Consultative Group on International Agricultural Research CGR-8 @ CGIAR Study Paper Number 8 Bangladesh and the CGLAR Centers A Study of Their Collaboration in Agricultural Research Carl E. Pray Jock R. Anderson M~~~~~~~~~~~~~~~~~~~~~ - L -7 H~~~~~ )t Aio.7.Z..s.1. {Ji9- g Consultative Group on International Agricultural Research CGIAR Study Paper Number 8 Bangladesh and the CGIAR Centers A Study of Their Collaboration in Agricultural Research Carl E. Pray Jock R. Anderson The World Bank Washington, D.C. Copyright ©) 1985 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. First printing October 1985 All rights reserved Manufactured in the United States of America At its annual meeting in November 1983 the Consultative Group on International Agricultural Research (CGIAR) comrnissioned a wide-ranging impact study of the results of the activities of the international agricultural research oganizations under its sponsorship. An Advisory Committee was appointed to oversee the study and to present the principal findings at the annual meeetings of the CGIAR in October 1985. The impact study director was given responsibility for preparing the main report and commissioning a series of papers on particular research issues and on the work of the centers in selected countries. This paper is one of that series. The judgments expressed herein are those of the author(s). They do not necessarily reflect the views of the World Bank, of affiliated organizations, including the CGIAR Secretariat, of the international agricultural research centers supported by the CGIAR, of the donors to the CGIAR, or of any individual acting on their behalf. Staff of many national and international organizations provided valued information, but neither they nor their institutions are responsible for the views expressed in this paper. Neither are the views necessarily consistent with those expressed in the main and summary reports, and they should not be attributed to the Advisory Committee or the study director. This paper has been prepared and published informally in order to share the information with the least possible delay. Carl E. Pray is a research associate on the Asia Agricultural Research Project, Department of Agricultural and Applied Economics, University of Minnesota. Prior to 1981 he was an Agricultural Development Council associate in Dhaka, Bangladesh. Jock R. Anderson, director of the CGIAR impact study, is a professor of agricultural economics at the University of New England, Armidale, New South Wales, Australia. Library of Congress Cataloging-in-Publication Data Pray, Carl. Bangladesh and the CGIAR centers. (CGIAR study paper ; no. 8) Bibliography: p. 1. Agriculture--Research--Bangladesh. 2. Consultative Group on International Agricultural Research. 3. Agriculture--Research--International cooperation. I. Anderson, Jock R., 1941- . II. Consultative Group on International Agricultural Research. III. Title. IV. Series. S542.B3P73 1985 630'.72'05492 85-22749 ISBN 0-8213-0642-1 iii SUMMARY The collaboration between the international agricultural research centers and the national agricultural research system of Bangladesh has been very important to the growth of agricultural research and agricultural productivity in Bangladesh, The collaboration started in 1966 with the beginning of IRRI's relationship with the Agricultural Research Institute. The number of collaboratively researched crops has expanded gradually to include wheat, maize, potatoes, some pulses and cassava. The most important types of contact included posting center researchers in Bangladesh to work with local researchers on local problems, regular visits by center staff, training programs at the centers or run by the centers, joint programs of germplasm testing, and publications. These activities have influenced the size and direction of Bangladesh research. The well publicized success of IRRI rice and CIMMYT wheat elsewhere in Asia made it easier for the national program to acquire budget allocations and resources from the East Pakistan and Bangladesh governments and from the donors. The increases in funding have been most rapid in foodgrains, potatoes, pulses and oilseeds, while the research on export crops has grown more slowly. The proportion spent on export crops has thus declined. Within the commodity programs, joint planning of research in on-farm research programs has influenced priorities toward farmers' needs. The institutional structure of research has been influenced by the centers' emphasis on multidisciplinary teams organized around a single commodity. The general efficiency of the research programs that are in toucb with the centers seems to have improved - training at the centers has encouraged researchers to be more practical in outlook and work style. There is more incentive to perform vigorously with the prospect of wider recognition through participation in international yield trials or in international meetings. Center outposted (special project) staff have resources that must be helpful in overcoming crucial research bottlenecks. The outcome of this collaboration has been more food production and probably improved income distribution. Modern varieties of rice and wheat iv have increased the demand for labor, which benefits the landless laborers who are the poorest group in Bangladesh. Increased food production has reduced the real price of foodgrains, which primarily benefits poorer consumers both in the cities and the countryside. Many aspects of institutional arrangements for agriculture research in Bangladesh reflect the influences of several international centers, ranging from the structure of the rice research institute to aspects of the ongoing debates on input and food pricing policies in the country. The centers are widely appreciated for their contributions. ACKNOWLEDGMENT This report reflects the ideas and concerns of many friends in Bangladesh although they may not all agree with our synthesis of the achievements of the centers. We share their general enthusiasm for the past and future contributions of the centers' partnership arrangements to the future welfare of Bangladesh. Many people helped us but we must particularly acknowledge the strong support and active personal assistance of Dr. Kazi M. Badruddoza, and that of Zafar Ahmed on an earlier version of this paper. We are also grateful to Dorothy Marschak for assistance in reaching this stage of publication. WEIGHTS AND MEASURES 1 acre (ac) - 0.405 hectares (ha) 1 maund (md) - 36.3261 kilograms (kg) 1 seer (sr) - 0.933 kg 1 (long) ton - 27.22 maunds SPECIAL TERMS USED Aman - rice planted before or during the monsoon (which begins in June) and harvested in November-January: B. aman is broadcast aman, T. aman is tranplanted aman. Aus - rice planted during March-April and harvested during July-August Boro - rice planted in winter and harvested druing April-June Thana - administrative unit comprising between 6 and 15 villages FISCAL YEAR The Bangladesh Fiscal Year (FY) runs from July 1 to June 30. v CONTENTS I Overview of Bangladesh Agriculture and Growth of the Research System 1 1.1 Importance of agriculture 1 1.2 Agricultural research systems in East Pakistan 1 1.3 Agricultural research in Bangladesh 7 2 History of IARC Activities 11 2.1 IRRI 11 2.2 CIMMYT 15 2.3 CIP 16 2.4 Other CG Centers 17 3 The Role of the CG Centers in the Evolution of the National Agricultural Research System 18 3.1 Size of research budgets 20 3.2 Commodity allocation of resources 20 3.3 Institutional structure 24 3.4 Goals and priorities within the commodity program 24 3.5 New research methods 26 3.6 Efficiency of the research system 27 4 Local Perceptions of the Centers 29 4.1 Nature of collaborative research in Bangladesh 29 4.2 Concerns 32 5 Impact on Other Institutions 35 5.1 Extension 35 5.2 Policy 36 6 Impact of Collaborative Research on Farmers and Consumers 38 6.1 Productivity growth 38 6.2 Income distribution effects of HYV technology 46 6.2.1 Long-run trends in rural income and asset distribution 46 6.2.2 Large vs. small farmers vs. tenants 50 6.2.3 Landless laborers 53 6.2.4 Consumers vs. producers 55 6.2.5 Women 57 7 Future Impact of Collaborative Research 59 8 Conclusion 61 References 63 vi ACRONYMS USED ADAB Agricultural Development Agencies in Bangladesh ADC Agricultural Development Corporation AVRDC Asian Vegetable Research and Development Center BADC Bangladesh Agricultural Development Corporation BARC Bangladesh Agricultural Research Concept BARI Bangladesh Agricultural Research Institute BBS Bangladesh Bureau of Statistics BIDS Bangladesh Institute of Development Studies BRRI Bangladesh Rice Research Institute BTC Bangladesh Tobacco Company CARE Cooperative for American Relief to Everywhere CGIAR Consultative Group on International Agricultural Research CIDA Canadian International Development Agency CIMMYT Centro Internacional de Mejoramiento de Maiz y Trigo CIP Centro Internacional de la Papa FAO Food and Agriculture Organization FSR Farming systems research IADS International Agricultural Development Service IARC International Agricultural Research Center IBRD International Bank for Reconstruction and Development IFAD International Fund for Agricultural Development IFDC International Fertilizer Development Center IFPRI International Food and Policy Research Institute JRI Jute Research Institute NGO Non-Government Organization RDRS Rangpur Dinajpur Rehabilitative Service SRI Sugar Research Institute UPLB Univesity of the Philippines at Los Banos USAID United States Agency for International Development WAPDA Water and Power Development Authority 1 BANGLADESH 1 Overview of Bangladesh Agriculture and Growth of the Research System 1.1 Importance of agriculture The agricultural sector dominates the economy of Bangladesh. It accounts for about half of the GNP, 79 percent of employment and 80 percent of exports. Since separation from Pakistan agricultural output has grown about 3 percent annually, which was just slightly more than population growth. Table 1.1 shows the area, output and value of the major crops of Bangladesh. The crop that dominates the agriculture of Bangladesh is rice. It is planted in three major seasons - aus, aman, and boro. The planting and harvest dates of rice and the other crops are shown in Figure 1.1. The major cropping patterns are listed at the bottom of Figure 1.1. The trends in area, production and yield of the major field crops are shown in Table 1.2. Wheat is the only crop in which there is evidence of rapid productivity growth. Yield increased 12 percent annually. Yields of the two major rice crops grew about 2 percent annually. 1.2 Agricultural research systems in East Pakistan The national research system (NARS) that East Pakistan inherited from the British had been disrupted by World War II and the Bengal famine. It was further weakened by disruptions occasioned by the Independence movement. The main Bengali experiment stations were in East Pakistan, but the research institutes were completely shattered during partition. Almost all of the jute research workers went to India (Chaudhuri 1949) and other sections were affected nearly as much. The long-serving Director of BRRI describes the period from 1947 to 1956 as "the Dormant Period." He has observed that due to lack of funds and manpower, the research was curtailed to the maintenance of 2 Table 1.1 Area, Output and Value of Major Crops Area Production Value Value Added (000 acres) (1000 tons) (million takas) (million takas) Rice (All) 25,106 18588 51246 48350 Aus 7,505 4162 8589 7594 Aman 14,762 10228 32468 30793 Boro 2,839 3598 10194 9963 Wheat 1,071 787 2105 2928 Maize 5 1 4 4 Tea 107 40 645 220 Tobacco 111 37 433 426 Jute 1,874 1049 2729 2566 Oilseeds 768 235 1186 1124 Pulses 816 211 479 711 Potato 238 902 1259 1142 Source: BBS (1982). 3 Figure .1 CROP CALENDAR OF BANGLADESH Existing Cropping Patterns -e : VEGETABLES P'ADDY tm3r cw ______ _ _ __1 ____ *--lEflfl** = -White Gourd PADOY BOR Gourd .-..3... --- -- Pin ~~~~~~~~~~~ - in a Gram 17 (fl i .addiSh Wh . . . = . .n . khOIWN 1 ,-- fl**~~fl*fl~~~ Turnip z2o 1kho I Mtor Soyabean tmt Pus tarduumr Ground-Nut -RUtTS Cround-"ut t = = _ - *^~~' ter and .* ,_ 1 _ _ _ _-i i--t g .-u~~sk Mel.. ;- 1 : Chill d | | PPineaople Chii I( OTHER CROPS Ginger Jute (local)_ Tumer 1 c :it:(Tossa Ex STlNG CROPPING -PATTERNS AUS/Jute A T. Amon 5. AUS + Mustard/Puls*s/Rabi Veg- 8 AUS and S. Amon Mixed AUS # T. Amon * Khesari (Lathyrus) etables/Tobacco/Cotton/potato 9. Boro Jute 4 KaTai (Pulse) 6. AUS and Arhar (Cajanus)/TiI 10. Sugarcane AUS and T. Anmon ?Iijes (Sosamun) Mixed 11. Sumwner vegetables 7. B. Amon * Mustard/Pulses Source: Moseman (1980). 4 Table 1.2 Average Annual Growth Rates in Crop Production Area Production Yield Rice 0.37 2.11 1.74 (Aus) (-0.14) (1.77) (1.92 (Aman) (0.37) (2.42) (2.05) (Boro) (1.85) (1.59) (-0.36) Wheat 19.75 32.21 12.45 Pulses -1.1 -2.8 (-1.8) (Gram) (-1.0) (-2.8) (-1.8) (Mash Kalai) (Masur) (1.8) (-0.7) (-2.4) (Khesari) (0.4) (-1.3) (-1.8) Oilseeds -0.2 -0.4 (Rape & Mustard) (-0.2) (0.3) (0.4) Potatoes 1.5 1.0 -0.5 Sweet Potatoes 0.5 0.0 -0.6 Tobacco 2.0 2.1 0.1 Jute -1.58 -0.02 1.56 not significant Note: FY70 - FY81 for rice and wheat, FY70 - FY80 for the other crops. Growth coefficients are average annul rates of change (in percent) estimated on the basis of fitted least-squares trend lines. Source: BBS data, IBRD mission estimates, ad Edward J. Clay, Rabi Crops in Bangladesh: A Situation Report on Production, Prices and Research on Compaative Profitability (Dacca, Janauary 1981), Table 2. 