There are various measures of the relative significance of local government, in Zimbabwe which give conflicting signals about trends (Helmsing (1991)). There has been a steady expansion in local, as a proportion of total recurrent expenditures, whereas the trend with respect to public sector capital formation has been in the opposite direction (table 6). On the revenue side, local government revenue current receipts has accounted for a small and diminishing percentage of total government receipts (only about 6% in 1986) (see Table 6).
Comparison with some other developing countries (Table 7) shows that Zimbabwe LAs account for a relatively high proportion of total government expenditure, but raise relatively little revenue, and are heavily dependent on the CG for finance. Although local authorities in Zimbabwe have a narrow revenue base, substantial and rising current transfers from the centre have allowed them to undertake significant expenditures. But the virtual absence of financial autonomy has limited (indeed virtually removed) their ability to take decisions on priorities independently of the central government.
3.2.1. Regional planning exercise
"... the production of annual plans is increasingly becoming a barren ritual, more demoralising with each new year, as very few of the proposed projects in any one year are implemented" (Mutizwa-Mangiza 1990).
In 1984 the Prime Minister's Directive outlined a new system of development planning at the sub-national levels. Since then documents representing Five Year and Annual Development Plans have emerged from each of the eight provinces. These, in turn, were the product of planning efforts at the village, ward and district levels. The procedure was intended to bring about local participation in the choice and implementation of development projects. The guiding principles of the plan-making process revolve around maximising participation, integrating bottom-up and top-down decision making processes, and enhancing consultations among political and technical structures.
The aims of the planning exercise have been expressed in the following terms (by Mashonaland East PDC 1991): To translate the broad intentions of the Provincial and National Development Plans into specific projects which the Provincial Council would wish to see funded; and to assist sector ministries in selecting projects for which funds should be requested from the central government for implementation.
Clearly, if these objectives were fulfilled the system of government in Zimbabwe would have proceeded far on the path towards local participation in development planning with locally expressed needs and priorities being realised in the National Plan. Local participation in decision-making would extend well beyond the formal responsibilities and powers of the District and Provincial Councils, to the whole of development expenditure in the area. 4 The task here, therefore, is to assess the extent to which the ideals of the system have been met in practice.
It appears that the results have fallen far short of initial expectations, and that a large proportion of the projects planned at the local level are not implemented. This is a fundamental problem of the planning system, which is compounded by many other (related) problems, highlighted by Figure 1. The following paragraphs explore some of the main weaknesses of the system, from the perspective of achieving significant local participation.
The most basic problem is that essential links in the chain between plan formulation at the local and national levels, and between the planning process and public sector budgeting are weak or missing. This is shown in Figure Two. The participatory planning mechanisms permit local formulation of priorities to be reflected (albeit after considerable screening) in Provincial Plans. But there is no systematic way whereby the Provincial Plans influence the National Plan. Indeed, the timing of National Plan and Provincial Plans is such that any direct influence is impossible, since the Provincial Plans are finalised after the National Plan.
A further problem lies in the gap between national (and regional) planning and budgeting. 5 The head offices of the sectoral ministries are the agencies which participate in the annual bidding process for fund allocations from the Ministry of Finance. Although their representatives may participate in the sub-national plan formulation process, the sectoral submissions need not correspond to those put forward in the Provincial Development Plan. Even the government has acknowledged that "it is an established fact that many local and provincial projects are not generated in a simple bottom up manner, but are formulated in the head offices of sector ministries and parastatals" (GOZ 1990). It appears that most key decisions concerning the operation of sector ministries have continued to emanate from Harare, and that the impact of local ministry officials on programs and budgets is limited, so that the sub-national coordinating teams end up largely as 'talk-shops' (de Valk 1990 p.79). Because the provincial and district development committees largely consist of local heads of line departments, the top-down vertical linkages dominate. There are few incentives to encourage DDC members to work together since instructions, salaries and promotion prospects all come down from Head Office. There is generally no system for collecting data at the local level, and vertical (sectoral) allegiances apparently make it difficult for DAs to obtain the required information from the ministries (see Mudenge 1992).
