The members of the Evaluation Team are very grateful for the time so
willingly given them by so many busy people during the course of the
evaluation. Everywhere we found frankness, honest self-appraisal, and
strong desire for success in-the Project-. Some of these many people, but
unfortunately not all, are listed in Annex C. Our apologies are extended
to those who have not been included. We also wish to record the heroic
efforts of Mr K. Mohammad Adil Asif on the word processor and Mr M. H.
Bhuiyan on the typewriter.
Jock R. Anderson (Agricultural Economics) University of New England,
Kim W-. Bridges (Systems Ecology) University of Hawaii at Manoa,
Oleen Hess (Agriculturalist) USAID(retired)
Carl E. Pray (Agricultural Economist) University of Minnesota.
Dhaka May 31, 1983
SFrom:, Research project II
kgric e 1 E al Evaluation
Re ortf the ay 1983 External
TReport Bo gl "0" sh\S
evidence that this is changing. A new Project Supervisor has recently
arrived and most of the IADS Specialists are now on duty, so that the
necessary manpower is available. IADS seems now to be in a position to
function more effectively and expresses its enthusiasm to work with
BARC to improve BBangladeshi agricultural research.
TECHNICAL SUPPORT SERVICES
BARC's International Program Service Unit (IPSU) provides administrative
support services to international-donor research projects. Its mandate
is to develop a strong local capacity to efficiently handle logistic,
accounting, fiscal management, office support and commodity procurement
functions. In some of these tasks IPSU has performed well. But in
others despite its good intentions, it has been unable to give the
level of support required by the Project. Particular difficulty has been
experienced because of inadequate attention to housing logistics,
financial disbursements and budget planning.
The causes for these problems are many, and responsibilities must be
shared by BARC and IADS. It is now time for a joint effort to see-how
the existing arrangements may be modified to better serve the needs of
the Project. What must be balanced are BARC's desires to have a strong
in-house support capability and the realistic need of an important
research Project, while recognizing the inherent constraints under which
langladesh government institutions must work.
Che central feature of the recent developments in the Bangladesh
agricultural research system, and the key to the ARP II itself, is the
strong orientation to on-farm relevance. This approach is new to
Bangladesh. It endeavours to.develop a more productive food and fiber
.growing system explicitly within the-constraints and environments faced
by the farmer. This is an inter-disciplinary thrust, under the lofty
banner of "farming systems research" (FSR).
"*SR is supposed to take a comprehensive view of a farming system and
This is fairly well reflected in the baseline surveys conducted in the
':rial'areas. At this stage, however, the mass of information collected
on farmers' circumstances has yet to be analyzed in a very fruitful way.
Presently, the emphasis is confined mostly to integrating various
::equences of crops under several improved practices and comparisons of
i.he profitability of alternative cropping patterns. The work would thus
he more accurately described as "cropping systems research" (CSR), to
indicate e the minimal attention to livestock activities and non-cropping
labor allocations, fisheries, forestry and horticulture.
The redirection of biological and social scientists' research efforts to
work in farmers' fields is a significant achievement.:The unified
approachh achieved through the National Coordinated Cropping Systems
Research Program is a major accomplishment of BARC and the Project.
Many research questions remain to be resolved, as is to be expected from
the inherent complexities, dynamics and uncertainties in the Bangladesh
farming systems, -and from the still evolving science and art of FSR. As
the CSR matures to FSR, it should yield a rich harvest of applicable
commendations to farmers, and of useful hypotheses fed back to
dentists in their continued efforts to bring good science to bear in
nerating productive innovations.
ECONOMICS AND SOCIAL SCIENCE
I hancement of research on economic and social science aspects of the
; iriculture of Bangladesh could take many forms. Problems span all
] :vels, ranging from farm management and production economics, through
m irketing and price formation, to analysis of agricultural policy. BARC
, id the Project have elected to concentrate on only the first level. The
i intentions of this part of ARP II were to focus on socio-economic
s dialysis in the farming systems research, and on research resource
allocation more generally, including attention to the distribution of
b nefits. Only the first of these has so far been addressed and, as t
0 -scribed in the FSR review in section 2.3, the economic and social
c imponent of the CSR is very underdeveloped.