5 plant resources and limited field and laboratory tests (Zaman 1975). Tea planters lost contact with their research institute, which was in India. A new tea research station was not established by the government until the late 1950s. Sugarcane improvement suffered because it was cut off from its source of genetic material. During this period, however, some research personnel were sent to the USA for PhD study. After 1956, well trained professionals returned from training, and research activity started to expand. However, the expansion of the Agricultural Research Institute suffered another setback in 1962 when the Pakistan government decided to take over the fields of the research institutes near Dacca to build the Second Capital of Pakistan. The land available for experiments was quickly reduced to about 4 ha and by 1965/66 no land at all was available for research. "Thousands of germplasm samples collected during the past 50 years at a tremendous cost were lost forever" (Zaman 1975). Financing of the Sugar Research Institute of Ishurdi was shifted from the provincial to the Pakistan central government. However, sugarcane research was returned to the province in 1966 but with no increase in budget. The 1960s were a period of growth for the Jute Research Institute which received a large increase in funds from the central government. However, its major research farm was also appropriated to build the Second Capital. It was left with less than two ha in the grounds of the Institute, and substations about 20 miles away from the Institute. In the mid-1960s the first foreign assistance for agricultural research started. The Ford Foundation funded IRRI to provide some technical assistance and to administer Foundation support for the development of a rice research institute in Joydebpur to replace the lands lost to the Second Capital. In 1970 USAID provided its first direct funds to research by partially funding the rice research program. Several other government institutions outside the regular crop research institutes started doing agricultural research during the 19608. The Academy for Rural Development at Comilla, supported by the Ford Foundation and others, accomplished a considerable amount of social science research throughout the 6 1960s. In the mid-1960s it conducted field trials on a number of rice varieties from SE Asia, China and Taiwan, several of which, especially Pajam, became very popular. It ran trials of IR8 and helped to popularize this semi- dwarf rice. It also introduced several varieties of vegetables and melons which gained widespread acceptance. In addition, an elaborate system for communicating and providing modern inputs through the village cooperatives was developed and supervised credit systems schemes were introduced. In the early 1960s the Bangladesh Agricultural University (BAU) at Mymensingh was founded with the assistance of the USAID and later also received support from the World Bank. By the end of the decade, research was under way in rice breeding and the social sciences. Another institution that developed during the decade was the Agricultural Division of the Dacca Atomic Energy Commission which later became the Institute for Nuclear Agriculture. This institution attracted very well qualified staff by offering relatively high salaries and prestige, but there is little evidence that it has produced any results with measurable impact. The major privately financed agricultural research program in British Bengal and Assam was the tea research system. At Independence, the tea research institutions were in India. The government then financed the Tea Research Institute in Sri Mangal in East Pakistan. However, there were still a few small, privately financed efforts to improve agricultural technology. Around 1960 the government gave permission for several of the sugar mills in North Bengal to import sugarcane varieties for testing on their mill farms. In the late 1960s the Pakistan Tobacco Company started some trials of Virginia tobacco varieties in East Pakistan. After President Ayub came to power in the late 1950s, the government took a more active role in agricultural development. In addition to the Academy for Rural Development at Comilla and the Agricultural University at Mymensingh, the government established two other institutions which supplied inputs and some extension advice. The first was the Agricultural Development Corporation (ADC) which provided fertilizer, pesticides and improved seed. 7 The second was the Water and Power Development Authority (WAPDA) which provided most of the irrigation facilities during the 1960s. It had its own extension service for certain project areas. The national agricultural extension service was also strengthened. Another deliberate step was raising the price of rice relative to inputs through government procurement and input subsidy policies. 1.3 Agricultural research in Bangladesh The War of Liberation disrupted work at all the research institutes. The Rice Research Institute (BRRI), thanks to strong leaderslip, survived better than most. The Sugar Research Institute (SRI) was virtually destroyed in 1971. It was shifted to the Bangladesh Sugar and Food Industries Corporation in 1973 and its research program was revived in 1974. The Central Jute Committee was abolished and the Jute Research Institute (JRI) was given autonomous status under the Ministry of Industries in 1973. The Agricultural Research Institute (BARI), which by 1974 dealt with only the "minor" crops, did not fare as well. In fact, it had no land of its own near Dacca to use for experiments until the mid-1970s. The Bangladesh Agricultural Research Council (BARC) was established in 1973 to plan and coordinate the research system as a whole, but it had little power. The Agricultural University (BAU) continued to teach but did little research in the period immdiately after Liberation. Since 1975 the research system of Bangladesh has been growing at a rapid rate (Figure 1.2). In the period between 1975 and 1980, the BARI budget grew by more than 50 percent annually. BARC and SRI grew by about 40 percent annually. BRRI, BJRI and the Forestry Research Institute grew by about 20 percent a year (Gill 1981). Only the Institute of Nuclear Agriculture showed signs of declining expenditure. There have been several important structural changes. BARI has been granted a considerable degree of autonomy from the Ministry of Agriculture. BJRI shifted to the Ministry of Science, then back to the Ministry of Industries, and has more recently returned to the Ministry of Agriculture. BARC has gained status and power as its budgets and technical FIGURE 1.2 Agricultural Research Expenditures on Crops (TK 000 at 1975176 Prices) 1950. 1951* 1952. 1953' i954 - 1955 . i956* r 1957 - 1958 . 1959 ' 0 1960 . * 1961* 1962 1963 5 1964, 1965 - 966. N 1967' 1968' 1969 . 1970. 1971' 1972 - 1973' 1974 . 1975 - 1976* 1977.- 1978 5 1979 * 196D0. 1986 * . 0 i59; 30161e' * *47 *- ** * * *' * **'9' ' 90 i5i258 4705 9 competence have grown. It is now playing an important role in encouraging communication between research workers, coordinating research efforts, and allocating resources between different institutes and research areas. The major non-government research efforts of the 1970s were by the Bangladesh Tobacco Company (BTC, formerly the Pakistan Tobacco Company), and by some of the non-government organizations (NGOs) such as local voluntary agencies, missionary and relief organizations. The Bangladesh Tobacco Company program has been very successful in transferring the production of Virginia tobacco to Bangladesh. There was no Virginia tobacco grown commercially in East Pakistan but, by 1976, Bangladesh was self-sufficient in Virginia tobacco production. This was due largely to an applied research program run on farmers' fields by BTC. The NGOs have had much smaller operations, but some of them such as the Mennonite Central Committee in Comilla and Noakali, RDRS in Rangpur, and CARE around Dacca have played useful roles in introducing improved varieties of vegetables, soybeans, rice and wheat, not to mention their other wide-ranging contributions including some innovative machines. In the national agricultural research system it is also important to understand the relationship of the research institutions to the ministries which control their funds and to the extension services which communicate their findings of research to the farmers. Figure 1.3 indicates the current structure of the research service. A research institute such as the Sugar Research Institute has to compete for funds in the Ministry of Industries against other uses of funds such as improvements for the sugar factories or edible oil factories. In the case of the institutes under the Ministry of Agriculture, the competition for funds is with other programs such as extension and the Bangladesh Agricultural Development Corporation which supplies inputs, and also with other research institutions. The extension agencies have been consolidated recently under the Ministry of Agriculture. The major extension service outside that Ministry is the sugarcane extension program. 10 Figure 1.3 Agricultural Research Organizations Ministries Institutions Ministry of Agriculture Bangladesh Agricultural Research Council Bangladesh Agricultural Research Institute Bangladesh Rice Research Institute Forest Research Institute Livestock Research Centre Fisheries Research Institute Bangladesh Jute Research Institute Bangladesh Agricultural University Institute of Nuclear Agriculture Ministry of Industries- Sugarcane Research Institute Sugar Mills Corporation Ministry of Commerce - Tea Board Bangladesh Tea Research Institute Ministry of Local Government Bangladesh Academy of Rural Development and Rural Development - at Comilla and Bogra Cooperative and Rural Development Ministry of Education Bangladesh University of Engineering and Technology Dacca University Chittagong University Rajshahi University Ministry of Finance and Bangladesh Institute of Development Planning Studies Science and Technology Bangladesh Council of Scientific and Industrial Research 11 2 History of IARC Activities 2.1 IRRI IRRI's first contact with the Bangladesh agricultural research system came in 1965 when the Assistant Director of IRRI was invited by the Government of East Pakistan and the Ford Foundation to visit. He was invited "to observe rice research, and offer suggestions on an accelerated rice research program in cooperation with IRRI" (McClung 1965). Soon afterwards, an IRRI staff member was placed in Bangladesh to initiate a presence that has continued to the present, except for a brief interruption because of the War of Liberation in the early 1970s. The number of IRRI staff members grew to four in 1970. They included a rice adviser, assistant rice adviser, a civil engineer who provided advice on construction projects and a plant pathologist. Only the rice adviser returned in 1972 after the obligatory exit for the War of Liberation. In 1974 the IRRI staff grew again and since then has consisted of 4 or 5 advisers stationed in Bangladesh. The 1984 staff consists of a research systems specialist and liaison scientist, deep water rice breeder, an agricultural economist and a rice production specialist working jointly with BRRI's staff. In addition to the long-term advisers, many IRRI staff have made short visits to Bangladesh. A major activity of the IRRI program from the earliest period has been training. In 1966 the staff of the Agricultural Research Institute consisted of about 130 people with at least a bachelors degree. Of these 3 had PhD. and about 36 had MAg or MSc degrees, mostly from Pakistani universities and colleges. In 1966 the first of a series of groups of 10 scientists were sent for training to the University of the Philippines at Los Banos (UPLB) and to the USA for post-graduate degrees (Alim 1966). There has since been a regular flow of trainees as shown in Table 2.1. In addition to agricultural scientists, by 1970 BRRI had sent 17 senior agricultural officials and planners on short-term training trips. These trips Table 2.1 Sunmary table of training cnMF!r CIP DR Year Degree Productim Specialized Total DqgrF Production Spcialized Total Degree Production Specialized ToLal related related related 1962 - - - 1963 - - - - - - - - - - 1 1 1964 - - - - - - - - - - - - 1965 - - - - - - - - - - I 1 1966 - - - - - - - - - - - - 1967 - - - - - - - - 2 4 2 8 1968 - I - 1 - - - - 1 3 3 7 1969 - - - - - - - - - 3 3 1970 - 1 1 - - - - - - 1 2 3 1971 - - - I - 1 1972 - - - - - - - - 2 - - 2 1973 - 3 - 3 - - - - I - - 1 1974 - 2 - 2 - - - - - 2 1 3 1975 - 2 - 2 - - - - 3 1 1 5 1976 - 6 - 6 - - - - 1 - 3 4 1977 - 4 - 4 - - - - 1 3 4 8 1978 - 7 - - - - - - 6 4 8 18 1979 - 7 - 7 - 2 5 7 2 7 11 20 1980 - 10 - 10 - 1 1 14 13 13 40 - 1981 - 9 2 11 - 7 7 14 10 12 16 38 1982 - 5 - 5 - 19 4 23 4 12 27 43 1983 4 5 - 9 - 23 4 27 9 2 2 13 1984 5 4 - 9 7 - 1. 8 14 6 3 23 lInGUL 9 66 2 77 7 52 21 80 71 70 101 242 Table 2.1 Summery table of trainirg (cont.) I rrA IBPGR ICARDk Year Degree Productim Specialized Total Degree Production Specialized Total Degree Production Specialized Total related related related 1962 - - - - - - - - - - - - 1963 - - - - - - - - - - - - 1964 - - - - - - - - - - - - 1965 - - - - - - - - - - - - 1966 - - - - - - - - - - - - 1967 - - - - - - - - - - - - 1968 _ _ 1969 - - - - - - - - - - _ _ 1970 - - - - - - - - - _ _ _ 1971 - - - - - - - - - 1972 - - - - - - - - - 1973 - - - - - - - - - _ _ _ 1974 - - - - - - - - - 1975 - - - - - - - - - _ _ _ 1976 - - - - - - - - - 1977 - - - - - - - - - _ _ _ 1978 - - - - - - - - I - 1979 - 1 - I - - 2 2 - 2 - 2 1980 - - 1 1 2 - 7 9 - 1 - 1 1981 - - 1 1 - - - - - I - 1 1982 - - - - - - - - - 1 - 1 1983 - - - - 2 2 - - - - 1984 - - - - - - - - - - - - IUI7IL 0 1 2 3 2 0 11 13 0 6 0 6 Source: Badruddoza (1984). 