3.1. The players: an overview
3.2. Criteria for relevance of CG research
3.3. An analysis of the players
In this chapter, the diversity of players in the field of agricultural policy research is sketched in a global overview of policy research resources, and the small but important part played by the CG Centers in this is explored, particularly for where it has maximal value in terms of international public goods and for strategic linkage to other parts of the CGIAR portfolio. The slender and probably often diminishing capacity of national agricultural policy analysis units to deliver the needed research products is examined. In the somewhat different field of management research, the particular cases of ISNAR, IIMI, and CIFOR are also considered.
There are many actors in the world of policy research and quite a few of these are devoted primarily to addressing rural issues. It thus behooves anyone considering the role of the CGIAR Centers in this global system to explore first who the actors are and the various perceived advantages they enjoy and, in this way, focus on the special contribution that might or should be made by the Centers.
The research community involved with policy and management is a large one, whether one views it globally or in more confining terms, such as for MDCs only, and it is still a large one for LDCs or for the tropical world generally as subsets of primary interest to CGIAR discussions. To overview this community, a simple categorization of the major groups of players is undertaken in Table 3.1. The intention is to provide a scheme for seeing how they align and to set the scene for subsequent commentary.
Table 3.1. Categories of Major Players in Policy and Management Research
1.1 International1.1.1 GOs (Nonh)National Development Assistance Agencies, e.g., DANIDA, ODA.1.1.2 NGOs
NAROs and EPAs, e.g., USDA ERS, FAS
Universities, e.g., IDS, UNUResearch agencies, e.g., WRI1.1.3 Official Development Agencies
Action Agencies, e.g., Winrock
Firms, e.g., World Economic InstituteFAO, UNDP, UNEP, Regional Development Banks, World Bank1.1.4 IARCs1.2 National
1.2.1 NAPAs, e.g., BIDS, TDRI
1.2.2 NARSsNAROs, Universities1.2.3 NGOs (South)
The next step in proceeding to an overview is to classify the fields of research that are under discussion in this review. A scheme is offered in Table 3.2 that introduces several categories of policy research and parallel categories of management research. These are intended to be self-explanatory, but the emphasis on agricultural research per se is obligatory, given the nature of this review. It will also be the case that some work does not fit neatly into just one defined category.
Table 3.2. Simplified Categories of Rural Policy and Management Research
|
Category |
Policy Research |
Management Research |
|
1 |
Macroeconomic & Trade Policy |
Trade Regulation (and negotiation) |
|
2 |
Agricultural Policy (including domestic agricultural and marketing policy) |
Agricultural Agency Management (including marketing parastatals, extension and national policy analysis management) |
|
3 |
Research Policy (including AR priority setting and funding, and population, migration, gender,...) |
Agricultural Research Organization and Management (public and public-private sector interface) |
|
4a |
Poverty Policy (including food security and nutrition policy) |
Food Program Management (including relevant tax and welfare management) |
|
4b |
Health & Education Policy (relevant to the rural sector, including agriculture) |
Public Health System Management (relevant to rural sector, including regulation of toxic products) |
|
5 |
Environment Policy (relevant to the rural sector) |
Natural Resource Management (including land, soil, water, forest, fishery, colonization, biodiversity) by public, private, and collective organizations |
With this step taken, the next one is to present some crude and necessarily highly subjective estimates of the numbers of full-time equivalent research workers in these fields around the globe. A global perspective seems appropriate, given the potentially high level of transferability of relevant human skills. Such a tabulation is presented in Table 3.3.
Table 3.3 Indicative Rural-Oriented Policy and Management Research Scientists
(in full-time equivalent (a))
|
|
Policy |
Management | |
|
North |
|
| |
|
|
GOs |
1800(b) |
1700 |
|
|
NGOs |
500 |
300 |
|
|
Sub-total |
2300 |
2000 |
|
South |
|
| |
|
|
GOs |
1000 |
2000 |
|
|
NGOs |
200 |
200 |
|
|
Sub-total |
1200 |
2200 |
|
International & Regional |
|
| |
|
|
ODA agencies |
200 |
150 |
|
|
IARCs |
58 |
18 |
|
Grand total |
3758 |
4368 | |
|
IARCs as % of total |
1.6 |
0.4 | |
(a) Data are subjective estimates (by Jock Anderson) intended to illustrate orders of magnitude of the numbers of involved research personnel.(b) Includes, for instance, 425 workers in the USDA and State experiment station system of the USA in 1992-93 (Cooperative State Research Service 1994) and 790 workers in the North nations of the British Commonwealth in 1986(Vernon, 1989).
