DEVELOPMENT INSTITUTIONS RESEARCH AND THE DEVELOPMENT BANK OF SOUTHERN AFRICA Mike Muiler Introduction To suggest that social research will be needed to support the transformation of South African society is to state the obvious. What is less obvious is how such research should be structured, in particular by whom it should be commissioned and managed. The purpose of this paper in reviewing the past, present and future of research in the Development Bank of Southern Africa is to offer a real world case for discussion. Much of DBSA's research to date has been done in support of the policies of its sponsors. As such policies are abandoned, it might be expected that the work done would be of little value if not downright wasteful. Yet there are examples of research which has been both useful and even potentially 'transformational' emanating from DBSA. In analysing the reasons for this, an important theoretical point is highlighted. There is a growing consensus, within the physical sciences at least, that research and development generates more benefits to society (or at least to the economy) when it is conducted close to the operational processes . The same might be expected of the social sciences. Indeed, the current fashion for Essential National Health Research , which seeks to focus on the operational processes within health systems rather than on biomedical issues, derives from a similar analysis. In this context, the operational field of the DBSA is a critical one for the transformation debate. The formal objective of the organisation is to promote redistributive development with a parallel emphasis on 'human development'. This is done through the targeted use of development finance which requires an intimate involvement in the operational side of 'development'. As we move from a focus on conflict through the problematic of transformation to the issues of development, this operational involvement will become more important. It will be argued that, because of its operational base, an institution like DBSA might in future be an important research player. Whether there is a role for DBSA as such will depend on whether a future government decides that an agency for development finance is needed and whether it proves possible to transform the existing institution's ideological base. The Development Bank of Southern Africa Some background details about DBSA and the organisation of research within it will, together with two case studies, help to illustrate these points. The DBSA was established in 1983 in response to ideas presented at the Carlton and Good Hope Conferences of government and business. It was an integral part of the regional TRANSFORMATION 18 (1992) *3 DEVELOPMENT INSTITUTIONS MULLER development policies of the time and as such was structured as a 'multilateral' regional institution of which South Africa and the four 'independent' TB VC states were members. Control is placed in the hands of a Council of Governors in which the majority of members are from the South African cabinet or its agencies. Routine control erf management is however carried out by a Board of Directors which is representative of business and other interests. Funding for DBSA came initially exclusively from the SA Treasury (apart from nominal share capital held by the TBVC members). These grants, which currently amount to R500 million per year, were further supported by loan repayments which began in 1986 and by borrowings from the local capital market in 1990. Loans were initially made almost exclusively to homeland governments and their agencies to finance infrastructure (roads, telecommunications, water resource and energy projects), business development (through Development Corporations) urban development and agricultural development (again through agricultural development corporations). Latterly, urban development projects in Black Local Authority areas have been financed as well as a few relatively small projects involving non governmental organisations. As a Development Bank, DBSA sees its role as more than simply a lender. Its business is stated as being 'to mobilise and provide loan finance, technical assistance and advice for sustainable development projects' but also, in addition, it 'has become increasingly involved in economic reform issues pertinent to the environment in which it operates'.3 This has included the provision of secretariat services for the 'Joint Financial Advisory Committees' (JFACs) which negotiate South African government budgetary contributions to certain of the TBVC/SGT governments. It has also encompassed facilitation in local government negotiations (particularly in Soweto and later the Central Witwatersrand Metropolitan Chamber) as well as involvement in the land debates as will be described below. The Role of Research Within DBSA In this context, it is possible to review the role of research within DBSA. First, the resources available for research can be identified. The use of loans raised in the market to 'gear' the grant income now enables DBSA to lend approximately R1000 million annually at subsidised interest rates. The income received from loan interest and other investments further enables DBSA to pay its staff related costs (over R70 million if travel costs are included) as well as to spend nearly R4 million on technical assistance grants