14 Table 2.1 Summary table of training (cont.) ICRISAT Year Degree Production Specialized Total related 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 -1 1975 _ 1976 - 1977 - 1978 - 1979 - 1980 - - 2 2 1981 - - 2 2 1982 - - 3 3 1983 - - 1 1 1984 - - 1 1 TOTAL 0 9 9 Source: Badruddoza (1984). 15 went to IRRI to see the development of new techniques and then elsewhere in the Philippines or Japan to observe new technology as it was applied by farmers. It can only be speculated that these trips were quite useful in convincing policy makers of the importance of rice research and the potential of the new high yielding technology. The influence of IRRI varieties and germplasm was felt almost immediately. In 1966 IRRI sent 303 selections of improved rice for testing. Before the first tests were harvested the government decided to order over a ton of rice from IRRI for commercial seed multiplication. This order included both IR8 and IR9, which had performed best in the initial trials. Since then IRRI genetic material has regularly been sent to Bangladesh including replacements for materials lost from the BRRI rice gene bank. 2.2 CIMMYT The first contact of CIMMYT with the Bangladesh research system was also during the East Pakistan period. In 1965 a kilogram each of Sonora 64 and Penjamo 62 were sent to East Pakistan from West Pakistan. These varieties yielded almost five times the provincial average wheat yield and matured more rapidly than the local wheats. On the basis of these results 160 kg of seed vas sent for testing in the 1966-67 season. These trials were also impressive. The CIMMYT representative in West Pakistan visited these sites in 1968 and provided some initial technical assistance to the scientists. The next year the CIMKYT representative, along with West Pakistani wheat breeders, sent a collection of 265 promising Mexican lines to be tested. The same year the first Mexican variety was released to farmers - based on 1212 tons of Penjamo62 (Zaman 1968). Since 1970, genetic materials have come directly from CIMKYT. Perhaps the second most important input from CIMMYT, after genetic material, has been training. This is epitomized by the three early wheat production trainees who have become the nucleus of today's Bangladesh wheat research team (Badruddoza 1984). During the period 1968-82, 46 Bangladeshis have spent at least one full crop cycle at CIMMYT in wheat-related training 16 courses. There were 15 trainees in experiment station management. Of those, 12 were from the Bangladesh Agricultural Development Corporation which undertakes seed multiplication of wheat and other crops for the government. Unlike IRRI, CIMMYT had no staff posted in Bangladesh until 1982. However, there were visits at least annually during the wheat season by CIMMYT staff. These visits have served to strengthen the research program in a variety of ways. They have helped the local researchers solve some of their technical problems. They have provided new ideas and advice, and generally helped to sustain researchers' motivation. In addition they played an important role in persuading the government to push wheat in its development strategy. The relationship with CIMMYT was particularly strengthened by frequent visits by Dr G. Anderson and Dr E. Saari who, with active support from senior national administrators, sensitized the government to the possibility that wheat cultivation in Bangladesh could be vital for food production. The government then decided to import 4,000 tons of Mexican varieties from India and Mexico in 1975/76, and wheat production was firmly launched. 2.3 CIP CIP started working in Bangladesh in 1979 when a consultant, supported by the World Bank and the Australian Government, was posted to the Bangladesh Agricultural Research Institute. Several CIP staff members have visited the project to provide special skills. By 1983 the project had sent two research workers for short-term training at CIP, seven officials of BADC and BARI to seed production courses in India and Scotland, and three research workers to Sri Lanka and the Philippines to study diffused light storage (Bangladesh PIP 1983). The first collection of CIP material was tested in 1980-81 and a supply of genetic materials has been available from CIP since then. 17 2.4 Other CG Centers IFPRI has carried out several studies on agricultural price policy and one study of the food aid program in Bangladesh. Several members of its staff make regular trips to Bangladesh. The main contribution of ICRISAT, IITA and ICARDA until 1984 has been the provision of germplasm of their mandate crops. A few Bangladeshi scientists have also received training at these institutions. 18 3 The Role of the CG Centers in the Evolution of the National Agricultural Research System The major actors in the development of the Bangladesh research system were obviously the Bangladeshi research personnel, administrators and politicians. Others have been important at certain times. The Ford Foundation played a very important role from the mid-1960s until the mid- 19709. During the 1960s, USAID invested heavily in the Agricultural University and in 1970, it began significant support of research. It has since become the largest foreign contributor of external funds for research. In the mid-1970s the World Bank began lending money for research, university and extension facilities, especially infrastructure. Many bilateral agencies have since started to support research. Table 3.1 mentions the donors active in 1982. The role of the IARCs is quite closely related to the activities of the donors - particularly the Ford Foundation, USAID, the World Bank, ADAB and CIDA. Most of the activities of the IARCs within Bangladesh are sponsored by donors rather than from either local funds or core funds from the centers. The centers have provided technical assistance, training, germplasm, networks of scientists and publications. Other organizations provide many of these same services in Bangladesh. Local institutions also provide many such services, but most are provided by foreign organizations. FAO has provided technical assistance, training, networks and publications. IADS, which is closely related historically and intellectually to the centers, and other consulting firms, provide technical assistance. Universities in industrial and some developing countries provides technical assistance, training, some networks and training. The research system also has effective contacts with non-CGIAR institutions such as AVRDC and IFDC. 19 Table 3.1 Estimated Research Expenditures and Associated Donors Research Annual Expenditure Average as Percent of Beginning Expenditure Value of Crop of Breeding (million taka) Program Donors (1982) Rice (BRRI) 26.46 0.08 196/67 IDA, Australia, Canada, Ford Foundation, USAID Jute (BJRI) 8.12 0.35 1978 ADB,UNDP Sugarcane (SRI) 6.76 0.39 1975/76 Australia Tea (TRI) 4.66 0.62 d/ UK, EEC BARI 54.50 0.89 (0.55)c/ T968/69 Wheat USAID, IDA Cerealsa/ (13.10)b/ 1.47 1978 Maize Canada (CIDA & IDRC) Pulses (4.40) 0.74 (0.37) 1979 IDRC Oilseeds (9.80) 0.91 (0.61) 1974 Sweden Spices (3.80) 0.28 Fruits (6.50) 0.32 Japan, ADB Roots and Tubers (7.10) 0.50 1977 White Netherlands Potatoes Other Vegetables (6.00) 0.99 (0.33) Japan Fibers/Tobacco (4.40) 1.55 ?Tobacco EEC (Cotton) BARC 4.34 USAID, IDA IRA 6.86 Sweden a/ Cereals include wheat, corn and millets. b/ BARI gives an estimate od the current allocation of manpower among crops; this was used to allocate BARI's expenditue among crops. c/ This uses value of crop adjusted as suggested by pray, 1980. d/ Relatively small selection. Program centers on Manyjuri and hybrid jates - they do not include commercial varieties. Sources: Gill (1981) and BARI (1980). 20 3.1 Size of research budgets The IARCs seem to have had some influence on the decision makers who increased the budgets for agricultural research, extension, education and subsidies during the late 1960s. First, the centers' success at developing modern rice varieties in the Philippines and with modern wheat varieties in the Punjab, and the publicity that they received, enabled local research advocates to convince administrators and politicians that research could be effective and that investment in research should be increased. Second, they were effective in convincing policy makers (recall the IRRI-sponsored trips to the Philippines and Japan to see HYVs in action). HYVs were also grown on the farms of politicians. Third, the early rice and wheat successes convinced many donors that research projects were useful to support. Many of the more senior research administrators in Bangladesh perceive the whole agricultural research thrust to be under active threat from the government, which may be suffering some of the same "fatigues" that are purported to be held in the donor community generally. Needless to say, a military government's perceptions of priorities does not always give due attention to agricultural research, which usually claims only a rather small part of the country's budget and, given its long-term orientation, is subject to regular scrutiny as to its appropriateness vis-a-vis other issues of national development or defense. The centers really have an important role to play in maintaining credibility of the research thrust in Bangladesh, and this may well prove eventually to be their most important input, namely, helping to protect the already strong national commitment to research investment. 3.2 Commodity allocation of resources It appears that the centers also helped the Bangladeshis to shift their research resources from cash crops to the major foodgrains. In the early 1960s, jute received the largest share of financial resources. By 1974, the budgets of both the Rice Research Institute and the Agricultural Research Institute had surpassed the Jute Research Institute (see Table 3.2). In 21 addition, in the 1970s, there was a considerable boost to the resources invested in BARI, which is concerned with the other crops of the IARCs such as wheat, maize, potatoes and pulses. When compared with the value of these crops, it still seems that there may be some under-investment in rice in Bangladesh (Table 3.3). The ratios of the expenditure on research to the value of the crop (research intensities) indicate that rice is still receiving far less than its share of the research, while tea and other food crops are receiving far more than their share. In column 1 of Table 3.1, the other crops research expenditure is arbitrarily partitioned on the basis of the number of BARI scientists working on each crop in 1979. This shows that cereals other than rice - wheat, maize and millets - receive a far bigger share of research than their share of total output. In fact, they had the largest research intensities except for tobacco and cotton. This shift of resources to the crops of the IARCs has been closely related to the centers' activities. As mentioned, the rice successes in the 1960s led policy makers to invest in rice research. In the 1970s, the wheat success story has been the major justification for both government and donor support for BARI. The hope for similar successes in maize, pulses, oilseeds and potatoes has motivated the large investment in that institution. Encouraging governments to encourage the culture of wheat, potatoes and maize in regions where there is little demand for them, or where the growing conditions are far from optimal, is clearly questionable. However, there are examples of areas of Asia that were once considered to be unsuitable for production or where there was little demand, where HYV rice and wheat are now grown. The case of wheat in Bangladesh itself is a prime example. Bangladesh now needs to step back and ask whether this same strategy will work with maize and potatoes. Perhaps too many research resources are being devoted to them? 22 Table 3.2 Allocation of Agricultural Research Expenditures (Tk 000 current) Crop/ 1950- 1960- Institution 1960 1965 1967 1974 1979 1980 Rice 176a/ 300 1,230 6,612 28,330 39,870 Other Food Crops 189a/ 200 n.a. 12,531 77,153 129,230 Jute 153 500 2,000 2,159 10,210 11,000 Tea 0 n.a. n.a. 2,233 6,015 5,110 Sugarcane n.a. 70 101b/ 550 10,330 13,590 BARC 0 0 0 n.a. 5,000 10,575 Livestock n.a. 644 n.a. 2,800 n.a. n.a. a/ Expenditure on botanists multiplied by 1.5 to account for entomologist, plant pathologist and others. b1/ Budgets Source: Pray and Ahmed (1985). 23 Table 3.3 Research Intensitya for Agricultural Research Crop 1950s 1965 1967 1974 1979 Rice 0.04 0.04 0.12 0.26 0.67 Other Food Crops n.a.b n.a. n.a. 2.20 8.00 Jute 0.29 0.61 1.78 1.50 2.60 Tea n.a. n.a. n.a. 20.60 9.70 Sugarcane n.a. 0.29 0.32 0.56 4.70 a - (value of agricultural research/value of output) x 100 at current prices b - n.a.- not available Source: Pray and Ahmed (1985). 24 3.3 Institutional structure IRRI appears to have influenced the organizational structure of agricultural research in Bangladesh in several ways. One of the first things done after IRRI's initial work in Bangladesh was the organization of a multidisciplinary rice research program within the Agricultural Research Institute. Eventually, local rice research workers, with the help of IRRI and the Ford Foundation, were able to set up the relatively autonomous BRRI under the Ministry of Agriculture. The new institute was modeled closely on IRRI (Alim 1966). Instead of the disciplinary divisions of ARI, the divisions of the institute are the same as IRRI's (a) rice varietal improvement (b) physiology, (c) agronomy, (d) pathology, (e) entomology, (f) soil chemistry, (g) economics and statistics (h) rice technology (i) agricultural engineering and (j) communications. In addition, considerable emphasis was placed on multi-disciplinary, problem-solving research. To provide this interaction among disciplines, BRRI set up seven task forces which met three times a year to discuss the problems on which to concentrate, how these problems should be tackled, and to assign projects or parts of projects to individuals for implementation. The task forces deal with varietal improvement, fertility management, cultural practices, cropping systems, adaptive research and training, pest management and agricultural engineering. 