These indicative world head-counts of P&M research workers suggest that a majority (but probably not an overwhelming one) of these workers are in the North, although significant (but unknown) numbers of these concentrate their efforts on the South. The subjective standard errors for the management researcher numbers are probably greater than for policy researchers, but if these data are even close to correct it is surprising (at least to this Panel) that the numbers of research workers in the management fields under review here are of the same order of magnitude (and perhaps more numerous) as more mainstream policy work. The CGIAR is a relative latecomer to this field, and it is still a very modest player. Another way of expressing this is to observe that the ratio of client partners to CGIAR researchers in management work is surely large and possibly overwhelming.
The matter of the CGIAR System effort being small is clearly portrayed in Table 3.3. The indicated small proportions of research personnel in the System places a premium on arguments as to just what it is that such a small cadre of research workers, no matter how fine and committed, can really accomplish.
The CGIAR has, over recent years, had several occasions to ponder afresh the criteria for inclusion within the CG portfolio. First, not unreasonably, the activity must be research or at least research-related. Research is interpreted to be a systematic approach to discovering new knowledge and thus to build on the past. Research-related activities offer opportunity for more fringe activities, some of which are surely relevant in considering policy research, where the distinction between policy analysis based on existing information and policy research involving the distillation of new information may be either fine or blurred, but it is seldom likely to be crystal-clear. Of course, the key consideration in this criterion is the quality of the research - minimally, international in at least one sense of this qualifier as used in international journals, etc.
Second, the research activity must be international in character and must contribute to a priority program that is consistent with the goals of the CGIAR. The international dimension in this sense requires that more than one LDC be involved, and that there is some movement of information or more material aspects across boundaries. Another important aspect of the international character is the potential further transferability (spillover) of new information across national boundaries. It is natural to inquire as to just how broadly applicable policy research might be. The answer will depend on the nature of the policy topic under consideration. It is conceivable that some useful policy research may have very little transferability outside the particular national context in which it is conducted. This is likely to be a rather special case, however, and it is more likely that there will typically be a high degree of applicability and relevance of policy analysis across a range of analogous or broadly similar national circumstances. One contemporary example of this is analysis of reform policies that will have wide relevance in many transition economies.
A third and important criterion for inclusion in the CGIAR portfolio is that at least one CG entity must be judged to be the best qualified institution to undertake the designated work. This will usually be reflected in a low unit cost per significant international research result, with benefit accruing from the rapid international exchange of information that is derived, along with positive interrelationships with other activities in the Center conducting the research. Naturally, it is expected that the potential payoff should be high relative to cost.
Thus, the policy research agenda is wide, ranging over objectives concerned with productivity, equity, sustainability, and environmental issues, all within national borders and beyond, and there is an analogous and somewhat overlapping breadth to the kindred issues in governance and management of public systems. Therefore, in reviewing the roles of the many different actors, it should be necessary to consider the quality of the research, the internationality of the work, and the comparative advantage of CG providers versus other institutions for conducting such work. An informal attempt at such an assessment is made in the following section 3.3.
3.3.1. Policy research
3.3.2. Research on research policy
3.3.3. Management research
In principle, it would be possible to make an assessment of all potential suppliers listed in Table 3.1, disaggregated by detailed institutional characteristics, according to the criteria used to assess relevance for entry into the CGIAR portfolio. This task has been resisted, however, because of the intrinsic difficulty of measurement, as well as the proliferation of specific institutional groups that would have to be included if there was to be any pretense of comprehensiveness. The alternative is to take a more parsimonious and selective approach to discussing possible advantages of different categories of providers.
In general, it is accepted that the CGIAR is a small but solid actor in the categories of policy research listed as #2 (agricultural policy) and #3 (research policy) in Table 3.2. In category 1 (macroeconomic and trade policy), the CGIAR is a small player in a rather large field. In categories 4a, 4b, and 5, the poverty, health, and environment, respectively, the CGIAR is also small, but it has special comparative advantages. For example, the initiatives undertaken by IRRI, CIP, IITA, and CIAT on integrated past management are extremely valuable, but the scope and policy implications of this work relating to potential health hazards, particularly those associated with agricultural chemicals, is broad, and the initiatives have been modest indeed. Inevitably, this theme of research will have to be significantly increased in future decades, especially as the consequences of pesticide and other agricultural chemical mismanagement and inappropriate use become more manifest.