T iere is great scope for BARC to improve the interchange of information
a id to foster coordinated rural research activities in social science.
T is would involve bringing together, perhaps through contract research,
.n riculturally-oriented economists and other social scientists most of
Soium are working outside the BARC group of institutions.
h ngladesh's capacity to do effective rice research has been well
c tablished through the successful operation of BRRI. The emphasis in
t e BARC crops research program is to develop such a research system for
o her food crops, and then promote their integration into rice-dominated
pi oduction patterns. The plan is to integrate all of the relevant
c ,ncerns, which range from breeding and fertilizer management, to
s ,cio-economic considerations, including the productive role of women
t'irough improvements in homestead gardening. Much of the integrative
w rk is to be accomplished through the farming systems research program.
I is discussed separately above.
Tic major successes to date have been in wheat, the summer pulses,
nimstard and potatoes, although much remains to be done. Further
a. hievements are close in some crops, including maize, kheshari and
c ibbages. The traditional disciplinary barriers to cooperative research
h .ve been broken down to some extent, but there is still much
c ,mpartmentalization.of interests, even in the cropping systems trials.
Elforts to encourage more inter-disciplinary problem-oriented work must
2.3 FARMING SYSTEMS RESEARCH
Ln planning rural research that is relevant to farmers, attention must
be given to non-agronomic issues, input constraints, market analysis and
household and area labor, and farm power availability by season, along
with the analysis of constraints to production and implementation,
verification and dissemination of research findings.
Conducting research on the small farmers' fields offers great promise
for developing appropriate location-specific production and management
technologies, so long as pre on-farm research has tested crops and
practices for locality adaptability and minimizing risks for the farmer.
Realizing the potential requires that a program be designed and
implemented for the complete cycle of research.
The recommendations for a given crop, practice, cropping pattern or
management technique must be considered in conjunction with other crops
or practices that precede or succeed them in rotation, or in the system
as a whole. Therefore, it is imperative that the research be based on
the relevant set of conditions and interrelationships existing on the
farms of the locality, rather than on the research station.
Tractor-based, ideal irrigation and input management systems on the
research station may show impressive 'results' but have little
applicability on the small holdings that, by necessity, are cultivated
with a hoe or animal power, have access to limited inputs and
irrigation, or rely on rainfall. Farming systems research on cropping
patterns and production .practices, regardless of the focus, must be
conducted in terms of the farmers' resources, not the government's.
resources. The farmers require safe bets that a new technology will
work. The scientists thus must know the local situation and be able to
view it from the farmers' point of view dictated by the circumstances
under which they exist.
In the final analysis, research is justified only to the extent that
applicable, profitable technologies, innovations and inputs are
developed, and that knowledge is communicated to and applied by the
The Project Paper envisages a multi-disciplinary small-farm research
program. The primary research thrust observed on the stations and the
on-farm trials for accomplishing this deals mainly with the agronomic
aspects, with some attention to water management. There is less evidence
of meaningful economic and social research and no evidence of livestock
and animal power research.
Cropping systems research (CSR) is presently the major activity under
the general heading of farming systems research (FSR). The emphasis in
this work, variously conducted by groups within BARC participating
institutions such as BRRI and BARI, is (a) to compare alternative
cropping patterns (CPs) or rotations and (b) to explore the potential
for intensification (usually to three crops a year).
The common themes in CSR are : (a) conduct of a benchmark survey to
record existing GPs and other farm practices and details including
cropwise profitability and the agricultural labor availabilities and
utilizations, (b) design of 'improved' CPs featuring intensification,
improved varieties, non-traditional crops (e.g. potatoes, maize,
vegetables, and sesame etc), improved cultural practices (ranging from
minimum tillage and more fertilizer, to use of insecticide and
pesticides, irrigation etc), (c) budgeting of costs and returns of
'improved' and existing CPs and (d) distillation of findings into
recommendations that take into account the circumstances of different
categories of farmers identified in step (a).