3.4 Goals and priorities within the commodity program The goals and priorities of the commodity research programs appear to have changed since they came in contact with the centers. It appears that, before 1965, rice specialists placed more emphasis on the t.aman crop and the aus crop with less emphasis on boro and b.aman. By 1961, 25 varieties of t.aman and 17 varieties of aus, 6 varieties of b.aman and 6 varieties of boro had been developed and approved for use by agricultural scientists of British Bengal and Esst Pakistan. With the success of IR8 in the boro season, more emphasis appears to have been placed on this crop, although t.aman and aus still received much attention. In contrast to the earlier period between 1967 and 1980, BARC reported that 6 varieties were released for the boro season, 3 25 for aus, 5 for the t.aman and none for b.aman (BARC 1983). More attention has been devoted to b.aman since 1975. Local rice workers had already been searching for varieties that were both resistant to lodging and fertilizer responsive before IRRI arrived. They were involved in the FAO Japonica-Indica crossing program that was aimed at producing fertilizer-responsive varieties. One change after IRRI's involvement may have been to screen varieties at higher levels of fertilizer. Another change seemingly associated with IRRI was the introduction of agricultural economists into multidisciplinary rice research teams. When IRRI staff arrived, there was no division of agricultural economics in the Agricultural Research Institute. The new rice research institute established a division of economics and statistics, and the project sent people for training in agricultural economics. The IRRI resident teams in recent years have included agricultural economists. Another change has been the introduction of cropping systems research by BRRI and IRRI staff in the mid- 1970s. This represents a significant step towards implenting a farming systems research orientation to work in Bangladesh. Both economics and cropping systems programs have helped to change research priorities in other areas by providing evidence that the prevalent priorities were incorrect or that there were new problems being faced by farmers. The studies of b.aman yields, which were carried out jointly by economists, entomologists and cropping systems agronomists, showed that the yield goals set by the breeders of b.aman were already being matched or surpassed by the yields of farmers (Clay 1978). This led to modification of research goals in the deep water rice program. The cropping systems workers played an important role in identifying zinc and sulphur deficiencies as important problems in some areas. This led to a major research and extension effort on these problems. The potato program has had a more dramatic reorientation than either the wheat or rice programs. Most of the 1980 research resources, for instance, 26 were concentrated on testing Dutch and a few Indian varieties under high fertility conditions and good water control. Now the program is selecting varieties from a wide range of sources, including a number of tropical varieties. A breeding program has been established which uses both exotic material and material from a collection of local varieties that has been established. In addition, completely new programs have been started on inexpensive diffused light storage facilities, reducing the cash cost of potato cultivation, producing potatoes through true seed, and developing a disease-free seed production facility in northwest Bangladesh. The dramatic shift is due to the joint efforts of the BARI staff in the new Potato Research Center, CIP and IADS. There was a quantum leap in the amount of potato genetic material screened. About 100 Dutch and Indian varieties were tested by BARI before 1980 (Ahmad 1980). In 1981-82, 2426 progenies from different families were tested in initial evaluation trials and 39 genotypes were studied in secondary and advanced variety trials (Sikka 1982). Undertaking research as a multidisciplinary team seems to have been a concept introduced by the centers, and used most widely at BRRI and increasingly also at BARI. 3.5 New research methods New techniques of rice breeding were introduced, as they were elsewhere in Asia. The use of new breeding techniques in maize is now practiced. In potatoes, CIP staff introduced new techniques such as the use of true seed to produce disease-free tubers, and possibly later for commercial production. Improved methods of screening varieties for resistance to pests have also been introduced. Cropping systems research was first introduced to Bangladesh by BRRI/IRRI and later something of the CIMMYT version was taken up by BARC and 27 BARI with the assistance of IADS. This gave institutional sanction and resources to on-farm and interdisciplinary research. 3.6 Efficiency of the research system A number of ways in which collaboration has affected the efficiency of the research system have already been discussed. There are, however, some further and more subtle effects. In general they are factors that motivate researchers to be more effective. The centers provide a new set of rewards for "good" research and in some cases have changed the definition of what "good" research is. "Good" research includes on-farm research that meets farmers' needs. Good research may or may not be publishable in a national or an international research journal - it depends on the center one is working with. Rewards include trips to other countries for conferences, scholarships for foreign degrees, approval and praise from a new group of peers - research and administrative personnel in the CGIAR networks in which Bangladeshi researchers participate - and ultimately perhaps a job at a centers or in other international organizations. In addition to rewards, the training programs of the centers emphasize certain types of research, such as on-farm research, which gives the research a more practical bent than perhaps would otherwise have been the case. While the definitions of good research may not have universal approval, it seems clear that they are superior to the old system where one was promoted primarily on the basis of years of service. This influence of the centers can be detected in the styles of the research workers within the different national programs, which have varying level of contact with such outside agencies. A strong contrast can be easily observed, because of their geographic contiguity, between BRRI and BARI. The substantial interaction with IRRI has led to quite different working styles at BRRI where people tend to work long hours, especially in the field, relative to those of nearby BARI where there has been relatively little after-bours work, and the more senior research personnel seldom get into the field. Part of this is due to the housing situation, since there is insufficient housing at BARI, but much of it is also due to contact with centers. The wheat 28 program is the exception at BARI, which manifests the influence of CIMHYT on work styles. There is evidence that a period spent at a center encourages a more committed and effective attitude towards problem-solving work. Again, some of the more negative attitudes that may be in need of correction may emerge from research workers' early educational experience at BAU, where there seems little philosophy of "hands on work" with regard to agronomic and other research activities. 29 4 Local Perceptions of the Centers 4.1 Nature of collaborative research in Bangladesh This section attempts to summarize the perceptions of the centers by the administrators and researchers of Bangladesh. It is based on discussions held by J.R. Anderson during a visit in March 1984 with the research administrators and program leaders of the national research system. In addition, a questionnaire was sent to a random sample of the research staff of the national agricultural research system. Forty scientists from a wide range of institutions and levels within those institutions replied in this first survey. The third major source of perceptions was the report of Dr Kazi M. Badruddoza (1984) "Training in the CGIAR System: The Case of Bangladesh" which reports on a survey of Bangladeshi scientists about the CGIAR training. In general, the centers active in Bangladesh are highly regarded for their independence and objectivity. The fact that they are perceived as having ample resources to assist research and to provide opportunities for scholarships and communication with foreign scientists reinforces the positive feelings about the centers. The two surveys indicated far more contact with IRRI than any other center. In the first survey IRRI had by far the greatest number of contacts, followed by CIMYT, CIP, ICARDA, and IITA. In terms of training, IRRI had trained 242 out of a total of 431 trainees in CGIAR centers' training programs. CIP was next with 80 and then CIMMYT with 77. The others had 13 or fewer trainees. The respondents in the first survey were asked to assess the importance of the centers in building up the national system. They felt that the centers were moderately active in Bangladesh. Not surprisingly, they rated IRRI and CIMMYT as the most active followed by ICARDA, CIP and ICRISAT. In addition, IRRI and CIMMYT were rated as the centers whose services were of most practical use and most helpful in building research capacity. There were too 30 few respondents to know which services were most important or which alternative source of information and services were important. The major group which indicated they had had little or no contact with the centers were faculty members at the agricultural university and the biological and soil science departments of the general universities. Some of them received publications but few had any other contact. They stated that they would like to but have not had the chance. The nature of the contacts came out especially in the longer discussions with senior officials. Most of the early center contact was with IRRI which, in the initial years, was perceived to be too "top down". Typically, IRgI staff were perceived as having too fixed an idea of what was needed within Bangladesh, without listening to the perceptions of local authorities. This was a problem with regard to some of the deficiencies in the varieties that were initially introduced, and with individuals being too dogmatic about a chosen research approach which frequently turned out to be inappropriate. It should be noted, however, that by the mid-70s, IRRI had learned from its early experiences. It had become much more sympathetic to the expressed local needs (e.g., for training) and is very highly regarded in Bangladesh in recent years. There are broadly similar experiences with regard to CIMMYT within the country. Perhaps profiting from the experiences of others, CIP has done a rather better job, in spite of some criticisms that it may be too blind to agricultural realities beyond potatoes. Such enthusiasms, however, are inevitable with commodity-specific centers. The germplasm received from the centers has been a major input into the research system. In the first survey it was the second most frequent contact point with the centers, following publications and slightly ahead of training. It was also rated as one of the most important services provided by the centers. Discussions with research leaders also emphasized its importance. A number of scientists mentioned the importance of even small amounts of external assistance. For instance, the external assistance to the rice 31 research program provided by the Ford Foundation, USAID, and Australian and Canadian sources oftentimes proved very useful in eliminating the impact of domestic constraints on operation. For example, there are many difficulties in importing scientific equipment through very restrictive input regulations. International organizations sometimes can help to circumvent these and thus to facilitate more effective research implementation. Similarly, there are many constraints imposed on travel arrangements of Bangladeshi research personnel. Most junior officers cannot use aircraft for domestic travel, which can conceivably hamper flexibility and perhaps the critical timing of, for example, a visit to an outbreak of a rice disease in a distant part of the country. Relatedly, there are many restrictions on travel outside Bangladesh by research personnel at all levels. Intervention by the centers can and does help to relieve some of the restrictions. The usefulness of these key inputs from the centers is, in part, related to the usefulness of foreign aid in general, whether it is distributed by the centers or other agencies. However, in the absence of the centers, it is not clear whether these resources would indeed be available to the country, and if they were, if they could be protected from the viscitudes of the regular budgetary process. Both IRRI and CIMMYT have actively fostered FSR-related type activities in Bangladesh. Local authorities see this as being important in broadening the perspective of agronomic and related research. They see a continuing need for the centers in such fledging work. There is still some distance to go in matching experimental orientation with on-farm reality. For instance, it seems that most of the research sites claimed to be "rainfed" have, in fact, been irrigated. The area of human capital improvement is the main topic of Dr. Badruddoza's report, and so some of the main points of his report are taken along with observations from our own field discussions. There is considerable agreement that this is one of the major contributions of the centers. Short- term training has been the most important in terms of numbers of people. 