Cover Image
close this book Laying the foundation - The institution of knowledge in developing countries
close this folder Part II The institutionalization of science in Latin America
View the document Chapter 3 Institutional development of basic sciences
View the document Chapter 4 Institutional failure or success: variables affecting agricultural research
View the document Chapter 5 Institutional development of research in economics
View the document Chapter 6 Institutional development of educational research
View the document Chapter 7 Industrial research and development institutions in brazil

Chapter 4 Institutional failure or success: variables affecting agricultural research

Jorge Ardila

The need for agricultural technology usually originates in the intermediate stages of a country's development and normally results from a policy decision. Initially, the research infrastructure tends to be a virtual state monopoly. As countries progress, the private sector begins to generate important research that complements state-supported work. This research is usually done in anticipation of economic gain and tends to be concentrated in industries that produce raw materials such as hybrid seeds, fertilizers, pesticides, agricultural machinery, and processing technologies. In later stages of development, the strategic importance of technology to development starts to be recognized, and public policy becomes more coherent.

 

Agricultural Research in Latin America

The development of agricultural research in Latin America can be divided into three stages. In the first stage, from 1940 to 1949, agricultural research was organized as a public service. Motivated by agricultural developments in Europe, some Latin American governments (notably Colombia and Uruguay) made formal requests to Austria, Belgium, England, France, and Germany for assistance in laying the groundwork for agricultural research.

This first stage of cooperation featured attempts to import and adapt foreign species. Although some experimentation was done, especially with fertilizers and pesticides, most countries did not have the capacity to innovate. During this period, political decision-makers demonstrated great interest in research as a way to boost agricultural production, but private research remained minimal. At the international level, the organized research system we know today did not exist.

The second stage of development, from 1950 to 1970, was characterized by external support for state research at the national level. In June 1949, inspired by the success of the Marshall Plan for European reconstruction after World War 11, the United States established the Point IV program to transfer technology and expertise in an effort to alleviate the food production crisis in many developing countries (Carter and Harold 1985). American universities belonging to the land-grant system provided agricultural extension services and in-house training. During the 1960s, 3 026 Latin American professionals were granted doctorates in agrarian sciences from universities in the United States, enabling most Latin American countries to consolidate the state research structures they have today.

Table 1 shows the importance of this program of technical cooperation and graduate training for Argentina, Peru, and Colombia. A total of 498 researchers were trained at the master's and doctoral levels, at an approximate cost of 34 million United States dollars (USD).

Table 1. Number of graduate-level (MSc and PhD) researchers from 1960 to 1978 in the National Institute of Farming Technology (INTA), the Colombian Agricultural and Livestock Institute (ICA), and the Universidad Agraria La Molina (UAM).

Institution and country

Personnel trained

Cost (million USD)

 

MSc

PhD

Total

Country

USA

Total

INTA, Argentina

84

27

111

3.7

3.6

7.3

ICA, Colombia

175

93

268

11.6

7.1

18.7

UAM, Peru

89

30

119

-

-

-

Total

348

150

498

15.3

10.7

26.0

Note: If costs for Peru averaged the same as for Colombia, the total investment would have been 8.2 million United States dollars (USD), of which the Peruvian government contributed 6.3%. Note that contributions vary greatly from one country to another.

Source Ardila et al (1980).

During this golden age of the system, research made a major contribution to the region's economic and agricultural development. Research institutions flourished and enjoyed strong government support. Investment in research yielded positive returns, with high rates of social profitability, as measured by the economic surplus generated, and its distribution among customers and producers. Regional investment in agricultural research rose 9.5% in constant terms during the 1960s, compared with a world average of 9.1% (Boyce and Evenson 1975).

During this phase, the private sector made some interesting ventures into the field of research. The establishment of the Centre for Coffee Research (CENICAFE) by Colombia's coffee-growers in 1938 is one example (Samper Genecco 1992). However, large-scale efforts in the private sector were rare, probably because public-sector research programs satisfied the needs of the productive sector Also in many countries, the agricultural sector was simply not large enough to support long-term research.