Research is one form of economic activity in which Weinberg's presumption, that the future will be like the past because in the past the future was like the past, seems generally applicable. In this regard, it is appropriate to reflect upon some of the success stories in policy research that would lead to a sense of optimism for tomorrow. Any such arbitrary list is bound to be unfair to those excluded and probably unjustifiable in terms of its implicit support for those connected with the included items. Nevertheless, to make this discussion more concrete, a few examples will be mentioned. The work by Krueger, Schiff, and Valdés, based at IFPRI and undertaken in conjunction with the World Bank, on the consequences of trade restrictiveness and the implicit taxation of agriculture in LDCs is notable. More micro-level implications of trade restrictions have been effectively pursued and analyzed by von Braun and others, also at IFPRI.
Given the concerns deriving from both poverty and environmental themes, the IRRI work on disadvantaged regions and the implications for IRRI's research program deserve particular commendation. On other themes, there is a proliferation of work that has a wide spectrum of disciplinary origins and analytical emphases. The ICRISAT VLS studies and the many subsequent detailed analyses that have been possible are a theme that has been mentioned elsewhere here. The CIAT work on cassava and its new role in market-led opportunities in Colombia and Northeast Brazil represent a rather different example. Recent CIP work on the successful spread of diverse potato varieties is also highly policy-sensitive and reveals successful exploitation of cogent socioeconomic research. Similarly, the CIMMYT and ICARDA analyses of research priorities for marginal areas have been a significant contribution to the refinement of research policy pertaining to this aspect of scarce resource allocation. On the livestock side, ILRI's work on dairy marketing in Africa has been demonstrably of high quality and applicability in contributing to the broader agricultural and food policy debate in Sub-Saharan Africa.
The examples could be replicated many times over, and only the combination of space shortage and ignorance of detail inhibits further documentation. The research reports of all the Centers need to be consulted for detailed documentation in this regard. Broad thematic areas are also worthy of mention, such as IFPRI's fledgling activities with its all-too-scarce partners in Sub-Saharan Africa and its endeavors to assess the wider technological and economic development implications of macroeconomic policy in Latin America (especially through the case studies in Argentina and Chile). As emphasized above, any such listing of illustrative success stories is bound to be inadequate, but it is allowed in the above to add to the concreteness of the discussion. The Panel's resources did not include adequate opportunity to document such past socioeconomic, policy, and management research endeavors in a manner that was in any way comprehensive or adequate.
Assessment of research advantage in any particular category of policy research is fraught with difficulties. As indicated implicitly in Table 3.3, the number of suppliers is large, even speaking in institutional terms and certainly in terms of total human resource capacity. As mentioned before, the result is that the overall CGIAR P&M research effort is relatively very small. When the CO System is selecting its priorities, it must do so with due regard to the advantages enjoyed by alternative suppliers, the adequacy of effort supplied by alternative suppliers, and its own niche opportunities to work in the area, presumably with the best suppliers, especially in the South but also in the North. Universities, for instance, have usually managed to carve out particular niches for some of their most specialized research and teaching enterprises. For example, Oxford University and Brown University have both developed considerable expertise in food-program management work even though there are many others similarly involved.
Thus, any CGIAR initiative in such an area must take appropriate account of existing capacity and orientation before it develops new research endeavors in such an area. And so it goes for almost every potential research policy theme. If the theme is the effectiveness of rural credit, for example, there are multiple potential suppliers, including such US-based universities as Ohio State. In the broad field of development studies, particularly those with a Sub-Saharan Africa orientation, again, there are many academic institutions in the North, both in Europe (e.g., Kiel, Wageningen) and in North America (e.g., Cornell, Michigan State, Stanford FRI) that have a significant commitment to such work, although it is often organized on a short-term project basis. It is no surprise that the CGIAR Centers are actively linked to a wide variety of such universities, both for post-graduate training, post-doctoral research opportunities, and peer interaction and collaboration on major research themes.
The happy conjunction of high potential and modest numbers of research workers is the essence of the case for continued investment in this type of work by the CGIAR. The CGIAR Centers will thus always have many potential partners, and they should actively seek these partnerships for maximum effectiveness. There could be natural collaborative research opportunities between, say, the World Bank, regional development banks, universities in both MDCs and LDCs, and national agricultural policy agencies. For example, the contemporary allocation of research resources within the World Bank surely reflects such a situation in collaborative work on measuring poverty.