The district and provincial development plans, which are the product of joint local-central efforts (in the respective development committees) have rather little connection with the actual budgeting and implementation of government projects and programs. For example, planning in the DDC & PDC is done on the basis of the calendar year, whereas budgeting is an entirely separate process which follows the fiscal year. The consistency or otherwise of provincial plans and sector ministry submissions varies. Although the secrecy of the bidding process renders comparisons difficult, the discrepancies are markedly less in certain areas - for example District Development Fund proposals6 - and worse in others. There is no legal framework to sanction ministries and departments that do not co-operate with the local authorities.
Thus Mashonaland East PDC complained in its 1991/2 Annual Development Plan of the lack of respect for the plan by the respective sector ministries who "always seem to have not only their own projects but also different priorities". How these priorities differ is examined in section 4.1 below. Figure 2 expresses this problem diagramatically, and shows the operation of parallel structures of decision making, up to the national level, where sectoral departments generally prevail. It is clear that the annual sectoral-based bidding process for central funds is the basic source of the problem. Whilst this is partly due to the operations of the sectoral ministries, 7 the situation has been exacerbated by inappropriate procedures. At least up until 1989, the destination of the Provincial Development Plans was uncertain. It was not even clear whether or not the sectoral ministries and Ministry of Finance actually received a copy of the provincial plans. It appears that the Provincial Development Plans remained in ineffectual limbo in the Ministry of Local Government. It is scarcely surprising, therefore, that these documents had little impact upon decisions about central government allocations and activities.
The achievements of the bottom-up development planning process have generally been disappointing. In many districts the DDC is "little more than a formality" - a forum for co-ordinating routine matters and airing views and problems, but with limited impact on the overall pattern of development in the area (Conyers and Mellors, 1989). A comprehensive survey of local attitudes toward the planning system found that the initial enthusiasm and commitment was rapidly waning - as reflected in disinterest in the process of project appraisal, poor attendance at DDC and Sub-Committee meetings and failure to hold such meetings (ibid). It has been observed in the Midlands that work by the PDC is hampered by continuing failure of the national government to clarify the role of the province in planning and resource allocation (PDSP, Sept. 1991). Clearly there is a danger of a self-perpetuating, worsening situation (as figure 1 suggests). The shift toward Strategic and Rolling Plans (discussed below) represents an attempt to avoid this.
Several PDCs have derived "performance rates" which represent central allocations as a proportion of total funds bided for. It was found not only that many project proposals in the plan were not funded, but many projects outside the sub-national plans were. Table 8 compares the allocations of three Provinces. There is a range of performance rates, with the Midlands obtaining a 78% success rate, far greater than Mashonaland East. Mashonaland West Province, in its Five Year Plan Review (1986-91), calculated that the projects implemented outside the Plan accounted for 114% of the provincial bid, whereas only 57% of planned projects were implemented. The Review described these figures as "shocking", and questioned the usefulness of the provincial planning process. Since the centre provides no justification of its project selection, provincial planners are unlikely to learn the reasons for their poor "performance rates" (e.g. whether the projects proposed were found technically deficient). In contrast, where district proposals are rejected by the PDC there is usually some indication as to why this was the case.
The lack of correspondence between local and national planning appears has been recognised by the government, which recently stated that "sector ministries have been requested to make sure that both national and provincial development objectives were reflected in their project submissions"(GOZ 1990). The main task of the newly created Regional Planning Division (within the National Planning Agency) is to ensure that provincial development priorities as well as projects submitted by provincial councils are effectively considered as part of sector ministry submissions. The RPD will also become a full participant on the PSIP budgeting Committees. However, the Ministry of Finance is not prepared to put forward financial guidelines on a provincial basis, on the grounds that the budgetary process would become unduly rigid. Such financial guidelines would lead regional planners to prioritise, and thereby permit more influence over development plans. Truly democratic local participation in investment decisions, however, would require financial devolution to the districts, via a block grant.
The problems do not lie solely on the side of the central government. The planning process has not been facilitated by the "submission of large lists of ill-conceived and poorly formulated projects" (GOZ 1990). In many cases, especially in the early plans, project requests were made without any detailed project abstracts, technical and administrative details or cost estimates. This was compounded by what is commonly referred to as their "shopping list" character. 8 Because of the absence of centrally determined financial ceilings, local authorities submit development proposals without any firm knowledge of prospective allocations. In some sectors, the number of projects funded seems to be proportional to the number of submissions made, 9 encouraging long lists of proposals. It is inevitable in these circumstances that there be an element of overbidding involved in project requests.