32 However, there is a continuing need for assistance to accomplish more degree training. The practical character of training at the centers was particularly appreciated. "Trainees are required to spend a considerable part of their time in the field or laboratory and learn practical research methods applica- ble to actual field conditions. This is a contrast to other training programs where emphasis is placed on classroom study. The general feeling and experience is that IARC training programs assist in development of practical vision and an ability to handle field programs more efficiently" (Badruddoza 1984). The centers have regularly played a "brokers'" role of placing people effectively in advanced tertiary institutions or in appropriate training. This brokerage role depends on informal contacts with the concerned institutions and in some cases also involves a period at a center for thesis work that is perceived to be more relevant than the typical thesis work conducted in temperate-climate advanced educational institutions. 4.2 Concerns There has been some criticism expressed in the country as to the degree of "ipampering" by the centers of domestic institutions. Part of this criticism relates to judgements about the inherent qualities of local staff who, in some cases, have been perceived as not really being adequately prepared for effective collaborative work. Some of this criticism is directed at the style of teaching at the Bangladesh Agricultural University (BAU) which, at least in some departments, may offer professional preparation of a somewhat limited level of achievement, especially in regard to developing capacity for independent thought. One high-level bureaucrat expressed a fresh perspective on combinations of multilateral and bilateral assistance. The concept was one of "mixing one's drinks" in assistance terms, because, when both are somevhat 33 competitively supplying assistance, the recipients can easily become intoxi- cated by the readiness with which resources are made available to them. The implication is that some research institutions have more resources than they can effectively handle at the moment. Some concern was expressed about representation on boards of governors of the centers. The main concern was that, if partner governments did not have direct representation on such boards, unfortunate gaps could develop betveen what centers perceived to be national priorities and those articulated within the government concerned. The criticism being expressed was that centers really needed an "official" brief from the partner governments to facilitate maximal coherence with individual national priorities. With regard to training, it is felt by some reflective observers in Bangladesh that the CGIAR system should switch its emphasis from short-term training to more advanced degree training, to help develop greater leadership capability within the national system. It was also felt that the in-country and regional training programs in Bangladesh by CIP are examples of what should be done in the future for short-term training. One significant issue that has not yet been adequately addressed in the research programs concerns soil salinity. Apart from some rice selection work in Barisal nothing of substance has yet been done to address this problem, but it is to be hoped that eventually the national and international institutions concerned will embark with some vigor on plant selection and a search for agronomic practices to mitigate the influences of high levels of salt on agricultural productivity. One suggestion that emerged from discussions with senior officials in Bangla-esh concerned the staffing arrangements of the centers with personnel from partner countries. The idea was that a period of service in an interna- tional center could be useful to a national research worker, but the tendency is for such people to be economically and bureaucratically "trapped" for longer periods, after which it is difficult for them to reenter the 34 professional service from which they departed. If centers had a rule for a fixed maximum term for all their staff, this would facilitate an exchange arrangement with national personnel into center staff with mutual benefit all around. It must admitted that this would force additional difficulties in an already challenging recruiting environment for all the centers but, given the tendency to experience differential turnover rates among different nationalities, there may well be merit in such a suggestion. The centers have been criticized by some for their lack of strong attention to developing "appropriate" technologies in the field of mechanical aids in general. The machines developed by IRRI, for example, are not as low in cost as they could be. For instance, some of the NGOs working in Bangladesh have developed extremely low-cost irrigation pumps which seem appropriate for their simplicity of construction and their reliance on mainly local natural materials, such as bamboo. IRRI has recently acquired a copy for study in the Philippines. Government agencies in general tend to be mostly concerned with the technological problems of large-scale farmers and tend not to concentrate on the problems of small-scale farmers. The complex problems inherent for international organizations working in difficult environments are epitomized by the work on deep-water rice. On the one band, IRRI has been criticized by many for not giving sufficient attention to improvement in this form of production. On the other hand, in spite of the considerable effort that has been addressed to this particular difficulty, which is particularly a problem in Bangladesh and other delta areas in Southeast Asia, there has been very little progress made. Such are the challenges of development. 35 5 Impact on Other Institutions 5.1 Extension The Bangladesh Agricultural Development Corporation (BADC) has improved the effectiveness of its seed production capacity through its contact witb the centers. The main direct contact has been through training at the Centers. BADC staff members have attended courses on seed production and farm manage- ment at IRRI, CIMMYT and CIP. BADC expanded greatly with the increased demand for improved seeds, fertilizer and irrigation that the HYVs of rice and wheat inculcated. In recent years, its importance has declined somewhat as private companies have been allowed to play a larger role in the distribution of inputs, but its rapid growth in the late 1960s and the decade of the 1970s was due in large part to the research activities both nationally and internationally. BRRI and BARI have been actively working to close the gap between research and extension. The IRRI rice production specialist works full time on this activity. Some extension staff from both the Department of Agriculture and the Water Development Board have gone to IRRI for training on various aspects of rice production. Since June 1974 more than 2000 people have graduated from BRRI's various, short courses. This includes 337 people who have graduated from the 4-month rice production course. Nearly 200 of the people from the 4-month course were mid-level technicians from the Extension Service of the Ministry of Agriculture. In total, almost 600 extension personnel have received some training (IRRI 1984). They also include agents from the Water Development Board, NGOs and BADC. BRRI staff have then provided technical backstopping to these agents on their return to the field. In addition, these extension agents are then involved in BRRI's various networks like the advanced-line adaptive research trials, the farmers' varietal observation trials, and the multilocation testing of the farming systems network. BARI is carrying out similar programs to improve the links with extension. They have been particularly successful in the northwestern 36 part of the country, where a World Bank project has been fostering these links as part of a project to strengthen extension. BRRI and BARI have had active contact with the agricultural programs of the NGOs. IRRI staff members have been active participants in ADAB (Agricultural Development Agencies in Bangladesh) which is the umbrella organization for many of the NGOs working in agriculture. BRRI and BARI staff have provided articles to the ADAB newsletter, made presentations to ADAB meetings, and visited the sites of NGOs in the villages. The wheat, maize, and millets, programs at BARI have also worked closely with those NGOs that have more technical capacity. The Mennonite Central Committee, in particular, has had close links with BARI as well as BRRI. It tests their technologies and participates in their training and workshops. Another important link with the NGOs is the BRRI cropping systems network. This network contains sites that are run by various NGOs. BRRI has provided training, technical backstopping and, in some cases, materials such as the seed of new varieties for testing. The network meets at least once a year to discuss results and to plan the next year's activities. This network provides the NGOs with the latest technology from BRRI which is spread through their organization, and provides BRRI with information about the problems of farmers. Zinc deficiency has been discovered at several of the NGO sites and applications of zinc fertilizer allowed some farmers to grow their first successful crop in years. 5.2 Policy Senior scientists from BRRI and BARI play an important role in formula- ting the seasonally intensive rice and wheat production schemes that the extension service and BADC carry out. Most of IFPRI's impact has been through and on organizations that are not primarily agricultural research institutions. These include the Bangladesh Institute of Development Studies (BIDS), the Planning Cell of the 37 Ministry of Agriculture, and the Planning Commission. The joint research project on food aid with BIDS shifted some BIDS resources to an examination of this topic that is very important in Bangladesh. IFPRI publications have been widely circulated as models of policy research analyses. In addition, the knowledge embodied in the IFPRI staff and its published research has influenced the policy of international organizations such as IFAD and the World Bank when their staff members have worked as consultants in Bangladesh. 38 6 Impact of Collaborative Research on Farmers and Consumers 6.1 Productivity growth Agricultural productivity has risen slowly in Bangladesh. The estimate of the growth in output for the period from 1948 to 1981 is 2.4 percent, while total factor productivity has increased by 0.78 percent annually (Table 6.1). Figure 6.1 shows the growth of output, inputs and productivity for the period. For the first decade after Independence from Britain, productivity declined. Since then there has been slow growth. Figure 6.2 indicates that there was almost no growth in land inputs, while current inputs grew from a very low base at a high rate - almost 6 percent annually. The data on individual crops (Table 6.2) show considerable diversity among crops and periods. The three crops which had the largest increases in yield per unit area for the whole period were boro rice, wheat and potatoes. The growth in the first two was surely largely due to the collaborative research of the KARS and the IARCs. The growth in potato productivity came largely in the 1960s before CIP existed. The impact of the collaborative work of the IARCs and the Bangladesh KARS has come primarily through the development of the new varieties and the practices that are related to them. The main wheat and rice varieties that have at least one parent from the centers are listed in Table 6.3. The spread of HYV rice and wheat is shown in Table 6.4. The problem of attribution in plant breeding work is epitomized by the Bangladesh wheat experience. The most important variety was Sonalika, bred in India using materials from CIMMYT and its predecessors based on "Mexican" material which actually had most of its origins in the USA and Japan. This is a clear case of successful transmission of technology across national boundaries, with assistance from multilateral support agencies such as CIMMYT. Related to the difficulties in attribution is the appropriate costing of the plant breeding materials that are used in the programs of selection for improved varieties, but which never find their way into the final selections. 