key species at the Zambezi Valley auction hunts

Increasing number of tourists, including sport hunters and associated revenue declined dramatically with land reform-related political instability. Declines, followed by an increase since 2001 in trophy fees requires further analysis

3. No extreme climatic


Frequency and distribution of droughts and other extreme climatic events

Mean annual rainfall

No clear links between 9 years of below (5 drought years) and 5 years above average rainfall (3 flood years) and CAMPFIRE performance &/or land use change. Nevertheless, food shortages are common in most CAMPFIRE districts on an annual basis

4. Favourable macroeconomic environment prevails

Macro-economic performance maintained or increased

Changes in gross domestic product (1990–2001).

Changes in per capita gross domestic product (1990–2001).

Z$:US$ exchange rate (1990–2001).

Proportion of total population in formal employment (1990–2001)

Real wages (1990–2001)

GDP shrank from $8,4 m. in 1996 to $3.3 m. in 2001 with GDP/capita constant from 1991 to 1997 at c. $700. By 2001 GDP/capita had declined to c. $238

Between 1990 and 1997 ZW$ devalued from ZW$2.47 to ZW$12.44 and to ZW$55 in 2001 when it was fixed

Proportion of population in formal employment has declined from 14% in 1980 to <9% by 2001 with real wages < 50% paid in 1980

Firstly, the commitment of Government to creating and sustaining an enabling policy framework for devolved natural resource management has not been achieved through legislative changes. More recent policy changes, especially those from PWMA indicate a re-centralisation of wildlife management.13 Most NR and land legislation still continues to ensure state control of resources and land. Furthermore, under an adverse macro-economic environment, PWMA and other NR agency budgets have declined dramatically in recent years.

Secondly, although the markets for wildlife products appear generally robust, particularly on State and Communal Land, a collapse of the wildlife industry on former large-scale commercial farmland, although not yet impacting significantly on CAMPFIRE, does have implications for both State and communal areas as all three are linked, each adding value to the other. National hunting revenues peaked at US$22 million per annum in 1998 but have since declined to US$16 million in 2001 (Booth 2002). This is also reflected in the number of sport hunting days sold, declining from more than 20,000 in the late 1990s to 18,000 in 2001.

Thirdly and importantly, the macro-economic indicators examined all point to declining economic performance. The increase in unemployment and the decline in real wages act to place increasing pressure on land and other natural resources in the communal lands of the country.

Direct and causal links between rainfall and CAMPFIRE are difficult to establish. The long-term impacts of cumulative and variable rainfall deficits (>1,000 mm by 2001) experienced over the past 20–30 years remain difficult to predict. It is climatic variability; however, that provides one of the strongest justifications for adopting wildlife, and other NR-based land uses as an alternative and sustainable strategy for social, economic and ecological betterment.


To further promote the principles of CAMPFIRE as originally envisaged, specifically meaningful devolution over natural resource production systems to producer communities Murombedzi (1992, 2001), as opposed to re-centralisation of authority, CAMPFIRE needs to emulate more recent experiences in the region. This includes the need to legislate for appropriate authority status and land rights at sub-district units of decision-making, preferably at the level of the village. Promotion of local level proprietorship, including the establishment of community trusts and/or cooperatives must continue. Whilst capacity is not a limiting factor, long-term leadership must be strengthened.

RDCs still retain excessive control, especially revenue retention, resulting in the intended primary beneficiaries being severely disadvantaged. Given the poor macro-economic indicators and without the appropriate incentives, these producer communities are likely to continue or return to former unsustainable practices on marginal agricultural land, confirming the analyses of Cumming and Lynam (1997); Conybeare (1998); Dunham et al. (2003) reported above.

This is compounded by the CA still precluding producer community membership and limiting such membership to RDCs, thereby continuing to avoid addressing policy issues such as devolution of AA to sub district level.

Child et al. (2003) conclude that the greatest contribution of CAMPFIRE has been the lesson that fiscal devolution leads to improved rural democratisation, governance and NR management. As this paper has articulated, this achievement has been, and will continue to be embedded in, and constrained by a number of fundamental issues, mostly in the policy arena, and yet to be satisfactorily addressed. Nevertheless, where devolution has been successful, promising results have been achieved. Whilst communities are able to manage funds, implement projects and contribute to wildlife management, appropriate and strategic interventions by way of technical advice and guidance are still required. Building long term relationships and trust, together with back-stopping and reinforcement, is invariably needed for longer rather than shorter periods of time.


For a full discussion of the CCG, see Child et al. (2003).


The WWF Multispecies Animal Production Systems (MAPS) Project.


Eric Loken, USAID, pers. comm., quoted in Child et al. (2003).


See WWF Natural Resource Management Support to CAMPFIRE (SupCamp) Project Documents.


See CAMPFIRE Programme Strategy Workshop Report 1992, Hunyani Hills & SupCamp Project Documents.


Although the terms “devolution” and “decentralisation” are used somewhat interchangeably here, Murphree (2005) defines “decentralisation” as the delegation of responsibility and limited authority to subordinate or dispersed units of hierarchial jurisdiction, which have a primary accountability upward to their superiors in the hierarchy, and “devolution” as involving the creation of relatively autonomous realms of authority, responsibility and entitlement, with a primary accountability to their own constituencies.


In Zimbabwe, Provinces are made up of Districts comprised of Wards. Wards in turn comprise a number of Villages. These spatially and physically defined groupings also reflect the lower level administrative structures of the country, namely WADCOs (Ward Development Committees) and VIDCOs (Village Development Committees).


Members of the CCG (subsequently, Service Providers and mostly NGOs), worked primarily through the WWCs and WWMCs.


Producer wards are used as a proxy for the area of wild land.


Note that habitat assessment by Dunham et al. (2003) was for all of North Gokwe District whilst that of Conybeare (1998) was confined to the Wildlife Corridor, an area set aside by North Gokwe residents for wildlife.


See Ingwe Safaris 2005 Year End Report to Guruve Rural District Council.


See Minutes of the Gairesi Development Trust & the Nyanga Downs Fly Fishing Club, 2005/06.


See Statutory Instrument (SI) 26 of 1998.


Copyright information

© Springer Science+Business Media B.V. 2009