The situation is exacerbated by the absence of strategic planning among the provinces and districts. Projects are usually presented as if all have equal significance. At the district level, project proposals are not properly scrutinised partly because of underqualified and inexperienced staff (Eldon 1990). Some Provincial Development Plans (e.g. Mashonaland West 1991/2) expressly concede the lack of prioritisation. Elsewhere (e.g. Gokwe District Council), local planners have developed an integrated strategy with comprehensive prioritisation (through a numbering system). Even at the ward level, there have been efforts to improve local planning - e.g. in Mberengwa District in the Midlands, the wards have produced comprehensive ward profiles and agreement upon ward strategies and projects in priority order (see figure 3).
The production and submission of plans by the VIDCO to the WADCO, and then to the District Council is necessary in order to enable the community to express its needs to the upper echelons of the national planning process. But there are some problems in operationalising village and ward level planning. A recent survey found that village plans were only available for one-third of cases in the sample (MC&CD, 1991). Formal Ward Development Plans frequently do not exist. 10 Where they do, they are often rudimentary and without prioritisation (see figure 4). This is at least partly because there is no budget for planning, which makes it difficult to, for example, hold seminars, collect data, or even telephone (Dzinoreva, 1992). Examining the District Development Plans, one frequently finds complaints about the late receipt of ward submissions (e.g. Kwekwe District Annual Plan 1987/8). At the council level therefore, the ward councillor often brings his own list of ward needs or, worse still, sector ministry staff initiate proposals. There are exceptions however - as in Mbweregwa District, mentioned above. In some areas, training of councillors and village leaders is being undertaken.
Another practical defect of the present system lies in the uncertainty and delays which characterize the disbursement of central funds. Sectoral ministries often delay in informing provincial and district staff of proposed projects or budgetary allocations (Conyers and Mellors 1989). District Councils have complained that annual plans have to be prepared before the results of the previous year's bids are known (e.g. Shurugwi D.C. 1988/1989 Development Plan). Even after a project has received central approval, considerable time may elapse before the amount is actually allocated. There are complaints that disbursement does not follow a pro rate basis over the year, and that the timing of the release of funds is unpredictable. Where funding is not received until very late in the year, as has been the case in Mashonaland West for example, it is difficult to implement the planned projects within original cost estimates. The supply of materials for projects is also hampered by cumbersome central purchasing authority procedures (Midlands First Five Year Development Plan Mid-Term Review 1989, p.9).
A further problem for district development planning has been posed by foreign donors. Many donors bypass the District Council by working through the sectoral ministries, or going straight to the village level. This has not enhanced the position of the District Council as an institution coordinating development efforts in the locality. Not only can aid disrupt co-ordination efforts and risk duplication, but responsibility for maintaining the completed projects is often handed to the district councils, thereby stretching limited budgets and capacity (Conyers and Mellors 1989).
The local planning system has had some benefits, despite the fact that many projects in local plans are not implemented. The process has promoted co-ordination, stimulated local thinking about community needs and priorities, and helped to secure external (donor and NGO) funding in appropriate areas.
A number of observers (Conyers and Cormack 1986; Mutizwa-Mangiza 1990) have found that horizontal co-ordination at the district level has improved considerably over the past decade. At independence, local government and the central field administration constituted a dual system, which acted largely independently of each other. The planning tasks of the District and Provincial Development Committees have led to greater interaction, although experiences have varied across councils. Close observers have suggested that the personalities involved, especially the District Administrator who chairs and organises the DDC, are crucial. In some cases, good working relationships have evolved between the political and technical arms of the council. It is reported that a strong "spirit of co-operation" prevails at the provincial level in the Midlands and Masvingo. In contrast, recently serious conflict emerged in Gokwe District Council between officers and councillors as to their respective responsibilities. The matter was resolved through the involvement of the Provincial Administrator and Governor and training workshops for councillors on their roles and responsibilities, and resulted in a much more positive atmosphere for Council work (PDSP 1991).