39 Table 6.1: Analysis of Agricultural Growth FY48-FY81 FY48-60 FY60-FY71 FY71-FY81 1. Total Production 2.40 0.41* 2.12 3.53 2. Total Conventional Inputs 1.62 1.15 1.82 2.56 3. Total Factor Productivity 0.78 -0.74 0.29* 0;97 4. Net Sown Area: (NSA) 0.12 0.22 0.43 -0.06 5. Net Cropped Area: (NCA) 1.10 0.08* 2.07 0.89 6. Male Agricultural Labor Force 1.77 1.79 1.41 2.77 7. Draft Animals 0.84 0.98 0.80 1.27 8. Value of Capital Services 1.19 1.00 1.44 1.58 9. Value of Current Inputs 5.60 1.77 8.77 6.49 10. Fertlizers 20.58 24.22 21.89 12.46 11. Irrigation 8.29 3.81 14.03 3.55 12. Labor Productivity 0.63 -1.38 0.71* 0.76 13. Land Productivity: NSA basis 2.28 0.19* 1.68 3.53 14. Land Productivity: NCA basis 1.30 0.33* 0.04* 2.64 15. Percentage of Area Under HYV Crops 62.94 0.00 89.50 14.45 16. Research Expenditures 10.61 12.73 13.36 17.67 17. Literacy 0.21* -2.48 1.55 1.62 * Not significant at the 95% level. Source: Pray and Abmed (1984). Figure 6.1: Index of total output, total input and total productivity (Usine Revised Time Series Data) (Base 1957/48 - 100) 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 0 198 * *.. . ,,-@,,,. . ... . I9W 195 5 " 1952Ar*. 1953 -e * 1955 , - _--.. f/* , .r' + 19'*I 1962 1961 4 - . .... . 1965! 1966 *~ . 1967 !* 1968 Vt -.1 - I.p 1969 !f- 1970! N- * 1971 *0 1972 I p I F 1973 sa- 1974 t vI 1975 o 0 1,70~~~~ 1979,la 1960.- 1961 * .a a- t Figure 6.2 Indexes of Inputs Used in Agricultural Production (Official Data) (Base 1947/48 = 100) s t 4* v* *e* ** @@ @* * X * * ** **.* *S* * *S@@* ** *t* *t*..**vvv@* ** * *...t ****.*..*** s @@ e, ee ,. .*. e pv. . @w 1948!*X' 1949 1950. 1951 i 1952 1953 1954; 1955 1956 a. 1957 1956 1959 1960 . 1961 ! -_ 1962 9 .- 1963 1*~ 1965 1966 . D..- _ 1967 o . 19668 - & 1969 1970 . 1971 !c 1973 1974 !. cr. 1975 * 'rt0 1976 % '-1 19797 -~. 197 01-' r ~~~ ;~ n0 1980 .. *.**O.***. * . * * * *. .S.. * 42 Banladesh Table 6.2 Anmal Rates of Grawth of Major Crops: Production, Area, Yield 1947/48- 1947/48- 1959/60- 1970/71- Crops 1980/81 1959/60 1970/71 1980/81 Rice Total Production 2.03 0.63 2.37 2.76 Area .88 0.43 1.62 0.67 Yield 1.15 0.20 0.75 2.09 Aus Rice Production 2.39 3.18 2.73 2.46 Area 1.60 1.92 3.25 0.24 Yield 0.79 1.26 -0.52 2.22 Aen Rice Production 1.04 -0.20 0.17 3.27 Area 0.13 -0.13 0.09 0.71 Yield 0.91 -0.07 0.08 2.56 Boro Rice PracucticL 7.98 1.42 16.06 1.54 Area 4.92 0.24 8.80 1.70 Yield 3.06 1.18 7.26 -0.16 Wheat roidction 10.47 2.33 11.94 24.94 Area 6.82 2.53 7.53 14.63 Yield 3.65 -0.20 4.41 10.31 Pulses Production -0.94 -2.21 2.55 -1.72 Area -1.12 -2.12 0.39 0.29 Yield 0.18 -0.19 2.16 -2.01 Oilseed Prodiction 1.56 1.27 4.96 -0.39 Area 0.21 1.19 -0.03 0.09 Yield 1.35 0.08 4.99 0.48 Production 2.66 1.46 7.38 0.53 Area 1.85 1.72 4.02 0.23 Yield 0.81 -0.26 3.36 0.30 Potato Production 6.24 0.37 11.23 2.02 Area 4.18 0.52 5.43 2.18 Yield 2.06 -0.26 5.80 -0.16 Tobacco Production 0.08 -2.48 4.25 2.57 Acea -0.01 -1.69 0.68 2.52 Yield 0.09 -0.79 3.57 0.05 Jute Production -0.06 0.14 1.40 -0.02 Area -0.90 -2.76 .24 -1.58 Yield .84 2.90 1.16 1.56 Tea Production 2.06 4.60 2.88 7.03 Area 1.58 0.54 3.57 -0.35 Yield 0.48 4.06 -0.69 7.38 Source: Pray and Ahred (1985). 43 Table 6.3 Rice and Wheat Varieties from CIMMYT and IRRI Variety Pedigree Parentage Purbachi Chen-chu-ai Pajam (not ackowledged as official) IR8 IR8-288-3 Peta/Dee-gee-woo-gen BR1 (Chandina) IR532-1-176 Peta*3/TN1/TKM-6 BR-2(Mala) IR272-4-1-2 CP-SLO*2/Sigadis BR-3(Biplab) BR27-10-1 IR506-1-133/Latisail BR-6(IR28) IR2061-214-3-8-2 IR833-6-2-1-1/IR156-149-1/IR737 BR-7(Brribalam) IRR2053-87-3-1 IR1416-131-5/IR22//C4-63 BR-8(Asha) BR167-2B-9 IR272-4-1-2/IR305-3-17-1-3 BR-9(Sufala) BR169-1-1 IR27-4-1-2-3-10/IR8 IR-5 IR547-2 Peta/Tangkai Rotan IR-20(Irrisail) IR532-E-576 Peta*3/TNI/TEM-6 BR-4(Brrisail) BR51-91-6 IR20/IR5-114-3-1 BR-5(Dulahhhog) Pure line BR-10(Progati) BR51-46-5-HR65 IR20/IR5-114-3-1 BR-i1(Mukta) BR52-67-1-HR88 IR20/IR5-47-2 In addition, the following four varieties have been recommended by BRRI for release, but the recommendaion has not yet been acted on by seed authorities. BRl2(Moyna) BRI61-2B-58 Chandina/IR425-1-1-3-8-3 BR14(Gazi) BR319-1-HR28 IR(d)/BR3 BR15(mohini) IR2071-199-3-6 IR561-228-1-2/IR1737//CR94-13 BR16(Shahibalam) IR2793-80-1 IR416-131-5/IR1364-37-3-1//IR1514A-E666 Variety Mexican Designation Mexicpak Siete Cerros (Kalyansona) Sonalika (53-388-AN x Pi"S"-IR)(B4946-A4-18 x Y53)Y503 Tanori 71 Jupataco 73 Norteno 67 Balak (selection from CIMMYT material) Doel (selection from CIMMYT material) Inia 66 Akbar Anada Barkal Kanclam Sources: Rice varieties - personal communication, Clarence Miller, IRRI agricultural economist at BRRI. Wheat: Dana Dalrymple and BARC Table 6.4 Acres of Rice ani Wheat Varieties Boro Aman Aus All Seasons Wheat Total HYV Pajam BRRI Total HYV Pajam BRI Total HYV HRRI All Rice %BRRI HYV Total 1968 156 156 156 24427 1 3 192 1969 360 360 5 5 17 382 24568 2 13 290 1970 580 0 580 29 0 29 43 652 25486 3 23 296 1971 857 18 839 200 83 117 17 1028 24494 4 33 311 1972 795 35 760 428 167 262 96 1117 22975 5 43 314 1973 1088 53 1035 763 260 513 114 1663 23796 7 53 297 1974 1454 70 1084 1032 333 699 197 2280 24410 9 72 305 1975 1629 88 1541 1064 416 648 489 2678 24197 11 81 311 1976 1587 105 1482 1182 500 683 667 2831 25525 11 218 371 1977 1338 123 1215 1278 583 695 800 2710 24419 11 288 395 1978 1587 131 1456 1283 666 617 883 2956 24778 12 389 467 1979 1650 167 1483 1923 821 1112 1002 3597 24992 14 583 642 1980 1788 224 1564 2154 1028 1126 984 3684 25106 15 1015 1071 1981 1845 216 1629 2376 1183 1193 1200 4022 25474 16 1412 1461 Source: Bangiadesh Bureau of Statistics, Agricultural Yearbook of Bangladesh, DIhaa. Various years. Ahmed and Pray 1980. 45 An appropriate costing of these related activities would somewhat inflate the rather low cost of research which has been attributed in attempts at evaluation of the net benefit of agricultural research (e.g., Gill 1982). The earliest estimate of returns to research in Bangladesh (Pray 1979) found rates of return above 30 percent. The results of that work have been updated (by C. Pray) to include more recent data on the spread of HYVs and on the shift in the supply curve due to research. The more recent data on the shift of the supply curve come primarily from a study by IFDC and BARC (Sidphur, Boanante and Ahsan, 1984). In addition, more accurate research and extension expenditure data was refined and used. The measure used in the process of computing rates of return is the total economic surplus from new technology, computed according to the framework and simplified parametic specifications by Hayami and Akino (1977). For brevity, these and other detailed assumptions reported by Pray (1979) are not repeated here. If it is assumed that research affects output with a 12- year lag, and that extension does not contribute to the spread of technology, the rates of return are between 31 and 37 percent. The fact that the important rice variety Pajam spread rapidly, despite the recommendations of extension that it should not be used, indicates that extension may not have had a great deal of influence in the adoption of at least some innovations. Alternatively, if it is assumed that both research in the previous 12 years, and extension expenditure in the previous year are influential in affecting output, the rate of return to research and extension expenditure falls to something between 23 and 32 percent. About 60 percent of the observed variance in a time series of agricultural productivity since 1960 can be explained by simple regressions on HYV area (Pray and Ahmed 1985). Research expenditure itself can similarly explain over 70 percent of such variation. These associations are also supported by analyses of cross-sectional district-level data for 1977 and 1981 (Pray and Ahmed 1985). In short, from such statistical association as well as from casual empiricism, there seems little doubt that agricultural research, 46 especially through the success of modern cereal varieties, has indeed been a signficant engine of agricultural productivity and economic growth in Bangladesh. The International Fertilizer Development Center (IFDC) and BARC conducted a study of about 2000 farmers in most districts of Bangladesh. In their study they estimated production functions for each crop season by variety. Their results (Table 6.5) show that, even at low levels of fertili- zer use, HYVs give yields superior to the local varieties. 6.2 Income distribution effects of HYV technology There has been considerable concern in Bangladesh about the distribu- tional aspects of HYV adoption. In this section an attempt is made to piece together the available evidence on the contribution of HYVs to changes in the distribution of income in Bangladesh. 6.2.1 Long-run trends in rural income and asset distribution By most indications, the per capita real income in rural areas has registered a decline over the past two decades, and the vast majority of the rural poor have been caught in the grip of increasing impoverishment. -'....this process of impoverishment may, in part, be a reflection of growing inequality in rural Bangladesh" (Islam 1981). In the late 1960s, Bose (1968) concluded that per capita agricultural income had declined over the entire period of the 50s and 609 except for 1963/64. More recently, Khan (1976), using energy requirements to define poverty, has shown that there was an increase in the incidence of poverty over the 15 years from 1960. Especially alarming was his finding that the "extremely poor" (those whose energy intake is only 80 percent or less of "requirements"), as a proportion of the rural population, increased eightfold over the period. 47 Table 6.5 Size of Holdings (acres) < 0.5 .5-1 1-1.5 1.5-2.5 < 2.5 2.5-7.5 7.5 > Tocal Area Under HYVs (lOOOs) 16 77 147 364 604 1079 531 2214 Percent of total gross croppd area under HYVs 10.2 10.5 9.8 9.1 9.4 6.8 5.7 7.0 Percent of total gross cropped ares fertilized 34.2 35.8 34.6 31.9 33 25.8 22.9 26.4 Percent of total gross cropped area irrigated 9.9 10.2 9.8 8.9 9.3 7.2 7.3 7.6 Source: BBS (1980) 48 However, other empirical studies do not support the conclusion that inequality has increased. Alamgir (1974) found that, while the percentage and absolute sizes of the rural population in poverty increased between 1966/67 and 1973/74, the Gini coefficient showed little change. The Agricultural Censuses of 1977 and 1960 indicate that the major change has been a concentration of holdings in the medium size farm (2.5 to 7.5 acre) category (Table 6.6). In addition, the average size of the small holdings increased and the proportion of total area held by small farmers increased. At the same time, the proportion of land owned by "large"-scale farmers decreased from 38 to 32 percent. Thus, the Gini coefficient for land holdings indicated lessening inequality between 1960 and 1977. The absolute number of landless is clearly increasing (Khan 1977, Singh 1979). There is also evidence that the percentage of landless households is increasing and that poorer farmers are selling their land (Alamagir 1976, Islam, 1981). Thus, although the distribution of land among landholders is less unequal, for the rural population as a whole, the distribution of the major asset in Bangladesh agriculture - land - may have worsened. Several researchers (Bose 1968, Clay 1976, Khan 1977) have found that real wages in agriculture declined between the early 1960s and 1970s, with especially sharp declines for the post-liberation period. There is, however, some evidence that the decline may have stopped and perhaps even turned around slightly (IBRD 1984). Smaller farmers and landless laborers have a high dependence on wage incomes (Alamgir 1976, Islam 1981). Therefore the decline in real wages led most scholars to conclude that the living conditions of the landless and near-landless (about half the rural households) must have worsened in the face of small increases in employment opportunities. Further, some scholars suggest that the larger-scale farmers enjoyed a substantial improvement in their living conditions by selling surplus grain at prices that were increasing faster than the general cost of living and through greater access to subsidized credit and purchased inputs. They hypothesize that, since the HYV technology is of the land- augmenting type and land is very 49 Table 6.6 Distribution of Land 1960 and 1977 0-2.5 2.5 - 7.5 7.5+ Total Percent of farm holdings 1980 51.6 37.5 10.7 100 1977 49.7 40.8 9.4 100 Percent of farm area 1960 16.2 45.7 38.1 100 1977 18.7 48.9 32.4 100 Average farm size (acres) 1960 1.1 4.3 12.6 3.5 1977 1.3 4.2 12.1 3.5 Source: BBS (1980). 