Even this limited work represents an improvement from previous methods
for testing new technology. In the past, new techniques were tested for
yield increases on the experimental stations and then tested by BADC and
BARI's On-Farm Trials Devision. These tests did not consider the impact
of the new techniques on other crops in the rotation, and they'did not
collect any economic information. They simply recorded the yield. Now,
at least, the entire rotation is being considered, and it is being
compared with farmers' cropping patterns on the basis of costs and
The IADS team inherited the present FSR sites and methodology from their
predecessors who had used the BRRI/IRRI methodology. They have been able
to make improvements in the field work and data analysis by working with
the scientists in the field and running short courses in statistics. Dr
Greene is about to start training the economists at the sites in partial
budgeting, which.should lead to site-level analysis that is more
appropriate. They will soon be opening up new sites. They hope to
develop them using the CIMMYT-based methodology which was quite
successful for IADS in Nepal.
The involvement of the anthropologist Dr Wallace at the BJRI site in
Manikganj will provide a model for greater involvement of
anthropologists and sociologists at these sites. He is collecting much
more comprehensive information about the village than are other sites at"
present, including information on animals and the spread of new
technology. The analysis of these data will hopefully provide an example
of what can be accomplished by moving outside of the narrow CSR context.
The philosophy of FSR is holistic in its declared ambitions of being
farmer-based, problem solving, comprehensive, interdisciplinary,
iterative and dynamic, and socially responsible. Accordingly, it must be
recognized that the present rather narrower focus on CSR is but a
partial step towards more comprehensive FSR. Amongst the aspects
presently seemingly neglected are (a) the interrelationships between CPs
and livestock (in terms of both provision of fodder to animals in
intensified cropping, and of traction and manure services from animals
to crop production) and (b) the linkages of farm labor to off-fa-m labor
Present labor accounting seems to be concentrated on agricultural
aspects only. Thus, for example, it is not clear that the additional
labor required for some of the 'improved' intensifications will actually
be forthcoming in a timely manner. There are related difficulties over
the costing of labor in computing total variable costs, gross margins
and benefit/cost ratios It seems that all labor, including different
categories of family labor, is valued at set rates that approximate
opportunity costs of full employment outside of agriculture (e.g. a rate
of Tk.25/day of adult male labor is used at the Hathazari site). If
actual opportunity costs of farm family labor are much less than this
for most of the time, this accounting procedure must understate the
actual profitability of all compared CPs, but especially the most labor
There may be other accounting difficulties in the present analysis of
profitability. The likely most serious one concerns costs of credit that
would be required by most farmers to be able to implement the 'most
profitable' CPs, especially those involving potatoes and intensive
production of HYV rice. Inclusion of realistic costs of such credit
would give a more accurate impression of profitability.
Relatedly, there is the-potentially important question of risk
associated with hopefully profitable 'improved' CPs. The scientific
staff working on CSR seem generally sensitive to this issue in their
discussions of the work, but their concern in not reflected in Project
reports and thus, presumably, in the ultimate stage of formulating
recommendations. Proper accounting for risk is admittedly not an easy
task, but it does seems that some pragmatic procedures are required to
bring explicit attention to this matter that is of such concern to
target farmers in their adoptive decisions. One suggestion is to report,
in addition to the presently. tabulated sample mean cost, yield and
return data, data recorded for the least favourable on-farm results.
While mentioning the reporting of mean data,it would be helpful to
readers of such reports if fewer 'significant' digits were reported
(e.g. if number like 36482 Tk/ha were rounded to, say, 36 Tk'000/ha),
and if an indication of sample variation was included (e.g. including
sample coefficients of variation in parentheses after or under the
respective mean would be of great informative assistance). Comment
should also be made on the very circumscribed relevance of benefit/cost
ratios as computed in CSR reports. Under diminishing marginal returns,
it is such an inadequate criterion for farm planning that it is probably
best left unreported.