Generally local planning efforts have generated substantial thinking about local development needs. The First Five Year Midlands Development Plan (1986-90), for example, is an impressive exercise in presenting basic socio-economic data on the province, and identifying sectoral problems and appropriate policies and programs. This in turn contributes to the strengthening of local institutional capacity to undertake monitoring, evaluation, and effective implementation. Similarly village plans, when available, in most cases clearly state problems, goals, and implementing methods (MCD&C 1991). Some local participants interviewed by Conyers and Mellors (1989) felt that the planning exercise put pressure on central government, and increased government awareness of development needs, especially in rural areas. A more directly tangible benefit has been the ability of District Councils to inform prospective donors of specific projects where local needs, estimated costs and so on have already been clearly identified. In this fashion, the "performance rates" of the District and Provincial Development Plans are sometimes improved by sources of funding outside the Public Sector Investment Program. Maungwe District Council, for example, has had remarkable success in securing donor pledges for project proposals which the central government had declined to fund, with the help of a local European volunteer working at the council.
Finally it should be noted that progressive improvements are being made to a development planning system which is still in its early stages. At the (sub)provincial level there has been a shift towards the production of Rolling Plans, which recognise the need for shorter and more realistic lists of project proposals to be spread out over a longer period of time (three years). This should also help in achieving consistency between one year's planned projects and the next. There have also been some attempts to develop Strategic Plans which collect and analyse data showing needs and potential in the locality and enumerate broad guidelines within which specific projects can be identified.
At the national level following the creation of the NPA (within the Ministry of Finance) (1988) (and within that a Regional Planning Division), Guidelines for Provincial Planning were published in 1990 which aim to improve both vertical and horizontal co-ordination in the planning system. 11
But the fundamental problem with the Guidelines' approach is the focus upon the province and neglect of the district level. It is asserted that "instead of emphasising plan preparation and investment co-ordination at the (sub)district levels, it is more effective to devote these human energies to the generation and formulation of better projects, and leave co-ordination and plan preparation to the provincial level" (p.28). The block grant proposed in the Guidelines (to amount to around 1% of PSIP according to the Minister of Finance) is to be allocated through the Provincial Councils on the advice of the PDC. The envisaged structure would not greatly further the devolution of power in Zimbabwe because the province is, to a far greater extent than the district, a creature of the centre, and only indirectly represents local interests. The Governor is not elected but personally appointed by the President, and the Provincial Council is elected only indirectly, consisting of District Council chairpersons and secretaries. Democratic institutions at the provincial level at present are thus weak, and central influence correspondingly powerful. Decentralization to the province will not, therefore, involve the transfer of substantial decision making power to locally elected bodies. One reason for maintaining central control over the province is that the provinces are, as one commentator observed, "large, powerful and ethnic...and potential threats to the unity of the nation state". More effective devolution would involve greater financial autonomy for the elected rural district Councils. The significance of the Guidelines is uncertain. But the focus on the province, rather than district for block grants appears to be the current government intention (Interview with Minister of Finance & Planning, November 1991).
3.2.2. District Development Fund
The main objective of the District Development Fund (DDF located in the MLGRUD) (enacted in 1981) is to "develop communal lands and such other lands as may be declared by the MLGRUD". It is financed through central grants, overseas development assistance, and revenues generated by hiring out equipment, ploughing lands etc. It is neither a full parastatal nor a direct government department. The Fund is involved in the development and maintenance of infrastructure at the district level, particularly roads and water supplies.
The Fund is charged with the construction and maintenance of primary and secondary roads in district council areas, whereas the Ministry of Transport is responsible for trunk routes. Several agencies are involved in construction of waterworks which then become the responsibility of the DDF for maintenance. To date DDF maintains 29,000km of rural roads and some 16 000 primary water supply points.
Project proposals for DDF funding are put forward through the local planning process. The DDF assists and liaises closely with the VIDCOs and WADCOs and the District Council and Development Committees (where it has offices) in formulating requests for projects.
Overall the DDF appears to allow much more local participation in decisions than most sectoral ministries. Project proposals are usually based on agreed district priorities. As with sectoral allocations, funding is allocated to the DDF, which then proceeds to divide the pool. It should be noted that the amounts involved in the DDF are relatively small - in 1985/6, total expenditure per person undertaken by the DC amounted to Z$ 33.48, compared to only Z$ 2.07 spent by the DDF.
The future of the DDF under the regime of the Rural District Councils Act is as yet unknown. It has been said that the Fund will become redundant (at least in its present form), and ought to be disbanded.