and nearly R6 million on consultants' services annually4. It is the staff resources together with the consultants and technical assistance budgets that provide the DBSA with its research capacity. Research was initially a high profile activity with a separate Institute of Develop- ment Research whose director was one of the four members of the 'top management' team. Theresearchdone was either that requested by client institutions or identified internally. It was carried out by DBSA staff and external consultants and researchers 44 TRANSFORMATION 18 (1995) MULLER DEVELOPMENT INSTITUTIONS More recently, responsibility for research activity has been brought back into the main structure of the organisation. A Centre for Policy Analysis which is part of the Policy and Strategy Complex of the Bank carries out the day to day administration of the research programme although research can be commissioned from - and executed by - any operational group. With this change, the focus was placed more on policy analysis than on research per se. The areas which have been highlighted by the organisation as important contribu- tions include: • Regional Development Policy (DBSA managed the research and policy formulation process, which has led inter alia to the revised approach to industrial development incentives, on behalf of the SATBVC Development Council of Ministers); • Urban Development and Housing (as part of DBS A's move into the funding of projects in black urban areas in 'white' South Africa, this has been a consistent focus and has included inputs to the 1986 White Paper on Urbanisation and the current de Loor Commission on Housing); • Land Reform (see below); • 'Poverty Policy* (including submissions to the government's Central Economic Advisory Services on approaches to increasing the labour absorption capacity of the SA economy and to the Calitz 'Poverty Committee' on various aspects of strategies for the reduction in poverty. In addition, there is the separate area of 'data research'. While the collection and collation of data has been largely for internal use, there has been a specific focus on bringing together data from all participating states on a comparable basis as well as an emphasis on collecting data by economic region rather than by state. Further, because DBSA has a mandate to consider economic development of South Africa as a whole (by virtue of the fact that its 'member states' coincide with' 1910 South Africa'), it has been able to collate information on a national basis which the formal agencies of the state were unable to do. This serves a valuable planning purpose to all who need planning data about South Africa as a whole since this is not available through normal statistical channels. Prioritisation and Allocation of Research Resources As in most large institutions, two processes run in parallel within DBSA with regard to the prioritisation and allocation of resources. There is a formal 'assignment' process by which research proposals generated from either an operational sector or from an external client are evaluated and the necessary resources assigned. As with the formal development funding process, this can be extremely cumbersome with, in some cases, more resources devoted to the process than to the final research. This has promoted the second, informal, process. A certain amount of ad hoc work is done on individual or departmental initiative without any formal screening or approval. Further senior management frequently calls on staff resources in particular, but also on external resources, when there is a need for urgent work on specific issues. TRANSFORMATION 18 (1992) « DEVELOPMENT INSTITUTIONS MULLER Even within DBSA, there is in consequence little transparent allocation of research resources or transparency regarding the actual use of resources since much of the activity is globalised under departmental staff budgets. Attempts are being made to introduce a more structured approach. Since this will inevitably entail both a greater control over the use of resources as well as transparency as to their use, there is a natural bureaucratic resistance and the process is proving to be a lengthy one. The case studies presented below do not deal with the allocation issues but rather relate back to the fundamental question of the type of research that is conducted and its potential to contribute to transformation in South Africa. Setting Apartheid in Concrete? - A Demographic Case Study Basic demographic data is a key planning tool for social and economic develop- ment So the publication by DBSA last year of a formal projection of the growth of the South African population was important. The document was, according to the introduction, intended to support government and private institutions 'who have been depending heavily on demographic estimates and projections in planning for social and economic development'. When the report was launched it was emphasised that it covered the whole of South Africa (i.e. including the TBVC states) although the formal title still refers to 'Southern Africa'. The inclusion of all South Africa's people in such projections is, as noted above, vital for economic planning. The freedom to address the real South Africa has however been considerably constrained in the past and these constraints are reflected in the