50 unequally distributed, the benefits from increased output and efficiency must acrue mainly to those who control the land. 6.2.2 Large vs. small farmers vs. tenants Several nationwide studies have been made of the use of new technology by farm size and tenancy status. The Agricultural Census showed (Table 6.6) that the proportion of cropped area under HYV paddy is inversely related to farm size. Thus, the farmer with less than an acre of land on average placed 10 percent under HYVs, while the large-scale farmers placed less than 6 percent of land under HYV paddy. The Agricultural Census data on fertilizer use show the same pattern - small-scale farmers fertilized one-third of their land compared to one-quarter by the medium and large-scale farmers. In 1979/80 the IFDC and BARC study (Sidhu, Baanante and Ahran 1982) indicated that the smallest-scale farmers on average used more fertilizer per acre than others (Table 6.7). More fertilizer per acre was used on sharecropped and rented land than on owner-operated land. For the nation as a whole, Januzzi and Peach (1980) found that owners, owners-cum-tenants and tenants all exhibited very similar rates of adoption of modern inputs - chemical fertilizers, pesticides, power irrigation, mechanical irrigation and other irrigation. Thus, there appears to be no overt disadvantage experienced by small holders or sharecroppers. The majority of the scattered micro-studies indicate that small-scale farmers as well as the large-scale have been responsive to the new technology (Muqtada 1975, Bose 1974, Faidley and Esmay 1976, Asadussaman 1972). Asaduzzaman (1979) and Rahman (1981) found that, while the proportion of adopters is lower among the small-scale farmers, once they adopt, they devote proportionately larger areas to HYVs. The IFDC/BARC study (Sidbu, Bannati and Ahaan 1982, 1984) is the only one available which estimates the impact of a "nev" technology - in this case fertilizer - on the incomes of different farm size groups. The results (Figure 6.3) indicate that, with the exception of the aus season, small-scale farmers enjoyed more net benefits per unit area than others. It was also 51 Table 6.7 Average Levels of Use of Fertilizer(s) on Owner-Operated, Sharecropped, and Rented Land in Munds Per Acre of Cropped Land Per Farm by Farm Size and Season, 1979/80, Bangladesh On mer-perated Land On Sharecwped Land On Cash Rented-In Land Boro ds Amsn Boro Aus Aman Boro Aus Aman Farm Size (FS) 1979/80 1980 1980 1979/80 1980 1980 1979/80 1980 1908 0 < FS < 1.0 1.23 0.83 0.79 1.41 0.68 0.77 2.67 1.78 1.26 1.0 < FS < 2.5 1.20 0.72 0.75 1.09 0.51 0.69 2.20 1.14 0.95 2.5 < FS < 5.0 0.93 0.68 0.85 1.04 0.57 0.68 2.55 1.87 0.96 FS > 5.0 1.12 0.78 0.79 0.95 0.77 0.74 0.75 0.97 0.93 All sw,1e 1.13 0.74 0.80 1.23 0.61 0.74 2.66 1.60 0.95 Source: Sidhu, Baanante and Ahsan (1982) Figure 6.3 Farm-Level Average Benefits of Fertilizer Use by Farm Figure 6.4 The Sample Distribution of Size by Season, 1979/80, Bangladesh. Benefits of Fertilizer Use Among Farmers by Farm Size, 1979/80, Bangladesh. O FERTILIZER NET BENEFITS 390 2.! FERTILIZER COST E3 FSSIO ACRE 380 2.0 f I.0'F3S2.5 ACRE 360 FARM SIZE I FS)' rs'TA2A .RE 340 I 2 FS '1.0 oc. [jJ2.5 F3S0.0 ACE 2 a I.OFS52.5 ac. Q FS-50 ACRE 320 - 20 3 2.5FS5.0 oc. 100 300 fl sFS54.5 oc. 2eO 2 20 BENEFIT/COST RATIO B/C 0 RS ~ ~ ~ ~~~I 1tZ1 I 234 1234A8N93Bs tB A690Au19 260ONS - UA 20 20O17-0 AU 98 MM18 33ot 240 -I E 70- P20- 1.9~~~~~~~~~~~~~~~~3. 160 .Ii1' .9 i4026. 140 - .e 8~*: 120 -j~ .(7j :9 too 1.4IJ~.~;.: ... 2~. 2. 1.2~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.75 40 * .*. *** *. 'ARM SIZE: 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 ~~AMAN WS7 WMF 19?9-00 kS 1980 AN"1 19W0 SEASONS: AM%AN 1979 BORO 1979 80 AUS 1980 AMAN 1980 ource: Sidhu, Baananti and Ahram (1982). 53 shown (Figure 6.4) that, by combining net benefits per unit area and farm size, more than half of the total net benefits from fertilizer use went to farmers with less than 2.5 acres in the 1979 aman and 1970-80 Aman crops, and about 40 percent in the 1980 aus and aman crops. Thus, fertilizer use did not worsen income distribution and, in the season when the most HYVs are used (boro), it actually improved the income distribution pattern among land owners. 6.2.3 Landless laborers One of the major advantages of new HYV technology for a country with a growing labor force is its greater requirement of labor. It is hypothesized that, with the application of land-augmenting or yield-raising technology, labor requirements per unit area and the demand for labor will increase. Further, a rise in incomes as a result of the new technology will lead to some substitution of family labor by hired labor. Until recently, the data on real agricultural wages indicated a decline from the level reached round 1964. Since this was the same time at which HYVa, the use of fertilizer and irrigation started to spread, some investigators concluded that the landless have failed to benefit from the increased productivity of the HYVs and that inequality in the countryside has increased (Islam 1981). However, this argument ignores the very large increase in the supply of labor due mainly to population growth. This large shift in supply during a time that demand for labor was growing (albeit slowly) due to modern technology, on balance pushed wages down. In the absence of HYVs, real wages would have fallen even farther than they did. Khan (1984) showed that real agricultural wages between 1949 and 1980 were positively related to agricultural productivity and the relative price index for agriculture, and negatively associated with the trend variable, which he associated with the rapid increase in supply of labor, the closed land frontier and slow growth in demand for labor. The most recent data on real agricultural wages seem to indicate that wages reached their lowest level in the mid-1970s. Table 6.8 shows nominal 54 and real wage indices as calculated by the World Bank, using official government wage data. The average index for the years 1973/74 - 1977/78 was 70, and during the next five years 80, - a 15 percent increase. While seemingly no one has analyzed this phenomenon thoroughly, it is likely that modern agricultural technology was responsible for part of the increase. The empirical evidence of increased demand for labor as a result of switching from the traditional varieties to the HYVs seems conclusive for Bangladesh. Virtually all micro studies found considerably higher labor requirements with the new technology: Muqtada (1975), Ahmad (1980), Ahmed (1976, 1977), Clay and Khan (1977), Hossain (1980), Faidley and Esmay (1976)), Sidhu, Baanante and Ahsan (1982) and Clay (1978). Ahmed (1983) found that the shift to HYVs not only increased hired laborers' real income but also increased their share of income, except where there was mechanized cultivation. It would seem that, with increased labor demand, the "trickle- down" effect of the HYV technology would be vindicated. However, there are several caveats. First, Clay (1977), and Khan and Clay (1978) have warned that the previous reviews of agricultural employment may have been too optimistic in that the evidence does not support employment elasticities (the ratio of the percentage change in employment for a percentage change in productivity per unit area) of more than 0.5 for the switch to lYVs. More importantly, some changes in cropping patterns resulting from the introduction of HYVs may have actually decreased aggregate employment. There are some cropping patterns where the introduction of the new technology led to the substitution of one new HYV crop in place of two traditional crops, leading to a lower aggregate level of employment. Second, the seasonal nature of employment in Bangladeshi agriculture may make some of the employment-generating effects of the new technology less beneficial for hired labor than the aggregate increase in demand for labor would seem to imply. It bas been found (Hossain 1981) that, when the demand 55 for labor increases, only 50 percent of the increase will be provided by hired labor. Clay (1978) noted "that most of the new rotations involve either a switch of activity into the winter months or substantial additional demand for labor during the winter months; but again, careful inspection shows that, in most cases, there is no marked increase in labor requirements during the slackest period (September-October). Intensification of crop rotation and the expansion of irrigated cultivation in Bangladesh in most cases significantly increases labor requirements, but does not remove the marked seasonality from crop production activity." In contrast to the situation in India and Pakistan, where agricultural research and new technology have been associated with labor-saving mechanization, Bangladesh has largely avoided this problem. The one type of mechanization which has been adapted widely is mechanized irrigation. This has increased the demand for labor, and it comes in all sizes from various types of hand and foot operated tubewells, to small mechanically powered tubewells which can irrigate about 15 acres, to large tubewells which can irrigate up to 200 acres. It is not clear that the government has always encouraged the use of the type of tubewell with the most appropriate capital to labor ratio, but at least it has not encouraged mechanical cultivation which many studies have found decreases the demand for labor without increasing output per unit area. At the agricultural census of 1977, there were 3454 tractors and 738 power tillers in use compared to 58,844 power pumps for irrigation (BBS 1981). One type of mechanization that the government has encouraged is labor-displacing-rice milling. This is particularly devastating to poor women who do almost all of the hand milling. However, there is no obvious connection per se between HYVs and power mills (World Bank 1983). 6.2.4 Consumers vs. producers In a closed economy with no government interference in grain markets, increased production of the main food grain due to new technology will push down real prices and pass a substantial portion of the benefits from research 56 to consumers. Bangladesh is not a closed economy and the government procures and sells grain each year. Thus, the actual impact of new technology on foodgrain prices and the distribution of income between producers and consumers will be determined in part by government decisions on (a) how much grain to procure internally, (b) how much to import, and (c) at the prices for both procurement and sale. Throughout most of Bangladesh's history, the governments' primary concerns have been with keeping rice prices relatively low and stable and ensuring that certain politically important groups have assured supplies of basic commodities at reasonably low prices. The ration system provides inexpensive grain to the important groups. When threatened with a general price rise, the government reacts by importing grain by whatever means are available. However, until recently it has not shown much interest in using price supports and procurement as a means of providing farmers with production incentives. The downward shift in the supply curve for foodgrains due to the introduction of new varieties has not been offset by increased internal procurement. Continued imports have kept prices down. Procurement has been about 3 percent of total supply since 1975/76, with the exception of 1980/81 when it increased to 7.7 percent, and imports decreased to about one million tons (World Bank 1982). Much of the benefit from the new technology has gone to consumers. The consumers who perhaps benefit most from new technology are those who do not have access to ration shops and who buy wheat and relatively low-quality rice. Most high yielding rice varieties grown in Bangladesh are considered low quality by Bangladeshis. Wheat is generally considered to be a grain inferior to any type of rice. The gains from changing technology are likely to go to the rural consumer who has to purchase grain in the market. This is reinforced by a procurement system which does not reach far into the countryside. Another important impact of the new technology on consumers is that it has reduced the seasonal shortages of foodgrains and thus decreased some 57 seasonal price rises. The major impact of new technology has been the improvement in production in the winter rice crops, and the wheat crop which is also grown in the winter. Thus, more grain is coming to market in late spring and summer which was the traditional time of shortage before the aus harvest. This means lower prices for consumers and, as noted, the consumers who benefit most will be the rural consumers who do not have access to the ration system. 6.2.5 Women Some concern has also been expressed about the possible effects of the new technology on a relatively under-privileged class of Bangladesh - namely, its women. Unfortunately, there are almost no data on the differential impact of HYVs on men and women. Traditionally, women have performed the bulk of the processing task for cereals. The fear is that their unequal status and seclusion will increase if labor-substituting technology, e.g., rice-milling, is widely available. Rice milling is spreading but research per se has done nothing to speed or to slow the spread. The rise in foodgrain production due to modern varieties should increase the scale of post-harvest operations (BARC 1975) and the demand for women's labor. However, this is a major gap in knowledge, and more research is needed on those and related issues. Table 6.8 AVEAGCE WAGC RATE INDICES B? SECTORS (SUMMARY). 1963/64 - 1982/83 (1969170 - 100) COL Index Rural Tonilal Indices of Industrial COL Real Indice /a Year Ceoeral Manufacturing Construction Agriculture Fisheries Workere Index General Manufacturing Construction Agriculture fisherie. 1963164 83.54 78.68 95.00 84.39 80.94 76.74 71.80 108.86 102.53 123.79 117.54 112.73 1964/65 87.87 82.42 104.01 86.67 83.83 81.03 .. 108.44 101.72 128.36 1965/66 88.12 83.52 100.42 63.95 89.20 82.45 .. 106.88 101.30 121.80 1966/67 94.32 88.83 100.38 98.49 95.08 90.88 92.31 103.79 97.74 110.45 106.70 103.01 1967/68 94.61 88.62 99.80 96.53 97.47 92.84 .. 101.91 95.45 107.50 1968/69 97.99 92.45 104.76 97.34 102.95 96.17 .. 101.89 96.13 108.93 1969/70 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 1970/71 105.30 107.28 102.05 105.10 104.82 103.98 105.48 101.27 103.17 98.14 99.64 99.38 n 1971/72 106.50 109.56 108.37 99.72 105.30 108.12 124.32 98.50 101.33 100.23 80.22 84.71 °° 1972/73 130.74 134.39 132.26 128.78 123.89 198.46 184.46 65.88 70.21 69.06 69.82 67.17 1973/74 173.07 161.23 201.10 185.61 156.18 268.03 259.91 64.57 60.95 76.18 71.42 60.10 1974/75 221.31 192.37 263.68 261.40 196.74 447.60 438.37 49.44 43.67 60.08 6t.03 44.88 1975/76 243.97 204.90 305.31 279.17 225.56 364.60 381.60 66.91 56.71 84.78 73.16 59.11 1976/77 252.07 222.39 310.75 267.74 237.04 354.03 374.13 71.20 63.02 88.50 71.57 63.38. 1977/78 279.81 251.30 333.89 311.63 250.95 419.27 434.38 66.74 59.94 76.64 71.75 57.78 1978/79 346.25 318.01 413.84 370.63 311.01 457.55 472.69 75.74 69.56 90.46 78.37 65.60 1979/80 432.90 389.04 491.69 443.31 451.43 525.68. 