Most of the CSR testing trials are still at an early stage so that
investigators, hopefully in collaboration with both farmers and
extension workers, are yet to enter the intended phase of composing
concrete recommendations. Some such work must begin forthwith if the
findings of the more advanced CSR are to be effectively exploited. There
are, however, likely to be some further difficulties in this phase --
difficulties that have their origins in the foundation of the CSR
programs. It seems that site selection has been primarily on the basis
of ease of access for scientific staff and visiting farmers (and other
agricultural tourists such as Evaluation Teams). Thus, in terms of
relevance to more remote, backward and difficult areas, this calls into
question the social responsibility aspect of this form of FSR. The
problem is exacerbated by the careful selection of cooperating farmers
for the on-farm trials in terms of their cooperativeness rather than
their representativeness. While it is important that a start has been
made somewhere, it is to be hoped that subsequent extension of CSR work
will be in direction that will correct the present bias towards more
favored locations and situations. These suggestions are not inconsistent
with Harwood's (1982) suggestions for BARC site-certification of FSR
sites, and for closer linkages with the extension service.
Involvement of extension personnel in the process from selecting the
nn-farm trial sites, planning and implementing the initial on-farm
trials, modifying the program in line with findings and subsequent
extending to the broader farming community is the exception rather than
the rule. It has been used most effectively in the MCC sites and to a
lesser extent in Ishurdi.
The crops selected at the regional stations for on-farn CP tests seem
appropriate for the respective areas. The cropping patterns, to the
extent that they have been tested in on-farm trials seem to have been
appropriate. They are, however, and should be, subject to modification
as findings indicate the necessity to do so.
Work is proceeding with introducing "new" crops such as maize, summer
pulses and vegetables. This should continue. Rice is the staple food,
maize is eaten by some people as roasting ears and for poultry feed.
These purposes will have a limited market demand for at least an initial
period since eating habits and food preferences change slowly. Extension
and research must be cognisant of the effective demand and balance this
with their encouragement of farmers to increase production of the new
crops. The emphasis should not be totally on the production/yield
potential. A common complaint at all locations was the inability to
analyze the data collected due to the shortage of the personnel, and by
implication, competence to do so. At all locations it was felt that a
mini-computer would solve this problem, but few had focused on the need
for and source of competence for computer programing, data analysis
and interpretation. Another problem would be the extreme fluctuations 'of
voltage and regular periods without power. The most appropriate solution
would be to reduce the mass of data and focus on the most relevant types
of data for site selection, cropping system performance, and research
system and farmer feedback.
The practice of providing inputs, such as improved seed, fertilizer,
insecticides etc free presents a potential problem when cooperating
farmers subsequently have to purchase them. It is unknown yet whether
this problem will, in fact, arise as extending the cropping patterns to
other farmers has yet to be done. It would not be unusual for the
surrounding farmers to expect, and possibly demand free, inputs like the
cooperating farmers received. Regardless, it is essential that the
Project and scientists bear in mind that implementation of the new
interventions by the farmers will not occur unless the resource demands
are reasonably within the farmers' economic, social, physical and other
The most important test, of any cropping is whether it is being adopted.
Thus, CSR workers should be periodically monitoring the area around
their site to find out if farmers know about the new cropping systems,
if they have- adopted these new systems and, if not, why not. This
information does not require a detailed questionnaire. It should be
quickly tabulated at the site, sent back to headquarters and used for
redesigning the next year's experiments. At the time of this evaluation
many site personnel did not know whether anyone was adopting the new
Problems and Constraints
The CSR represents one of the most highly coordinated achievements of
the Project. The trials have been effectively coordinated in terms of
procedures and style of operation across the six participating
institutions (BARI, BJRI, BWDB, BAU, SRI, MCC) with the exception of
BRRI. Even here, however, there is a degree of coordination because of
the approach adopted --- stemming no doubt, from the apparently common
roots of the CSR programs in the IRRI tradition.