The third stage, from 1970 to 1990, witnessed the decline of the state model and a crisis in agricultural research brought about by institutional, economic, social, and political changes, both domestically and internationally There was an urgent need to transform the research model to meet new demands, to face a new economic order, and to adjust to the reduced level of resources available.

A system of international research centres emerged during the late 1960s. It grew out of the experiences of England and France with tropical crops in their colonies as well as those of the Rockefeller Foundation in Mexico in the 1950s and 1960s. The system was composed of bilateral and multilateral assistance agencies and private foundations in developed countries, under the umbrella of the Consultative Group for International Agricultural Research (CGIAR) and its Technical Assistance Committee (Samper Genecco 1992).

This new research system was part of a strategy to channel international cooperation more efficiently. Efforts at institution-building had failed in many countries, and early successes were losing their effectiveness. Although the international centres were intended to complement the national systems, they were initially seen as competitors or replace meets. This led to a reduction in resources for important national research programs.

At the same time, the private sector, represented by interest groups, associations, and federations of producers, became more actively engaged in research. It did so partly to compensate for the weakened state of public institutions and partly to meet the demand for new technologies in which the public sector had no direct interest because they were felt to be designed primarily for private benefit.

These two main groupings - the international centres and the private sector - although contributing to the enrichment of research, also served to lessen and even displace state involvement in the production of technology. This effectively reduced the state's potential to socialize the benefits of technological change in the interests of greater equity.

Furthermore, the international agencies and the private sector offered more professional opportunities for researchers trained by the public sector, normally at much higher salaries than those paid by the state. Consequently, researchers left the public sector in droves. In addition, much of the international technical cooperation of the 1960s relied on resident American researchers. The end of this practice further complicated the staffing of state research institutes and threatened a specialist shortage.

The threat was temporarily averted by the establishment of graduate programs in agriculture in countries such as Argentina, Brazil, Chile, Colombia, and Mexico. When these programs were subsequently dropped, specialists were sent abroad to be trained once again. This time fewer funds were available, and most of the training opportunities under previous technical cooperation agreements had disappeared. The new programs arranged with foreign governments or with foreign loans never matched the former level of training activity. In some countries, the new graduate programs in public research institutes barely managed to replace the specialists who had left for other organizations.

This led to a drastic reduction in graduate specialists working in most public institutions. Brazil was the exception because it had never stopped training specialists through the Brazilian Agricultural Research Organization (EMBRAPA). Other consequences included a slowdown in the production of innovations and increased training costs, which exerted even greater pressure on already tight budgets. With the lowering of public-sector salaries, research institutes were forced to replace outgoing researchers with nonspecialist staff, which reduced their chances of producing high-quality innovations and further contributed to their loss of social recognition.

In addition to this human resource crisis in the public sector, budgets were drastically cut by government in an effort to reduce fiscal deficits. For example, in Colombia in 1990, research staff of the Colombian Agricultural and Livestock Institute (ICA) received only half the amount for equipment and materials that they had received in 1970. If, in 1970, operating costs comprised 30% of expenses, in real terms, then a 15% budget reduction amounted to a 50% reduction in operating capacity (ICA 1990).

The state research infrastructure is suffering seriously from underinvestment. As a result, the national expenditure needed to restore the system to its former level of functioning is likely to be substantially less than the economic benefit lost because of the lack of technological change in agriculture.

In addition to these institutional factors, several economic considerations deserve mention. The energy crisis of the 1970s raised the cost of producing goods that required imported inputs derived from oil and rendered much technology obsolete. The national research institutes (NRls) in Latin America continued to develop technological systems based on modern inputs and to produce costly capital-intensive technology to improve physical yields.