Compared to other developing countries, the government of Zimbabwe receives a relatively large share of national income (28%) in the form of taxation. This is mainly collected by the central government. The provincial level does not have any role in revenue raising (nor expenditure) while a relatively small share of public resources are mobilised by District Councils. References here to "local" revenue and expenditure are to District Councils. In 1986 local recurrent revenue represented 6% of total revenue (less than in 1980, when it stood at 9%). Given that local authorities represent 20% of total public expenditure, this indicates a correspondingly large role for central transfers. In contrast, IMF figures suggest that Zimbabwean local governments enjoy a large degree of financial autonomy, financing 59% of their own expenditures from local revenue (in 1986). However the IMF figure is an average calculated across the three types of local governments. The wealthier Rural and Urban Councils are much more financially autonomous than the DCs where the majority of blacks reside (table 9).
Urban Councils raise 80-90% of recurrent revenue from their own sources of taxes, fees, licences, tariffs and rents, with the balance coming in the form of central grants. The capital budget is usually financed through loans from both the central government and the Local Authority Pension Fund. Two councils, Harare and Bulawayo, are allowed to borrow on the open market and can issue stocks and bonds.
In stark contrast, in the District Councils locally generated revenue accounts for only 14.7% of local expenditure (table 10). Government transfers have risen from 50% (1980) to 73% (1986) of local government receipts (Helmsing, 1991). The District Councils are empowered to collect rates, charge fees and impose levies, but central approval is required for revenue raising activities, and local revenue sources are limited to the sale of beer, secondary school fees and charges on items like lease rents and business licences. The limited nature of local taxation in the District Council areas is partly due to the relative underdevelopment of the Communal Lands but it also reflects the ambiguous attitude of the central government towards increased local financial autonomy. Any new imposts or increases in levies must be approved by the centre, initially by the Ministry of Local Government and then by the Ministry of Finance. In the Midlands, for example, the PDC regards the constraints imposed on the district councils by central government as a "major limiting factor" (Midlands Province, 1989). Gokwe District Council has been unsuccessful in its bid to introduce a crop levy at a rate of $1 per $250 marketed output which would, on preliminary estimates, raise around Z$800 000 per annum. 12 The same council complains of the centrally imposed reduction in a local charge on bus operators using the town's bus station from Z$3.00 to Z$0.50, "making collection barely worthwhile".
The most important component of local revenue is secondary school fees. One study found that in 1984/5 school fees constituted 60% of local revenue in the country as a whole (Helmsing and Wekwete 1986). Substantial variations occur across districts.
There has been a sharp decline in the yield of the development levy (descendant of the poll tax which was collected by traditional leaders). The very low level of collection (table 11) is attributed to the perception of unfairness in levying an equal amount on every adult, and the belief that the money will be "wasted" on administration and councillors' allowances rather than spent on development projects. Moreover there are few effective sanctions against non-payment (whereas default on such fees as business licences leads to closure and eviction). Some DCs have started to require presentation of a development levy receipt before issuing birth certificates (e.g. in Buhera). There have also been allegations of failure on the part of VIDCOs to pass on the levy to the District Council. In some areas collection has been transferred back to the chiefs and headmen. There are early indications that this may be effective; in Gokwe, where the development levy had recorded a 25% collection rate in 1990/1 (amounting to less than Z$45000), kraalheads had, for the first two months of the financial year, collected over Z$183 000.
The government has emphasised income-generating projects (including beer halls, grinding mills, guest houses, poultry and other business ventures), as a means to increase local resource mobilisation. In practice it appears that this has not been a success in terms of revenue raising. Income generating projects accounted for less than 1% of District Councils' revenue in 1985/86. Even where a significant share of revenue is raised this way (as in Maungwe D.C. in 1989/90 where it represented 10% of income), there can be high running costs which have not been taken into account. Yet although the revenue effect may be limited, the net impact upon local economic and employment activity may well be positive (see Helmsing (1986)), and efforts may be justified on this basis.