On closer inspection, however, perhaps the degree of coordination
achieved is not necessarily all desirable. This question can be broached
under four considerations:
(a) uniformity vs diversity FSR is not an 'agreed' approach in terms
of the particular methods that might best be employed and thus the
rather uniform approach to CSR that has been 'coordinated' or otherwise
imposed does not provide much opportunity for discovering advantages of
(b) duplication vs parallel development to the outsider, it seems
that there is a lot of the same sort of CSR work going on. This
impression of 'coordinated duplication' must, of course, be tempered by
the degree to which there is systematic coverage of the major
agro-ecological zones of the country (especially in terms of flooding
depth, soil type, and farming system), but the team was not persuaded of
this latter virtue;
(c) analytical vs encyclopaedic research perhaps reflecting the
still-early stage of the work, the present emphasis on accumulating data
rather then on testing hypotheses does not give a strong impression of
'science in action' The impression is more of a series of ad hoc
fertilizer cum variety trials without formal controls or unconfounded
component effects. The feedback loop between benchmark survey data to
technology design seemed generally absent, as was that between trial
results back to component tests;
(d) centralized vs decentralized interpretation related to the
foregoing considerations is the extent of responsibility for those
actively engaged.in the, CSR field programs. The situation varies .from
site to site but there 'is a tendency for those closest to the data (the
site coordinator agronomist and, where in place, the socio-economist) to
receive the plans, to conduct the surveys, trials and summaries (with
vigor) and then to hand over the 'results' for consolidated reporting
without the analytical attention mentioned in (c). It would be healthy
for these regional' staff to take on more of a scientific role in
interpretation, design, hypothesis testing etc and thus preserve more.
of a proprietory interest in the work. Again the main exception is MCC
in Noakhali and Comilla and that is the
actual recommendations being made.
The team was concerned that each CSR experience seems to be described
and discussed in isolation from other previous and contemporary
experiences and, most distressingly, from previous FSR. It is as if
there is no literature that is pertinent to the work. There is, but it
is seemingly not readily available to the scientists engaged in the
work. Library resources, in the FSR field especially, must be built up,
and access to these resources, facilitated for all involved research
Minimally, all FSR site offices should be supplied with a copy of the
(USAID sponsored) Shaner et al. (1982) review of FSR*, as a low-cost but
fairly comprehensive entry to the world literature. Desirably, BARC at
least should have an active subscription, to and retrospective holding of
the journal "Agricultural Systems" through which it to be hoped that
Bangladesh scientists will eventually submit their FSR work to
* Shaner, W.W.Philipp, P.F. and Schmehl, W.R. (1982), Farming Systems
Research and Development: Guidelines for Developing Countries, I'
The Team believes that the orientation to conducting research on
farmers' fields with the active involvement of farmers in the management
of the trials is commendable for its enhancement of relevance of
scientists' activities, for predisposing relevance of experimental
conditions, and for easing the two-way transmission of information
between research workers and farmers. However, it recommends that the
scope of the CSR work be broadened to encompass wider FSR
considerations, be strengthened to enable more explicit testing of
hypotheses, and be interpreted to emphasize adaptive feedback to
research planning and determination of priorities in future work.
Persuant to these general suggestions, the Team makes the following
several recommendations :
There should be more integration of the cropping research program with
the work of scientists on the regional stations or at Joydebpur. The
results of the cropping systems trials and other surveys such as the
proposed adoption surveys which focus on constraints to adoption should
be written up quickly and reported to scientists concerned with these
crops. The constraints that are identified then should be the basis of
further on-station work'. The IADS Specialists can play an important role
in improving this communication process.
Training of FSR workers at all levels should be activated, especially
for the 'scientific' staff including socio-economists. This should
include field trips to some of the more advanced CSR sites like the MCC
sites in Noakhali and Comilla. As their skills increase, more of the
research design and data analysis can be done on site.
To the extent that training is done overseas, some institutions should
be selected partly for their differences from the prevailing IRRI
approach (e.g. University of Reading, Lincoln College, CMMYT, Colorado
Participating institutes should move to ensure that all FSR projects are
staffed by at least one economist as well as an agronomist. More formal
economic modelling should be undertaken to extend the analysis of the
emerging CSR data (e.g. farm planning models to orchestrate efficient
whole-farm resource allocation). This may require some short-term
Consultancies as well as an appropriately interested second Speciatist
appointment in agricultural economics.