These economic variables accompanied major changes in worldwide patterns of consumption. Increased urbanization and personal incomes generated a demand for new and improved products and for research to develop them. The demand for research was further intensified by the liberalization of Latin American economies and the need to boost the international competitiveness of local production. Economic liberalization put pressure on the NRls to expand their range of research subjects and regions; it also focused research on raising the export-earning potential of products. This expansion has further reduced the operating capacity of the national institutes.

Fundamental quantitative and qualitative changes are currently taking place in the demand for agricultural technology. The new trend has serious implications for the entire research system, affecting the topics selected for study, as well as the equipment and specialists required. The research system faces the new demand that agricultural development be sustainable over the long term. Bottlenecks created by earlier research and the occupation of new land must be removed. Problems such as soil erosion, compaction, and salinity as well as the pressing need to replace pesticides rendered ineffective by resistance must be solved. Because the private sector has no direct interest in financing this basic research, it must be supported by weakened public institutions.

Latin America is currently engaged in the political process of decentralization and democratization designed to achieve greater efficiency in the use of public resources and a more equitable distribution of services. This will require moving resources from research to support technical assistance and technology transfer at the farm level.

Although renewed private-sector interest has resulted in some notable successes, it is still the public-sector NRIs that have made the most significant contribution to agricultural research. Their deterioration may result in slower economic growth and a restricted contribution to increased agricultural productivity. The scarcity of research resources, the decline in quality, and the new demand for technology have combined to reduce technology's contribution to regional growth in the last 10 years, and threaten to lead to technological stagnation.

These factors have had a negative impact on the region's capacity to produce needed technological change in the field of agriculture. With the institutional model in decline, Latin America has an additional disadvantage in that recent progress in biotechnology and genetic engineering may be widening the technology gap between the region and the industrialized countries.

 

Characteristics of Agricultural Research Institutes in Latin America

This section offers a preliminary review of the diversity of institutions in agricultural research in Latin America, with some comments on their interaction from three points of view: the economic nature of technology; the level of participation of technology in stages and processes of agricultural research; and the level of coverage of the organization's objectives and user satisfaction.

Institutions have many kinds of relationships with each other. Sometimes the decline of one institution is accompanied by the rise of another (substitution), and sometimes the success of one depends on the success of the other (complementarily). Institutions may also compete with each other for resources, or over research products.

Institutional transformation in agricultural research may be attributed to such factors as the benefits to users; political changes, normally associated with an attempt to improve bureaucratic efficiency; and major changes in the availability and price of resources required for production (Ruttan 1982). Several economic and political reasons explain the creation, development, and transformation of research institutions, always making the distinction between institutional variance (reasons for creation, transformation, or disappearance) and institutional failure or success.

Economic nature of technology

Both the state and the private sector have an interest in obtaining part of the economic surplus generated by technology. The goal of the state is to make technology more widely available to its citizens, and the goal of the private sector is to use technology for profit-making production.

Technology may be in the public domain or it may be associated with the appropriation of economic benefit. Examples of the former include varieties of plants developed to ensure successive identical generations, usually by propagation, as in potatoes and sugarcane. Examples of the latter include hybrid seeds that cannot be used to grow successive generations because of falling yields, forcing the grower to buy from the hybrid breeder who retains the genes for crossing. Here, it is obviously in the private sector's interest to promote hybrids and in the state's interest to promote varieties. Some biological techniques, such as improved varieties resistant to pests or diseases, substitute for the use of chemical inputs. Here, the state has a central interest in research; private industry, in production.

The introduction of biotechnology and genetic engineering has enabled major chemical manufacturers to control the producers of improved varieties and hybrids and to create seeds that respond only to certain inputs, which they produce. This forces the buyers of hybrids to also buy the inputs. As a result, the seed producers of Latin America are in danger of losing their market.

Technology is also built into equipment, processes, and raw materials that are sold at prices set to recoup research costs and generate profits. Private enterprise is particularly interested in this type of technology, and it is often the only sector capable of making the large financial investment required. The four main models of research institutions are described here.