Nonetheless District Council total revenues have increased significantly in real per capita terms - from Z$9.48 in 1980/1 to Z$32.64 in 1985/6 - and the contribution of local to total district revenues has risen (Helmsing 1991). This is partly due to a deliberate central cut-back, but also results from the diversification and higher yields of local revenue sources (Helmsing 1989). The major study on this topic, based on a sample of 11 district councils, found that the share of local revenue in the total had increased from 4% in 1980-1 to 15% in 1985-6 (Helmsing 1991). A case study of Gokwe, one of the better endowed and more successful district councils, found that locally-generated revenue, as a proportion of total local revenue, rose from 34% in 1983/4 to 64% in 1987/8 (Mutizwa-Mangiza 1990), and to 68% in 1990/1. There are significant variations between different District Councils (table 10 and 12). These figures do not capture the large extent of self-help contributed in both cash and kind, organised by the VIDCOs and WADCOs and directed towards the reconstruction of schools, improving rural water supplies and other local facilities (Muizwa-Mangiza 1990).
Overall, however, District Councils are heavily financially dependent upon the centre; there are large central grants in the areas of education, administration, health and roads, each of which are strictly tied to particular expenditures. The various sectoral ministries engage in an annual bidding process to the Ministry of Finance, and then transfer specific amounts to finance district activities.
A specific example of the type of tying involved can be drawn from the field of education. The central Ministry of Education administers grants to District Councils under the heading "Tuition". The Ministry has construed this term very narrowly, to encompass only those expenditures deemed to relate directly to classroom learning - i.e. textbooks, chalkboards etc. The District Councils have unsuccessfully argued that the grants should be available for such necessary expenditures as payment of clerks involved in local school administration. This proposition was recently rejected in the courts. Thus the discretion of District Councils with respect to the tuition grant is strictly limited.
In summary, poor economic resources, combined with very limited tax powers have led to low local revenue generation. The DCs are very heavily dependent on tied grants from the Central government.
Finally brief note is made here of two experiments where the degree of local financial autonomy has been significantly enhanced, a wild-life management project (CAMPFIRE) and an aid project (the Provincial Development Support Project in the Midlands).
CAMPFIRE involves the devolution of responsibility for and revenue from wildlife management to local "producer communities". This has generated substantial financial benefits. The CAMPFIRE program seeks to allocate earnings from wild-life management directly to the people involved, i.e. to the wards and villages under council jurisdiction. The District Council receives a levy for operating expenses of around 10-15% of total profits - the rest of the proceeds are spent in the localities affected.
The legal basis for the program lies in the conferral of "appropriate authority" status upon the District Council under the Parks and Wildlife Act 1975, with the rights to decide on how to utilise wildlife and to benefit fully from the revenues generated. By January 1991, twelve District Councils had been accorded this status, some of whom were receiving annual wildlife revenues in excess of Z$400000, which represented a fourfold increase in own revenues.
A fundamental principle of CAMPFIRE that "the unit of proprietorship should be the unit of production, management and benefit" (Murphree 1991), and that benefits must exceed costs in order for people to have an incentive to adopt sustainable wildlife management (the costs include damage caused by wildlife, refraining from crop production and appropriate land management). In addition the unit of proprietorship is to be as small as practicable, which effectively points to the ward as the highest level of government to be directly involved. In order to be granted "appropriate authority" the District Council must demonstrate its intention to delegate authority to producer communities and to promote the development of institutions that facilitate decision making and the just distribution of revenues at these levels. According to Murphree (1991), successful CAMPFIRE implementation has depended on the extent to which district councils have followed this directive. He also perceives a "centralising" tendency on the part of councils seeking to retain authority and benefit, limiting the role of producer communities.
There are a number of instances where there has been a fairly thoroughgoing devolution of decision making power to the local level under CAMPFIRE. In such cases the council may distribute the revenue equally among all the wards in the district, or on the basis of the site of the animals. The latter is in accordance with the input/benefit principle. Beitbridge District Council, for example, allocated wildlife revenues to three VIDCOs and two WADCOs within its jurisdiction. "Non-producing" villages and wards were excluded. Almost 60% of revenue went to Chikwarakwara village, where most of the safaris had occurred and where, for the first time, the local community was allowed to decide itself how to use the funds. In the other four Beitbridge cases the District Council consulted the relevant VIDCOs and WADCOs, then determined which community projects would be financed. In Chiwarakwara, the villagers themselves debated the most appropriate allocations. They decided to finance three community projects - a new grinding mill, the school, and repayment of arrears on the school building fund. In addition a cash dividend of Z$200 was paid to every household.