Reduction in the amount and simplification of data collection to
digestible levels will save considerable time and improve utilization.
The existing farm/plot size, cropping patterns and intensity, production
inputs and practices, source and quantity of water, amount and source of
labor and farm power, financial situation and/or access to credit, and
positive and negative social aspects should be adequate initially.
Expansion and refinement of data collection should be initiated later as
the human resources, equipment and/or competence to analyze and 'utilize
the data are available. The Team noted with interest the observation of
Dr Barker in his consultancy report "Everywhere the tendency has been to
collect too much data without thinking through the objectives and
analysis. We at IRRI have been as guilty as everyone else on this
Extension agents should be involved in designing the trials, monitoring
the results and monitoring the adoption of the'new technologies.
2.4 ECONOMICS AND SOCIAL SCIENCES
The Project Paper is not very clear about intended achievements:in the
fields of Economics and Social Science (ESS). On p.6 it endorses the
1980 report of the joint GOB/AID/IBRD review team which states:
"Agricultural economics and social sciences must be more fully involved
in problem diagnosis, planning research, monitoring and interpreting
results relevant to farm conditions." The contract specifies that the
agricultural economists will train local staff, organize a central unit
to process and analyze data, and "advise on socio-economic criteria
relevant to research activities and priorities". They will do "analysis
of the issues related to appropriate technology and employment of
landless and marginal farmers as well as women and family members"; They
will also "design and execute surveys on farm practices, problems,
employment opportunities and division of labor"(Contract, Appendix A.
The major achievements in the ESS area include (a) the initiation of
contract research projects, (b) teaching short courses on descriptive
statistics for people working with farming systems, (c) selection of
candidates for Doctoral and Master degrees, (d) assisting
anthropologist Dr Wallace to initiate his research, and (e) bringing Dr
Barker (Cornell) in as a Consultant.
The major topics chosen for contract research were important ones for
small farmers technology assessments of wheat, maize, potatoes and
other tubers; cropping systems, and irrigation management. The initial
Project proposals were not impressive. Recently, the application forms
and selection procedures for new projects have been greatly improved.
The IADS Specialists have helped to integrate the ESS cropping systems
research program with the Crops cropping systems research program. Dr
Greene has been upgrading the statistical skills of the scientists at
the sites and is now starting to work on their economics skills. If
successful, he will have made an important contribution to CSR because
the staff will have a better understanding of what they are doing, and
will take more pride in their work. Finally, the regular visits of Dr
Manzano and Dr Greene to the sites allow them to identify weaknesses in
design of field work, and encourage the staff to improve their methods.
Since Dr Gisselquist arrived, the role of BARC in the water management
area has increased. He has submitted several proposals to the Member
Directors which would further strengthen BARC's role in this area -
these include funding for a bibliography, staff support and library
facilities. The role of the Specialists is discussed..in more detail in
Social science research is useful if it -helps scientists allocate
research resources more effectively, helps farmers allocate their
resources more efficiently, or helps policy makers develop more
effective policies. The closest ties at present are with other
scientists. At BARI and BRRI, agricultural economists do discuss their
plans and results with other scientists. However, they seem to have
little impact on the plans of other agricultural scientists.
The agricultural economists of the BARC group of institutions do not
appear to be closely linked to policy-makers. The Team did not see any-
policy studies for the Ministry of Agriculture.This is probably due to
the fact that these institutes simply do not have sufficient well
trained social scientists.
The linkage with farmers is also weak. Most BARC-related social
scientists are located in Dhaka and Joydebpur. They go out occasionally
to interview farmers.The only social scientists in branch stations of
BARC institutions seem to be the economists associated with cropping
systems work. However, the regional economists did not seem to be
developing close contacts with farmers.On one CSR site the economist did
not know if any farmers had adopted the new patterns. They understand
little about the potential of cropping systems for solving farmers'
problem. Also they seem to play little role in developing the regional
stations' research program.