Institutions that liberate technology - Public-sector agricultural research institutes, regional research centres, and international institutes are typical of this organizational model in Latin America (Tables 2, 3, and 4). These institutions are complementary, although most NRIs do not possess the infrastructure needed to reproduce the products of the international research system.

Table 2 National agricultural research institutes in Latin America.

Country

Institute

Year created

Argentina

National Institute of Farming Technology (INTA)

1957

Ecuador

National Institute of Agricultural Research (INIAP)

1959

Mexico

National Institute of Agricultural Research (INIA)

1960

Venezuela

Fondo Nacional de Investigaciones Agropecuarias (FONAlAP)

1961

Colombia

Colombian Agricultural and Livestock Institute (ICA)

1963

Peru

Servicio de Extensión y Promocion Agraria (SlPA)

1963

Chile

National Institute of Agricultural Research (INIA)

1964

Brazil

Brazilian Agricultural Research Organization (EMBRAPA)

1973

Source Pineiro and Trigo (1983).


Table 3. The international agricultural research centers.


Table 3 concluded


Table 4. Regional research centres with links in Latin America.

In terms of investment, it is estimated that the system of international centres spends approximately two and a half times more than national institutes. The investment of the CGIAR-affiliated centres is estimated to be about 200 million USD plus a similar amount for the other institutions listed in Table 3. Latin American NRls probably spend another 200 million USD between them. The economic surplus generated for the region runs into billions, which explains the interest in agricultural research.

International centres help countries keep up with developments in basic food production, a task that is especially critical in light of the current technological revolution in biotechnology and genetic engineering.

Institutions that appropriate technology - These models are generated by local or international private industry or agroindustry and are frequently linked to the research and marketing efforts of large transnational corporations. These companies are likely to be involved in breeding improved seeds and manufacturing fungicides, pesticides, fertilizers, other forms of soil treatment, or animal feed and concentrates (Pray and Echeverria 1989).

Included here are firms that manufacture and distribute machinery and implements, companies that produce the active ingredients and raw materials later assembled in satellite countries, as well as firms that provide support services (such as software for agricultural applications) and technical assistance for the pseudotransfer of process technology. (Pseudotransfer of technology occurs when research is conducted abroad, but development and promotion are carried out at home. This happens when inputs and machinery are imported and assembled with their technology built in and then sold locally. In vitro growing techniques for certain species are an example.)

Some NRIs work on technology that can be appropriated by the private sector. This includes farm machinery and irrigation systems, plants for research, and inputs such as nitrogen-fixing bacteria. Recently, NRIs and agroindustrial interests in Latin America have undertaken joint ventures to produce and market certain types of improved genetic material, machinery, and implements. The government contributes to the development of the technology and shares the profits with the private-sector partner.

Institutional models of restricted release - Here, technology is either copied, created, or adapted and then used by the group funding the research. Private institutes dedicated to applied research financed by producer associations are typical of this model. Examples include the Executive Commission of the Plan for the Economic Recuperation of Cacao (CEPLAC) in Brazil and the Centre for Coffee Research (CENICAFE) in Colombia.

In another version of this model, firms engaged in the processing of farm products support research to improve their raw materials. In Colombia, CENICANA, the Research Centre for Sugarcane, is an institute for research on sugarcane, and other countries have similar institutes funded by the manufacturers of cigarettes, chocolates, beer, and wine (Pray and Echeverria 1989). National and international consultants such as Chemonics and Winrock International, which release their knowledge to a restricted public and make cross-border transfers of technology, also fit this model.

Institutions liberating technologies between countries - These integrative institutions promote the dissemination of knowledge through the exchange of research as well as germplasm and other materials. Work often goes on concurrently and in a coordinated fashion in a number of countries (for more information, see Trigo 1988; IICA 1990; Plucknett et al. 1990). This category includes the following research networks (Ardila 1990):

International system networks, performing work that originates in the international research centres and is coordinated at the national level by an NRI;

· Cooperation networks among countries, such as the Central American Cooperation Program for Crop Improvement;

·Networks related with specific work under development by regional centres such as the Tropical Agricultural Research and Training Centre (CATIE) in Turrialba, Costa Rica, the Caribbean Agricultural Research and Development Institute (CARDI), in St Augustine, Trinidad and Tobago, or the Agricultural Research Foundation of Honduras (FHIA);

·Cooperation programs involving two or more countries, such as the Inter-American Institute for Cooperation on Agriculture (IICA), also linked to the work of international centres. Examples are the programs for the Southern Cone (PROCISUR) and for the Andean Region (PROCIANDINO).