document Its projections make the remarkable assumption that the process of eviction of black South Africans from 'white' South Africa is to continue. The spatial distributions presented have been calculated using a methodology that takes the homeland growth rates experienced during the 'bad years' of forced removals and extrapolates them, albeit at a declining rate, until 2035. What are we to make of this? Does it suggest that the DBSA, widely recognised as among the more enlightened members of the state's constellation of institutions, is planning to set apartheid in concrete in the 21st Century? The answer is (we hope) more mundane. It is not that DBSA is part of a conspiracy to maintain apartheid nor that forced removals remain on some hidden agenda. This example simply reflects the extent to which social research is guided by institutional ideologies, be these overt or implicit. What this means is that the technical demographer, in applying the formal tools of his trade, may reflect old ideologies until such time as the conse- quences of change are specifically incorporated in the formal methodologies used. In this case the consequence has been that an initial opportunity to produce projec- tions that will help plan the development, oreven the transformation, of South Africa has been missed. More important however, the technical basis that has been laid will make it relatively easy to correct the initial assumptions and to generate a useful product in the future. 46 TRANSFORMATION 18 <1992) MULLER DEVELOPMENT INSTITUTIONS Agriculture and Redistribution - A Transformational Input? If the demographic work can be caricatured as an attempt to set apartheid in concrete, it should be contrasted with DBSA's contribution to the land debate. This received most attention through a paper presented (overseas) under the title 'The Potential for Black Smallholder Farmers Participation in the South African Agricul- ture Economy This paper represented a radical break with the past. It essentially presented a vision of South African agriculture within which black smallholder fanners would play a major role. Points of key importance were that: • it was assumed that agriculture in South Africa had to be restructured; • it was argued that black smallholder farmers could be at least as efficient as white commercial fanners; • it identified nearly 8 000 000 hectares of land that could be made available for black smallholder farmers including controversial categories such as 'indebted land' currently held by white farmers; • on this basis it was argued that at least 500 000 smallholder farmers could be accommodated. While the proposals can and have been criticised on a number of grounds (it has been argued for instance that what is on offer is largely more marginal land), the paper was important. It brought together data on land use and productivity that was not generally available. The conclusions it drew from this data represented a major break in approaches to land and agriculture for an agency of the South African state. This can be demonstrated by the extent of criticism that was received from traditional agricultural interest groups, the farmers' unions in particular. A revised version of the paper has subsequently been produced in which the focus on land transfer mechanisms was reduced and the numerical data on small farmer productivity and total farmer numbers and land areas involved were omitted. Among Uie key factors which produced an environment in which this policy research was produced were: • the failure of homeland agricultural models based on large farms and settlement schemes and the consequent development and introduction of the 'farmer support programme' model by DBS A in its homeland area of operations which follows closely decades old approaches elsewhere in the Third World ; • an acute awareness of the structural nature of poverty and unemployment and the extent to which it was growing ; • the structural crisis in white agriculture and its relatively small contribution to the national economy coupled with an awareness that current policies were failing to exploit its potential to generate important benefits for the broader society ; • the political primacy of the land issue in South Africa; • the strong sense of exclusion from international practice and experience of small farmer development (which is proving relatively successful elsewhere in the region with notable successes in Zimbabwe and pathbreaking theoretical research in Swaziland) and the related need to make a credible contribution in an international forum. TRANSFORMATION 18 (1992) 47 DEVELOPMENT INSTITUTIONS MULLER It might prove interesting to examine in more detail why these particular pieces of research were done at this juncture. For the purposes of this paper however, the intention has simply been to focus attention on an example of useful policy research that has emanated from DBSA. Comment It would be easy to dismiss DBSA sponsored research as the increasingly irrelevant product of an apartheid institution. That would however be too simple as the case studies demonstrate. In the current period, South Africans can - and must - begin to look beyond the short term conflictual approaches to the longer term needs of transformation and development This does not imply that there is no longer conflict