A number of major issues have been ignored by BARC. One role that ARP
II proposed to support was to use economic analysis to help plan
research through assess-ing the distribution of benefits and determining
the efficient allocation of research resources. The first Internal
Review recommended that BARC accelerate studies on returns to research.
However, BARC has done little in this area during the grant period.
Other issues of major importance have been proposed by various teams,
but so far BARC has not been able to respond. The 1980 GOB/USAID/IBRD
Team and Cushing(1983) emphasize the importance of rural employment and
the landless, and it is also explicitly mentioned in the ARP II
contract. The 1982 Internal Review and consultant Dr Barker recommended
that BARC support more research on price policy. BARC's only activity in
this area is Dr Ahsan's research on fertilizer with IFDC. Much of the
data that are being collected at the farming systems sites could be used
to deal with policy questions, but they have not been so used.
Technical assistance may not have been ideally used thus far.Dr Greene
is currently doing useful work. Many Bangladeshi scientists can
profitably use a course in descriptive statistics. It is a necessary
step which should enable the cropping systems sites to operate more
effectively. However, as Dr Barker stated in his report the major
constraint to agricultural economics research in Bangladesh today is the
ability to decide what questions to analyze" (p 9). It is in this area
that the expatriot economists should be able to make..heir .greatest
contribution. They can help develop this ability in Bangladeshi social
scientists through joint research projects, assisting in the preparation
and review of research proposals, and through short courses on research
methods. If they choose to do short courses, the emphasis should .be on
identifying important questions, developing them into testable
hypotheses, and -then testing them.
Dr Wallace seems to be an inexpensive bonus for the project. He has an
interesting research program which should help to broaden the CSR and
also put the project in touch with social scientists other than
agriculturall economists. The second agricultural economist has not yet
been hired. IADS has not been able to come up with an acceptable
The contract r-esearch also does not appear to have been well used. In
preparing the research proposals, scientists should have identified
important topics, reviewed the literature, and decided on the best
procedures for testing the hypotheses. There is little evidence that
this was done in the first 10 IDA and USAID contract research projects.
However, if social scientists utilize the new forms in the Yellow Book,
they will have to deal with these issues in the future.
The titles of the USAID financed projects illustrate the lack of
precision that runs throughout the proposals i.e "The Economics of
Irrigation in Bangladesh." or Socio-Economic Research on Jute Farming
in Bangladesh". The proposals themselves are equally vague and'
unrealistic about what can be accomplished in 2 years with $20,000. The
the result has been the Team's observation that, in FSR, agricultural
economists are compiling mounds of data that are not being used to
answer important questions. These data could be used to help to guide
research planning on station, to assess the impact of new technology,
and to answer some price policy questions, but this is yet to happen.
The contract research ha-s made a limited contribution to the integration
of the research system..The grants were given out to a small number of
institutions and these institutions do not include most of the top
social scientists in Bangladesh. The ten projects were divided between
5 principal investigators at 5 institutions. The cropping systems'
projects brought BJRI, BARI, and BWDB closer together, but there was
little other integrating effect. About $120,000, or 44 percent of the
total IDA and USAID financed contract research in ESS, was given to
BARI. No grants were given to BAU, which has a strong economics faculty,
or to BIDS, Dhaka University,.Chittagong University, or Rajshahi
University. All of these institutions have active research programs in
ESS and should not have been neglected, as seems to have been the case.
By April 1983, ESS has only been able to spend 26% of its total
scheduled expenditure. The major shortfalls are in Specialists' salaries
and commodities (especially the computers). Degree training is also way
behind, but this will solve itself because candidates have been selected
and are starting to depart in May 1983. The commodities problem could be
solved relatively quickly if the problem of customs duties is settled.
The problem of the Specialists' salary is because the second
agricultural economist has not been hired. Consultants have not been much
used, in part because BARC cannot get government approval.