The strong links between CGIAR Centers and the World Bank are also indicated in Table 3.2 research categories 2 and 3, as well as 5. There are many yet unexploited opportunities for involving LDC universities and NAROs in analogous collaborative research work that, depending on its nature, may also involve a three-cornered arrangement with an appropriate IARC. Indeed, quite a few such arrangements are already in place.
The point has been that CG policy research is a small but important element of the global policy research effort. Within the CG System, however, IFPRI stands as a relatively large and important player in this modest but influential work, and thus it deserves explicit attention. The niche claimed by IFPRI (at least according to its Medium-Term Plan) is defined by its perceived comparative advantage in conducting process-defined work according to the following aspects:
(a) It has a critical mass of policy analysts that can be organized flexibly according to the complementary efforts required for focused task forces.(b) There is a purposeful dedication to research without distraction by other activities such as teaching within university departments and project management within other public agencies such as the World Bank.
(c) Close links are maintained to technological generation units through cross-Center participation within the CG System.
(d) Traditionally disparate social and economic disciplines work together within the Institute and its assembled research teams.
(e) Finally, there is the record of past achievement and perceived impartiality in its policy research work.
Even a casual reading of IFPRI publications, supported by knowledge of its staff and institutional structures, strongly supports most of these claims. Some claims, however, could usefully be challenged along the following lines:
(a) Arguing that full-time dedication is a niche-defining attribute can be dangerous if indeed there are strong complementarities through engagement in other activities, such as teaching policy analysis or conducting policy dialogue with borrowing countries as in, say, development banks. Indeed, a case might be made that the most productive long term policy research environment would be within agencies that have a multiplicity of policy-related activities, for as long as research does not become an activity marginalized by other tasks.(b) The claimed advantages of strong links with technological development work, such as is accorded through CG membership, does not adequately recognize that many other potential players have similarly close working relationships with biophysical agricultural research systems, both within and without the CGIAR. Indeed, some of IFPRI's critics have charged that, in spite of its CG membership, at least until recently its links with other Centers have been fragmentary and insufficient. This is, however, a time of change and the current situation now looks quite positive in this regard.
(c) The claim of interdisciplinary scope in IFPRI's teams is a thin straw if it is contrasted with the way some policy-oriented work is undertaken at other CG centers where, by the nature of the mandate of the Center and its staffing, much more diverse teams can be and are put in place. The situation is now, however, changed (if not clouded) by the new-style inter-Center initiatives in which IFPRI has been such a busy and active partner, and thus its System-wide effective transdisciplinarity has been much extended.
(d) The Panel has not had an opportunity to assemble comparable data to assess the productivity of IFPRI relative to other institutes and agencies around the world, but it does not stand out as singularly impressive in terms of volume of published papers per full-time research fellow equivalent. It might be argued that the style of work at IFPRI, with its typically long in-field work in LDCs followed by collaborative interpretation of primary data into a policy analysis mode, means that what is produced is relatively thorough-going work and not a multitude of quick analyses, such as may be more the style of some other agencies. This claim then is worthy of further empirical testing with some attempt to correct for the different types of research products and their different inherent qualities and resource requirements. The claim of impartiality is probably quite reasonable, at least to the extent that partners in LDCs do strongly attest to this feature of IFPRI's work as being one of its most appreciated strengths, and certainly its working style has sought to emphasize this aspect of IFPRI's work. Documentation of this issue is available in various forms, including the CGIAR Impact Study of the mid-80s (Anderson, Herdt, and Scobie, 1988).
Research on research itself and related policy is something that has been done from time to time in various parts of the CGIAR System, and, of course, extensively outside the System. ISNAR was the lead Center for conducting research on research policy within the CGIAR, at least until recent times, although increasingly IFPRI's recent activities are complementing those of ISNAR. One basic requirement of any such analytical work is the availability of cogent data on national research systems, and this has been a commendable contribution of ISNAR in assembling such a database and implementing procedures for updating and maintaining a system for monitoring the state of NARSs. Much remains to be done in exploiting this database analytically to yield insights that may better inform decision makers concerned with investment in research systems, particularly public systems, but also in comprehending the nature of the broader research portfolio that involves the private sector as well, which is increasingly the case in many parts of the world.
The case for CGIAR investment in this work is strong, given its lack of popularity as a theme in the large policy world outside the CG system where little has been done, and much of what is available is for a very restricted set of OECD countries, such as USA, UK, Netherlands, and Australia. There are demonstrably pressing policy issues that must be addressed as the investment in national research systems in many countries is diminishing to crisis levels. These, for instance, stifle achievement in research systems where a large fraction of the scarce resources available goes to salaries only and then often disturbingly late. The productivity of such systems is unlikely to be sufficiently high to justify the existing public investment, and the crisis is bound to become worse before there is any significant improvement. Accordingly, policy analysis of options open to countries is of utmost importance and must be addressed by many agencies, including ISNAR and IFPRI, but also concerned donors and development agencies, such as the World Bank.