The development and transfer of technology

This section describes institutions composed of small groups working separately to contribute to the development and transfer of technology. The process includes the following stages (Ardila 1987, 1989):

· Acquiring and copying technology that requires no adaptation;

· Developing basic research to expand knowledge and production opportunities;

· Adapting, validating, and adjusting technology for specific applications;

· Developing technology to make it commercially available to a large market;

· Promoting technology; and

· Maintaining technology already on the market.

Argentina, Brazil, and Mexico are the only Latin American countries to participate in all these stages. They are the only ones to distinguish between different research functions and to set up specialized scientific units with substantial budgets and first-rate staff and equipment. In contrast, the international centres focus almost entirely on basic research in defined areas. They do little applied work and international promotion, and their work requires completion by the countries to which it is delivered.

Most research institutions in Latin America are not actively involved in the first two stages (copying and developing technology). They tend to be engaged more in adapting, validating, and promoting technology as well as in protecting national inventions by restricting the import of new technologies. They may also take on ambitious projects to extend, transfer, and promote technology.

There is a serious lack of coordination in the region between research (adaptation, validation, and applied research) and efforts connected with the extension, promotion, and massive development of technology. Frequently, innovations go no farther than interesting research results. Although both research and development functions can be carried out successfully by one organization, they can also be divided between different bodies (as with EMBRAPA and EMBRATER, the Brazilian Technical Assistance and Rural Extension Corporation). These functions can take place under one roof and one management authority (as with the National Institute of Farming Technology (INTA) in Argentina) or under one roof and two management authorities (as with ICA in Colombia). In general, the more the functions are separated, the more difficult it is to make a positive impact on development because of the additional coordination required.

Institutional objectives and users

An institution is characterized by clearly defined objectives that can be associated with identifiable groups of users. In agricultural research, the users are farmers and the objective is technological development. Some institutions, however, have more than one objective and more than one kind of user. This multiplicity can lead to an overload of functions and complex operational structures that reduce effectiveness. The NRIs in Colombia, Mexico, and Peru are examples of this; in addition to research, they perform control and supervision functions, provide technical assistance, and run sanitation campaigns.

Institutional users can be broken down by region, product group, type of research problem, and type of user. Some institutions are designed to serve the research needs of a specific region. For example, IAP (the Research Institute on Peruvian Amazon) serves the Peruvian Amazon and COA (Araracuara Corporation) covers the Colombian Amazon. Institutions also differ widely with respect to the type of products and problems they deal with. The NRIs tend to work on a wide variety of products, regions, and problems; other institutions restrict themselves to a single product and a few research problems. The work of the International Laboratory for Research on Animal Diseases (ILRAD) in Kenya on trypanosomiasis and theilerosis in animals is an example of the latter. In general, an institution's chance of success is increased when its objectives and coverage are well defined and somewhat limited.

In addition to those mentioned so far, other organizational models also support the system's viability. The most important is the foundation model promoted by USAID (Sarles 1988). Foundations allow producers to exercise real influence in directing resources and activities to meet the sector's needs. This model ensures the creation of bonds and articulations in research because boards of directors include representatives of different research institutions. This promotes collaboration among them and strengthens their research, promotional, and educational activities.

Foundations have a more flexible and less bureaucratic management structure, and they are not as influenced by politics. Their role is not to undertake research, but to support institutions that do. In recent years, USAID has supported foundations in the Dominican Republic, Ecuador, El Salvador, Guatemala, Honduras,Jamaica, and Peru. Initial evaluations of the foundation model point to difficulties in ensuring stable, long-term domestic funding, a factor that may jeopardize its future.