Constraints to Progress
These problems have arisen due to a variety of constraints on BARC -and
IADS. First, the Member-Director of ESS is also acting as the
Member-Director for Technical Support Services and is unable to give
full attention to Social Sciences. Thus, he has little time to keep up
with the literature in Agricultural Economics, to do research himself,
Dr 1- develop communication between social scientists in different
The social scientists working on agriculture are scattered through a
wider range of institutions -than are other agricultural scientists.
AlmoFt all agronomists work- in the BARC group of institutions, but most
of tle best rural social scientists are outside the BARC group. An
additional complicaton is the split that exists between agricultural and
nonagricultural social scientists. In agronomy, virtually every one
studied at -the same institution (BAU or the Agricultural College) In
social sciences they come from more diverse institutional backgrounds.
Thus, promoting cooperation and communication among social scientists
seems, in the Bangladesh context, to be much more difficult than in
A third constraint is the inability of IADS to recruit acceptable
candidates for the second agricultural economist position and the two
water management positions. This means that Dr Greene must do a
considerable amount of work with BARI people and Dr Gisselquist has to
spend much of his time on the non-economic aspects of water management.
A fourth constraint is the shortage of well trained agricultural
economists within the BARC group of institutions. At present the only
PhD agricultural economists are located at BARI, BARC, and BAU. The Jute
Research Institute does not even have an agricultural economics section,
and so agronomists are trying to do the necessary analysis. Most
regional stations and substations also do not have any agricultural
A fifth problem has been the reluctance of GOB to approve some nominated
In.keeping with the needful completion of the panel of Member Directors,
the Member-Director for Social Sciences should devote full time to
developing the social sciences. This is an extremely important area.
Social Scientists are widely scattered throughout the country and
improving communication could be very important in preventing needless
duplication of research and stimulating more research of higher quality.
The Team also recommends that IADS make greater efforts to recruit
qualified people for -the economics and water management positions.
BARC should make better use of the economic skills of the specialist
agricultural economist. The Team suggests that he should be spending
more time trying to develop the economic analytical ability of
Bangaldeshi scientists, and less on teaching them statistics. The
statistics might be turned over to an organization like the Rural
Development Academy Bogra, the Graduate Training Institute at BAU-, or
the Institute of Statistical Research and Training, Dhaka.
The Team recommends that BARC should try to encourage communications
between agricultural economists in the BARC group and social scientists
at BIDS and the general universities. This could be accomplished through
available in Bangladesh and teach him how to do research here. Thus,
when he returns he is ready to work. This has the additional benefit of
producing research on Bangladesh problems.
The Team suggests that short courses on writing research proposals and
research reports be instituted. Courses on research management could
also be useful.
One of the expressed purposes of this project was to train scientists to
work with farmers. Thus, BARC should consider funding and expanding the
BAU internship program which was initiated last year. This program, with
the support of E and RP and BAU, provided three week internships for 125
outgoing students in five BARI substations in the Northwest. Faculty
members also visited these substations to evaluate the program. Thus,
both students and faculty members got some exposure to work on
substations, and the T and V system in the Northwest. The weakness in
the current program is that it does not put students directly in touch
with farmers. An expanded program should include several weeks for doing
a survey of villages around the station. These surveys should deal with
the relationship between the farmer-and the substation such as: Does the
farmer visit the substation? Does he have any influence on the program
of the substation? Has he ever learned anything from the scientists?
seminars, improved bibliographic and professional literature services,
contract research, and membership on the technical advisory committees.
The Team recommends that BARC try to bring the social scientists from
outside the BARC group into contact wih agricultural scientists (who are
not social scientists). This could be done through sabbaticals,
seminars, short-term training and contract research.
The Team suggests that BARC should broaden the range of research topics
that it supports. It should work towards more social science
participation in planning research through ex post and ex ante analysis
of research investments and their impacts on various groups. BARC needs
to encourage more research on the impact of new technology on the
landless, women, and marginal farmers. Studies on the diffusion of
innovations are another largely untouched area.
The team suggest that BARC encourage more social scientists to spend
sabaticals at BARC institutions as Dr Wallace has done.