As is revealed in Table 3.3, the ability to conduct research in the field of management is rather restricted, given the few human resources devoted to such work. Since the CGIAR has focused on some key public management research issues, it is thus appropriate to address some of the special cases and their particular problems.
Management of public agricultural research agencies is clearly well within the mandate of ISNAR, although the focus to date has been on a relatively narrow concept of agricultural research, agricultural as opposed to fishery and forest, for example, and (for historical reasons) on primarily publicly managed national research organizations. The management problems of these agencies are considerable, in part because of a lack of clear guidance about the best organizational forms and managerial procedures. Since such agencies are largely within the public-service sector, they are usually compromised by the necessity to conform with public-service rules and procedures applicable within each country.
In an increasing number of cases, agencies have been recast as autonomous or at least semi-autonomous bodies that enable them to disconnect from some of the strictures of traditional public-service systems, such as promotion solely according to seniority, one illustrative example that is clearly counterproductive in a scientific research environment. Notwithstanding the many difficulties and problems that remain to be addressed, the resources thus far devoted to such work have been extremely small and the work has not yet been structured in a very systematic cross-country learning mode. Much effort has been devoted to policy dialogues among key decision makers and managers in national research systems, but this in itself does not lead to critical analysis of what works and what does not. Accordingly, much remains to be done in this field, and it should be accorded high priority in the public-management research agenda of the System. Without investigations that point the way more clearly, investment in this vital developmental field will be to little avail, and agriculture will thus not deliver its promise.
Other management themes within the system are diverse, but two are of particular significance, given the focus of the two institutions that recently joined the system, namely CIFOR and IIMI. Other centers are also engaged to some extent in issues related to public management, such as CIAT in Latin American land policy and ICRAF in the management of publicly, privately, and commonly owned agroforestry resources. In this brief summary and the longer discussion in section 7.7, however, the focus is on CIFOR, IIMI, and ISNAR.
The difficulties facing managers of forest resources in LDCs are manifold, and the social losses through inefficient management of public forest resources are probably huge, although not well documented. In part, the inadequacy of current information is related to the prevailing weaknesses in national forest research systems and the sometimes compromising links between forest departments and such research units, typically where the research unit itself is part of the department that is suffering the problem. Other contentious issues related to both biodiversity preservation and gainers and losers in the use of public forest resources have surely made even more difficult the progress of investigatory work on management.
The issues to be addressed are not primarily the traditional technical aspects of tree and forest management per se. Rather, the issues are the public management of the natural resources and the public control of the human resources involved. Hence, CIFOR's emphasis on devolution, on developing alternative institutional arrangements for the management of local/communal forests when public agencies no longer have the capacity, finances, and/or mandate for direct management of such resources. Governance, and the new roles in civil society, and management with NGOs and the private sector, all figure prominently in CIFOR's37 current strategic plan.
The CGIAR investment in this work is a modest beginning for tackling a large and significant global problem. The issues facing fishery resource management in coastal areas and inland water bodies are similar in many respects, and the System must look to the global experience (as embodied, for example, in FAO) and to ICLARM's fledgling efforts in this regard for guiding future investment in this important field of public responsibility.
With the growing shortage of water for all purposes in the world, especially for the most densely populated areas, the recent attention to improved water policies is clearly appropriate and significant, especially in regard to irrigation water and especially in Asia. IIMI has primarily addressed such polices. The inefficiencies of many irrigation management systems have been well documented in a wide range of studies in recent decades, and the establishment of IIMI to focus on this issue, working through collaborative mechanisms with national irrigation authorities, seems to have been timely and important. This is not the place to endeavor to assess just how successful this work has been, but from a strategic perspective, the work is surely vital and must be conducted with renewed vigor as the problems magnify. With limited space for adding new irrigation areas and the political difficulties over building new major reservoirs, the existing resources must be used more efficiently to economically underpin the demands that will be placed on these irrigation systems for their contribution to future food and fiber productivity. The lessons of improved management systems for water resources are many and are addressed at greater length in the companion strategic review of natural-resource management research in the CGIAR. In brief, effective involvement of water users and linkage between both upstream and downstream residents in river basins are key to improved management and to greater social benefits through the operation of such public and community systems.