548.71 82.45 74.10 93.63 80.80 82.28 1980/81 492.09 452.27 544.60 482.01 529.28 568.10 617.70 86.62 79.61 95.86 78.04 85.69 1981/12 566.16 515.25 616.30 566.88 617.15 656.15 696.61 86.35 78.56 93.98 81.38 88.59 1982/83 598.11 558.02 677.20 558.35 638.93 683.56 .. 87.50 81.63 99.07 81.68 93.47 not awallable. /a The general. manufacturing and construction wage indices are deflated by the COL index for industrial vorkers. while the agriculture and fieheries wage lndices are deflated by the rurel COL index. Source: World Bank (1984) 59 7 Future Impact of Collaborative Research The most important yield increases from rice breeding will be in t.aman and aus. The new varieties BR-10 and BR-l1, which were recently released for the t.aman season, have several characteristics that are popular with farmers and thus they appear to be spreading rather rapidly. Their main characteri- stic is that they can be transplanted several weeks later than BR-4 but still get produce high yields. This allows farmers to transplant aman rice following late floods, or jute and late aus rice. The main advantage of the local improved varieties has been their ability to outyield the BRRI varieties when planted later in the season. Now the local varieties will be replaced with BR-10 and BR-li in several places. High yielding short-duration varieties for the aus and boro season have just been released and BRRI hopes to release its first improved variety for b.aman in 1985. (IRRI 1984). The government is now making fertilizer with zinc and sulphur available. Gill (1983) reports that, in experiments in 18 districts during the 1981 aman season, sulphur in fertilizer increased yields of rice (BR-4) by over 400 kg/ha and zinc increased yields by nearly 400 kg/ha. He continues: "Informed but unofficial estimates suggest that around 3 million acres are potentially zinc-deficient and 2 million acres potentially sulphur deficient". This suggests that the potential increase in output in the aman season would be 800,000 tons. More recent experiments indicate that sulphur deficiency is more widespread and zinc deficiency less widespread than previously supposed. As of January 1984, both zinc and sulphur were available from BADC and their appropriate use was spreading slowly (IRRI 1984). The growth of wheat production is slowing down but there is room for more production in less favored regions. Also more genetic diversity is being built up. The recently released varieties Balak and Doel have much more resistance to rust than Sonalika which currently is grown on about 70 percent of the wheat area. 60 There will be further reductions in imports of Dutch seed potatoes and increases in yield due to improved seed potato production. Disease-free seed is now being produced in Degibani in northern Bangladesh. This will allow the production of lower cost seed potatoes of the Dutch varieties. These seeds are free of diseases (mainly viral) and hence give higher yields. High- yielding varieties of local and exotic stock are in the pipeline which will increase yields of high quality and thus high-priced potatoes. The cost of storage is being substantially reduced through refinement of techniques in collaboration with CIP research staff. 61 8 Conclusion The collaboration between the international agricultural research centers and the national agricultural research system of Bangladesh has been very important to the growth of agricultural research and agricultural productivity in Bangladesh. The collaboration started in 1966 with the beginning of IRRI's relationship with the Agricultural Research Institute. The number of collaboratively researched crops has expanded gradually to include wheat, maize, potatoes, some pulses and cassava. The most important types of contact included posting center researchers in Bangladesh to work with local researchers on local problems, regular visits by center staff, training programs at the centers or run by the centers, joint programs of germplasm testing, and publications. These activities have influenced the size and direction of Bangladesh research. The well-publicized success of IRRI rice and CIMMYT wheat elsewhere in Asia made it easier for the national program to acquire budget allocations and resources from the East Pakistan and Bangladeshi governments and from the donors. The increases in funding have been most rapid in the foodgrains, potatoes, pulses and oilseeds, while the research on export crops has grown more slowly. The proportion spent on export crops has thus declined. Within the commodity programs, joint planning of research in on-farm research programs has influenced priorities toward farmers' needs. The institutional structure of research has been influenced by the centers' emphasis on multidisciplinary teams organized around a single commodity. The general efficiency of the research programs that are in touch with the centers seems to have improved - training at the centers has encouraged researchers to be more practical in outlook and work style. There is more incentive to perform vigorously with the prospect of wider recognition through participation in international yield trials or in international meetings. Center outposted (special project) staff have resources that must be helpful in overcoming crucial research bottlenecks. 62 The outcome of this collaboration has been more food production and probably improved income distribution. To quote Dr Kazi M. Badruddoza (1984, pp. 74-5): "The success of the rice research and production programs in Bangladesh was, among other things, due to a large number of trained scientists from IRRI. In fact, the establishment of BRRI in 1968 was considerably influenced by the collaboration of the local rice program with IRRI. Over the first decade of its existence, BRRI released 15 new varieties and the area under modern varieties grew from 271,000 to 1.78 million ha, a growth rate of 25 percent per annum. Varieties released by BRRI now cover an estimated 15 percent of the total rice production area and produce 31 percent of the total output." "...Credit for success of wheat cultivation goes to a few dedicated scientists of BARI, who were trained at CIMMYT and also to the persistent interests shown by the late Dr. Glenn Anderson of CIMMYT and follow-up visits and activities of CIMMYT staff... In the six years since 1975 and 1976, when the thrust on field- scale introduction of wheat started in a traditionally rice growing country, wheat production increased several fold. Bangladesh now produces about 1.4 million tons of wheat annually compared to 40 thousand tons in the early 1970s. This is a success story which is rare anywhere in crop production. It would not have been possible without the dedicated team of scientists trained at CIMMYT and continued interest by CIMMYT staff." Modern varieties of rice and wheat have increased the demand for labor which benefits the landless laborers who are the poorest group in Bangladesh. Increased food production has reduced the real price of foodgrains, which primarily benefits poorer consumers both in the cities and the countryside. Thus the poor seem to have benefited significantly from the modern varieties. While still poor, they would be even worse off in their absence. At present there is insufficient evidence to say whether or not their relative position has improved due to these varieties. Many aspects of institutional arrangements for agriculture research in Bangladesh reflect the influences of several international centers, ranging from the structure of the rice research institute to aspects of the ongoing debates on input and food pricing policies in the country. The centers are widely appreciated for their contributions. 63 References Ahmad, K.U. (1980). "Excotic Pototo Varieties for Bangladesh" mimeo, BARI Ahmad, S. (1980). Output and Factor Market Effects of Technological ProQress. Population Growth and Public Investment in Bangladesh Agriculture. Unpublished Ph.D Dissertation, Yale University, New Haven, 1980. Ahmed, I. (1976). "Technical Change and Labour Utilisation in Rice Cultivation: Bangladesh", Bangladesh Development Stuides, 5(3), July 1976. Ahmed, R. (1979). Foodgrain SuPPYl, Distribution and Consumption Policies Within a Dual Pricing Mechanism: A Case Study of Bangladesh. IFPRI, Washington, D.C. Almagir, M. (1974). "Some Analysis of Distribution of Income, Consumption, Savings and Poverty in Bangladesh", Banlgadesh DeveloRment Studies, 2(4). Almagir, M.(1976). Bangladesh: A Case of Below Poverty Level Equilibrium Trap. Bangladesh Institute of Development Studies, Dacca. Alauddin, M. (1982). "Inputs and Returns to Agricultural Research in Bangladesh", Journal of Management. Business and Economics, 8(2). Alim A.B., Zaman and Johnson, LPV (1966). "Progress Report on: Accelerated Rice Research Program of East Pakistan January-August 1966" mimeo, Dacca, September 24, 1966 Asaduzzaman, M. (1979). "Adoption of HYV Rice in Bangladesh", Bangladesh Development Studies, 7(3), Monsoon, 1979. Bangladesh Academy of Rural Development (BARD) (1975). Socioeconomic Implications of Introducing HYV in Bangladesh: Proceedings of Seminar. Comilla, November 1975. Bangladesh Agricultural Research Institute (BARI) (1980). Second Five Year Research Programme. BARI: Joydebpur, 1980. Bangladesh Bureau of Statistics. Agricultural Census of Bangladesh 1977, National Volume, June 1981. Bangladesh Bureau of Statistics. 1982 Agricultural Yearbook of Bangladesh, Dhaka, 1982. Bangladesh Potato Improvement Program. "Progress Report 1982-83" Dhaka: BARI, 1983 64 Bose, S. R. (1968). "Movement of Agricultural Wage Rates in Bangladesh", in A. Mitra (ed) Economic Theory and Planning, Dhaka. Bose, S. R. (1968). "Trends of Real Income of the Rural Poor in East Pakistan 1949-66", Pakistan Develogment Review, 8(3), Autumn 1968. Chandhuri, S.D. (1949). "Jute Research Laboratories of Eastern Pakistan," Agriculture Pakistan, 1(1), September 1949. Chawdhry A.R. (1961). "Agricultural Research Achievements in East Pakistan 1906-1961" Dacca; East Pakistan Press. Clay E. (1976). Institutional Change and Agricultural Wages in Bangladesh" Bangladesh DeveloDment Studies 4(4) October 1976. Clay,E. J. (1978). et. al. "Yield Assessments of Broadcast Aman (Deep-Water Rice) in selected Areas of Bangladesh in 1977" BRRI. Clay, E. J.(1978) "Employment Effects of the HYV Strategies in Bangladesh: A Rejoinder" BARD: DACCA. Faidley, L. AND ESMAY, M. L. (1976). "Introduction and Use of Improved Rice Varieties: Who Benefits?, in Stevens Robert D., et.al. (ed). Rural Development In Bangladesh and Pakistan. University of Hawaii Press: Honolulu, 1976. Gill, G. J. (1981). Operational Funding Constraints in Agricultural Research In Bangladesh. BARC, Dacca, April 1981. Gill, G. J. (1982). "Agricultural Research in Bangladesh: Costs and Returns", BARD/IADS, Dacca, 1983. Hossain, M. (1980). "Foodgrain Production in Bangladesh: Performance, Potential and Constraints", Bangladesh Development Studies, 8(1&2). IRRI (1984) "BRRI-IRRI Project Developments; IRRI Team Annual Report for 1983" mimeo Dhaka: BRRI: 1984. Islam, R. (1981). "Trends in Rural Income Distribution in Banlgadesh", in Mahmood, Wahiduddin (ed). Development Issues in an Agrarian Economy-Bangladesh. Centre for Administrative Studies: Dacca University, Dacca, 1981. Januzzi, T.F. and PEACH, J.T. (1980). The Agrarian Structure of Bangladesh: An Impediment to DeveloRment. Westview, Boulder, Colorado, 1980. Khan, A. R. (1977). "Poverty and Inequality in Rural Bangladesh", in International Labor Organization. Poverty and Landlessness in Rural Asia Geneva, 1977. 65 Mc Clung, A. C., (1965). "Accelerated Rice Research Program for East Pakistan in Cooperation with the International Rice Research Institute" May 1965, mimeo Ford Foundation, Dacca. Mosemann, A.H. (1980). "Bangladesh Agricultural Research System: Report of the Review Team" BARC: Dacca. Muqtada, N. (1975). "The Seed-Fertilizer Technology and Surplus Labour in Bangladesh Agriculture". Bangladesh Development Studies, 3(4), October 1975. Pray, Carl, (1979). "The Economics of Agricultural Research in Bangladesh" Bangladesh Journal of Agricultural Economics Vol. II No. 2 December. Rahman, Atiqur (1981). "Adoption of New Technology in Bangladesh Agriculture: Testing gome lypotheees", in Mahmood, Wahiduddiu (ed), Development Issues in an Agrarian Economy - Bangladesh. Centre for Administrative Studies: Dacca Univeristy, Dacca, 1981. Sidhu, Surjit, S., Baanate, C. and Ahsan, E., "Agricultural Production, Fertilizer Use, and Equity Considerations: Results and Analysis of Farm Survey Data. 1979/80" Muscle Shoals, Alabama: IFDC 1984. Sidhu, Surjit, S., Baanate, C. and Ahsan, E., "Agricultural Production, Fertilizer Use, and Equity Considerations: Results and Analysis of Farm Survey Datas 1980/82" Muscle Shoals, Alabama: IFDC 1984. Sikka, L.C. "Bangladesh Potato Improvement Program: Progress Report 1981-82", BARI, 1982. Singh, I. (1979). Small Farmers and the Landless in South Asia. World Bank: Staff Working Paper No. 320. Washington, D.C. February 1979. World Bank 1983 "Bangladesh: Selected Issues in Rural Employment" Washington, World Bank World Bank 1982 "Bangladesh: Foodgrain Self-Sufficiency and Crop Diversification." Washington, World Bank. Zaman, S.M.E. (1975). "History of Agricultural Research in Bangladesh", in Integrated Rural Development, Proceedings of the Seminar, November 29 - December 3, 1975. Zaman, H (1968). "Progress of Wheat Research in East Pakistan During the Period 1958 to 1968" Agriculture Pakistan XIX. ISSN 0257-3148 ISBN 0-8213-0642-1