in South African society. On the contrary, there are many clear conflicts of interest that need to be resolved. But these conflicts are becoming increasingly transparent. And as the negotiating fora within which they can be identified and mediated begin to emerge, there is a growing need to provide the technical basis - starting with simple demographic and economic data - on which to negotiate. The identification, prioritisation, resource allocation to, and management of, the necessary research thus becomes a critical issue. Much energy has gone in to the establishment of independent research capacity, perhaps more than has gone into research itself (hopefully, a useful lesson about the inefficiencies of the aid process). This has been necessary while access to information has been ideologically con- strained. The disadvantage in the present political context of the proliferation of research structures does not simply lie in the wastage of resources. It is often that the products are not accepted as anything more than reflections of the ideological position of their sponsors. In the field of policy formulation, there will always be a need for each interest group to maintain independent analysts. This does not imply however that there is no scope for cooperation in research to provide the base data on which analysis can be done. It could be argued that too much analysis and policy formulation is currently being done on too small a data base precisely because there are not structures to promote cooperation in conducting the basic research which must underlie the process. There would seem to be a clear and immediate need to begin to match the information requirements of what is still an extra-parliamentary opposition with the data and research resources of the state institutions. Control of these is after all the long term goal of the democratic movement and establishing the research agenda should be an early part of the process of establishing the policy agenda. In an activity like research which is necessarily formalised and structured, it would surely be possible to devise appropriate mechanisms of management acceptable to all parties. In the longer term, it will be necessary to review the status of existing institutions of which the DBSA is a case in point DBSA is a state institution, a member of the family of parastatals (with agencies such as the IDC) through which the state has <« TRANSFORMATION 18 (1992) MULLER DEVELOPMENT INSTITUTIONS historically intervened in the economy. As such, should it be decided in the course of negotiations and after that there is a role for a development finance institution like DBSA, the key issues will be those of setting its new objectives and establishing acceptable systems of priority setting, management and control. That will be the main item on the agenda. But the debate and the decisions need to be informed by the requirements of transformation. In that context, the need for research and the potential 'value-added' of an institution like the DBSA must be recognised. There are few other organisations whose brief is to view the heights of the economy through development oriented glasses while keeping its feet firmly on real-life multi-sectoral ground through financing operations which are guided by social objectives. As indicated in the introduction, it is this combination that has the potential to contribute to the process of transformation and development; it is this that should guide any assessment of the value of an institution like DBSA as a focus for research. Social research, as in the physical sciences, is likely to be more productive for society if it is organically linked to the operational processes that generate research issues. There is an immediate need in South Africa to establish structures which could promote research that would be useful and acceptable to all parties engaged in the various negotiation processes. In the context of the longer term transformation debate, it is also important to highlight the need for crosscutting intersectoral research driven by social objectives. These conditions suggest that a development finance institution such as the DBS Acould play a useful future role in the generation of research issues and the promotion of the research process. Such a role would however depend on prior agreement as to the need for such a multi-sectoral development finance institution and the establishment of acceptable control struc- tures for its overall activities. NOTES 1. See for instance the discussion in CSIR, 1991 which refers to the work of Rosenberg and Kline in relation to linear 'science-push/market-pull' versus 'chain-linked' models of technological innovation and Jon Tumey (1991) 2. See Commission on Health Research for Development (1990) 3. See DBSA (1991a) 4. 1990/91 figures ex DBSA (1991a). 5. See DBSA, (1991b) 6. See Oiristodoulou NT and Vink N, (1990) 7. See Brand SS, Christodoulou NT, Van Rooyen CJ and Vink N, (forthcoming). 