 

An Agenda for the Future

We now have improved methodologies for evaluating the contribution of research results to economic growth. They have been refined to take equity into account - the distribution of benefits to different social sectors. Methodologies have been developed to evaluate impact at the product and regional level, and the work of Robert M. Solow has provided methodologies for evaluating research from the aggregate economic point of view (Romano 1987). Less work has been done on methodologies to evaluate equity, although the subject is now receiving closer attention. These three levels of methodological development are illustrated in the work of Solow (1957), Griliches (1958), and Schmitz and Seckler (1970).

The decline of the institutional research model has been advanced as a key reason for the failure of technological management to contribute to economic growth. Therefore, it is useful to identify the variables linked to institutional success. Institutional success occurs within a framework of efficiency - technologies that help to attain global objectives - and effectiveness - the achievement of results through good management of the organization and its resources.

This section outlines factors that may help to explain the success of agricultural research institutions. The variables thought to influence success are: the way in which the institution complements other institutions, the type of technology it produces, and its internal organization (Table 5). Each is analyzed in detail.


Table 5. Variables hypothetically associated with institutional success and failure.

Although the international centres have focused on basic technology, the NRIs have concentrated on applied technologies for public use, and the private sector has concerned itself with profit-making technology, there is a complementary relationship between all three. This affects the success of the system as a whole and each institution separately.

 

The system of international centres is constantly expanding the frontiers of knowledge through basic research and connections with specialist centres in developed countries Because the NRIs are unable to perform this function in Latin America, their work is assisted in great measure by the international system. If that interaction ceases, the efficiency of the whole system will suffer and so will food production.

Improved biological material usually comes out of the international system. Very few countries in the region have the capacity to improve their gene banks. Private enterprise has worked with applied technology, but only after obtaining improved germplasm from the international system and the NRIs. Therefore, private interests rely on the international system to provide improved material so that the yields of key crops such as rice, maize, wheat, potatoes, soybean, sorghum, and grasses can be maintained or increased.

Private enterprise in developed countries is different because it has the capacity to develop technology very rapidly for both machinery and inputs as well as genetic material. In this case, the NRIs import and adapt technology that perpetuates their dependency Because the NRIs are mainly occupied with improving plants, this complementary relationship has a decisive influence on success. The same is true of private enterprises associated with producer groups, although they are more interested in direct participation in research that goes further than producing improved varieties, responds to real needs and interests, and matures in the short term. Research institutes working on machinery and inputs face a different situation. But they barely exist in Latin America, except for some that are satellites of companies in industrialized countries.

Besides the complementarily between countries and the international system, countries in the region also complement each other. Important efforts are beginning in this respect, to some extent compensating for some of the defects and omissions of the international system, especially in fields different from basic foods (such as in the development of tropical products for export). It is clear that there is no such thing as technological independence and that interaction between the system's components is essential to institutional success.

Institutional success is closely linked to the practicality of a technology's new application. Technological change that does not meet real needs can result in institutional failure. A technology can, therefore, be measured by its contribution to economic growth, equity, and the conservation of natural resources. Contribution to economic growth refers to the overall aggregate impact of technological change on the allocation and use of a nation's resources. If the change does not produce a substantial surplus for the country and uses more of its scarce and more expensive resources, it represents a failure.

We may see a continuation of the institutional failures of the 1980s and 1990s if the NRIs cannot match the level of institutional activity of international centres and private enterprise in Latin America; that is, if they can no longer complement the other two elements in the system.

It is generally agreed that technology produces favourable results if it helps to improve the situation of a country's poor. Equity means that the benefits of technological change should be distributed fairly throughout society. Thus, the measure of institutional success or failure at this level will be largely determined by how equity is defined. Technology is considered a failure if it adversely affects sustainability and conservation of natural resources. Evidence seems to suggest that failure is more common than success.