The Chikwarakwara case study presents an interesting example of division of responsibilities among different levels of government. Once the central government had granted appropriate authority status, the Beitbridge District Council negotiated a contract with a local safari operator, and undertook such functions as policy formulation, the co-ordination and marketing of wildlife, and training and awareness campaigns. The producer community (VIDCO) was responsible for (i) protecting the wildlife and day-to-day management, and (ii) deciding how to spend the revenue earned by the wildlife. Child and Peterson (1991) record significant consultation between the village and district levels. In this case the District Council provided technical assistance to the VIDCO in such matters as cost estimates for project proposals, but did not dominate village committee and community meetings. It is notable that the Province did not play any role in this process (and was not even informed of developments), a fact which was resented by the provincial leaders (PA and Governor) who were very critical of CAMPFIRE (Child 1991, p.22).
The CAMPFIRE project has succeeded in simultaneously meeting a number of objectives - including wild-life conservation and generation of resources for local communities. Although distributed very unevenly across the country, the revenues have been concentrated in marginal lands where harsh climatic and agricultural conditions inhibit agricultural activity, and has therefore been equitable in impact.
The MIDLANDS project: In the Midlands Province since 1989, untied capital funds have been made available (through the British ODA) to support local development projects being undertaken by Gokwe District Council. Gokwe district has a relatively buoyant local economy based on agriculture and the exploitation of natural resources (see Mutizwa-Mangiza 1992).
The overall objective of the program is to promote effective planning and implementation at provincial, district and local levels. Specifically the program aims to (i) strengthen the capacity of District Councils to plan and implement their own development programs; (ii) improve the capacity of provincial level institutions to support planning at the district level; and (iii) demonstrate the benefits of a decentralised planning, implementation and management system. To this end, there are two basic components: capital funds being made available to Gokwe District Council, and technical assistance at the district and provincial levels, working as facilitators in this process.
Progress has been monitored closely. There have been improvements in integrated policy planning, where Gokwe D.C. has produced a Three Year Strategic and Rolling Plan (1991-94) which, contains much shorter and realistic lists of projects in priority order and assesses implementation capacity. The Council has appraised alternative sources of local revenue, but further progress here has been blocked by the central government. The level of the block grant, which depends on council requests subject to local implementation capacity, has increased steadily over the three years. It presently represents a sizeable proportion (20%) of the total District Council budget. Giving the district greater control over project funds has, according to the District Administrator, generated considerable pressure for greater effectiveness, responsibility and accountability in financial management - so that after some early management problems, a system has developed to manage and monitor project expenditures carefully (Dzinoreva 1992).
An important thread of colonial rule, which has been maintained since independence, lies in the notion of physical planning. The relevant department is located in the MLGRUD, where it formulates statutory land use documents. Local authorities are empowered to regulate economic activity within their jurisdiction, through the regulation of land use. The primary objectives of the colonial regulatory framework were to protect the formal sector (white-owned businesses) against informal sector activities, and to ensure an orderly development of towns and districts which were clean and safe. It is argued below that these controls have stifled local economic activity, with particularly adverse repercussions for the poor.
5 This problem is becoming increasingly recognized; see, for example, Proposals for Strengthening Development Planning, Midlands Province, Nov. 1989.
6 The District Development Fund (DDF) finances certain items of local infrastructure - seep. 24.
7 Virtually every review conducted by the provincial and district levels stresses this point. As one participant stated, sectoral ministries should ensure that "the projects submitted through their own ministries are the same as those included in the provincial plan" (P.A. Manicaland, Parliamentary Workshop on Regional Development, 1988).
8 Thus, for example, Manicaland's proposed provincial expenditures in 1988 exceeded Z$1billion, which is greater than the entire national budget!
9 As observed by Mashonaland East, Provincial Five Year Review, 1986-90.
10 Or are only produced on a very irregular basis. In Maungwe in late 1991, the District Administrator was relying upon Ward Development Plans which had been produced at least five years prior.
11 The Guidelines propose improvements in the links between provincial planning and public sector expenditure allocation.
12 The concept of crop levies has even been agreed with farmers organizations, on the condition that the revenue be spent on improving the rural road network.