8. Compare, for instance, my article for the Manchester Guardian (25.7.1974) which reviewed African experience in helping small farmers and the description of the fanner support programme in van Rooyen, Vink and Chnstodoulou, (1987) 9. See Ugthetn and van Niekerk, (1990) 10. See van Rooyen J and MatchetcheC, (1991) REFERENCES Brand SS, Christodoulou N X Van Rooyen CJ and Vink N, -'Agriculture and Redistribution: A Growth with Equity Approach', Development Southern Africa (forthcoming). Christodoulou NT and Vink N, (1990) -'The Potential for Black Smallholder Farmers Participation in the South African Agriculture Economy', Conference on Land Reform and Agricultural Development, (Newick Park Initiative) Commission on Health Research for Development^ 1990) -Health Research : Essential Unk to Equity in Development, (Oxford University Press) CSIR, (1991) -Science and Technology Policies and Economic Development DBSA (199 la) -Annual Report. TRANSFORMATION 18 (1992) <9 DEVELOPMENT INSTITUTIONS DISCUSSION DBSA (1991b) -Southern Africa - Population Projections 1995 - 2005 Iigthelm and van Niekeik, (1990) -'Unemployment: The Role of the Public Sector in Increasing the Labour Absorption Capacity of the South African Economy', Development Southern Africa, 7 (4) November 1990. Tumey, J (1991) -'What Drives the Engines of Innovation?'in New Scientist, 16.11.91 Van Rooyen J and Matchetche C, (1991) -'Approaches to Analyse the Agricultural Sector's Role in Regional Development in South Africa', (presented at) Agricultural Economics Association of South Africa Conference, (Port Elizabeth) Van Rooyen, Vink and Christodoulou, (1987) -'Access to the Agricultural Market for Small Farmers in Southern Africa: the Fanners' Support Progamme', Development Southern Africa 4 (2), May 1987. DISCUSSION FIROZMANJI: Perhaps if I may be permitted to speak to you as somebody inside a development agency, with an uneasy experience that money talks. Development agencies with funds in their hands can not just influence but I suggest set the agenda. And I think this is the issue which I think can be useful to try and look at, saying, 'Where is the agenda being set, by whom, for whom, for what?' MIKE MULLER: When we achieve a political structure wherein people are relatively happy with the institutions and their objectives, a good way to organise research is in fact to locate it close to the operational processes and not necessarily in some committee, be it in Ottawa or be it in Nairobi or be it in Pretoria. MALE VOICE: It seems to me the issue that the role of the state and private sector research institutions in the transition period needs to be specifically addressed. Because there's a lot of energy in these agencies going into restructuring these institutions, in some ways in more progressive directions, but in many respects merely aimed at ensuring their own reproduction. It seems critical, from an outsider's standpoint, that in this period, among the issues that should be pushed quite actively are opening these organisations up to external scrutiny. OK. Let's see what you're funding, what you're doing. How willing are you, in fact, to redirect work? MIKE MULLER: If you just look at ideology at the moment as change without pain, that these institutions can be transformed and somehow we will also manage to reproduce the glitzy people without too much damage to them. RENE LOEWENSON: I'd just like to take the opportunity to say a little bit about Zimbabwe and the role that was played by the Whitsun Foundation in the mid seventies which undertook research in agriculture, in finance, trade, investments. These statistics became the basis for government policy to transform. Some of the private sector concerns, which are being echoed in here, were also present in the Whitsun Foundation prior to Independence in 1980. If you read some government policy documents two years hence, five years hence, after Independence, word for word they repeat the words of the Whitsun Foundation. I would submit to you that the research that was done by the Whitson Foundation in that period was wholly on transformation. But in the end, as we know, if anyone knows about Zimbabwe, the land experience and rural development problem has been wholly untransformation- al. 50 TRANSFORMATION 18 (1992) DISCUSSION DEVELOPMENT INSTITUTIONS MALE VOICE: I think we are losing a central point, that there's a role for the state - the heading of this session was the transformation of the state institutions. Who is actually going to fund more general theoretical research. We haven't addressed that question this whole session. Or is it just going to be the universities, and who's going to fund that? DAN SMIT: I was suggesting that there was a research model which had revealed itself to be actually quite effective. And that was a model in which there was an institution which had a particular set of agendas. Those agendas were agendas related to a constituency on the one hand, and to a development objective on the other. And that when you have an institution which was geared in that kind of way, one often had very positive and productive research input. And that research input would very often be innovative in its own right. I would agree, however, that that's not sufficient, that basic research has to occur some place. That very often the innovation within institutions like the Urban Foundation or any development institution occurs within paradigms. What I would insist, though, is that the kind of model that the Urban Foundation could be, in a particular kind of research, its a model worth trying to repeat in the structures that we set up in a democratic South Africa. 51 TRANSFORMATION 18 (1992)