Sixteen variables can be linked to an institution's success. The first seven are related to the context of the institution and the last nine to its internal management, organization, and policy.

1. Policy - Government policy, although not expressly formulated for research institutions, does in fact affect them, particularly with respect to funding for research and technology transfer and to organizational characteristics, functions, and objectives. For example, evidence shows that policies promoting an export-oriented economy also tend to support research and resource allocation for exportable goods.

2. Funding - Regardless of the legal nature of the institution, if the source of funds is limited, so is the institution's research capacity. Recently, the debt crisis and economic adjustment policies in Latin America have exacerbated the financial difficulties of the publicly funded NRIs.

3. Communications - All technologies require some degree of massive replication to reach producers. Latin American seed producers accomplish this task very efficiently. For other kinds of technology, particularly those not in the public domain, research institutions must develop communication links to ensure that technology reaches the producer.

4. Strategic alliances - In avoiding initiatives designed to broaden their funding base, publicly funded research institutions jeopardize their capacity to survive government cutbacks. Institutions need partners to perform complementary functions. Joint ventures always contribute to success.

When funds are scarce, partnerships within and between countries can be valuable. If operating costs comprise 30% of research costs, an institution can look to partners to cover the smaller budget items, or contract out its research capacity to other institutions. Depending on the type of institution receiving the contract, this arrangement can greatly expand an institution's capacity to achieve results. International cooperation on research is an example of using scarce resources wisely and simultaneously promoting equity among nations.

5. User participation - Institutions structured to allow the direct participation of users in research efforts tend to be more successful because they are more responsive to user needs. Several research institutions have created producer councils with the power to approve research plans and allocate funds. Governing councils can also be made up of government representatives rather than producers and serve as bureaucratic coordinating mechanisms.

6. External affiliation - Research institutions connected with larger institutional complexes have better chances for success. In Colombia, for example, coffee and sugarcane research institutes are part of a more complex system of production and distribution.

7. Country size - The larger a country and its economy, the better its chance of institutional success in research. Studies have shown that even the smallest crop research team in Central America costs far more that the crop's financing capacity, assuming that producers contribute 1% of the harvest toward research. This is one of the reasons that foundations and associations of regional centres are so vital to research.

8. Staff turnover - High staff turnover tends to delay research and raise its costs. An institution's ability to achieve significant results and use its resources efficiently depends on its capacity to retain qualified staff.

9. Staff quality - There is some evidence to suggest that high-quality technology requires highly qualified staff but this relationship must be investigated further.

10. Adaptability - An institution that can adapt easily to changes in its environment is more likely to be successful. Changes in research needs do not happen abruptly; they tend to evolve over time. Also, quick reactions to policy changes may lessen an institution's efficiency.

11. Capacity to change - Sometimes an institution's internal mechanisms act as obstacles to desired change and reduce its capacity to adapt to the environment. Changes in direction that are not backed by coherent internal decisions rarely produce good results.

12. Complexity and coverage - Institutions with a wide variety of objectives and clients may be more inclined to experience operational difficulties, internal competition for resources, and a greater likelihood of conflict, all of which lessen their chances of success.

13. Growth limits - Institutions that place clear limits on their growth are more successful than those that constantly expand their programs and coverage. The NRIs are a case in point. The fact that their programs and geographical coverage have grown far more quickly than their funding has created bottlenecks and hurt their performance.

14. Balance - A fair division of institutional resources between the development and transfer of research normally brings good results. In many cases, however, the balance of the two functions is not clearly defined, and resources for research exceed those devoted to transferring technological innovation.

15. Critical mass - Institutions must develop a critical mass of resources to achieve reasonable levels of success. If resources fall below a certain critical mass, an institution may be advised to reallocate funds and eliminate some of its activities.

16. Stability - Because research is a long-term activity, it needs stability to achieve its objectives. Frequent changes in objectives, organization, management, and staff can lead to institutional failure. Stability is therefore a prerequisite for success.