C I I M Y T'S E A S T E R N A FR I C A
CENTRO INTERNATIONAL DE MEJORAMIENTO DE MAIZY TRIGO
INTERNATIONAL MAIZE AND WHEAT IMPROVEMENT CENTER
CIrPIUYT's mandate is to work with national programs in developing
and diffusing new technologies in maize and wheat. In pursuing
these ends the Center focuses on improving the genetic material
available to national programs, on investigating the interaction
among improved genetic material agronomic practices, and agro-
climatic factors; and on training nationals in breeding and produ-
CIIPMYT's economics section supports the activities of the crops
programs in developing and diffusing new technologies. It is
recognized that many elements impinge on farmers' behaviour and
on their response to new technology. Some of these forces are
natural, related to weather and soils. Other are biological,
related to the interaction between plants and their environment.
Still others are economic related to markets and to the allocation
The farmer reaches decisions about his farming activities by
blending these elements. Research aimed at developing technologies
must incorporate the same elements or run a substantial risk of
being irrelevant to farmers. C11MiYT's economics section emphasises
incorporating the economic dimensions of farmer circumstances at
the earliest stages of programs for developing and diffusing new
technologies, "that is in planning the adaptive research work."
CIMMYT believes that appropriate technology is the key to successful
i agricultural development efforts. In rural development programmes,
huge institutional and infrastructural edifices, perhaps covering
roads, water supplies, marketing, extension and credit services are
built up around a chosen technology to enable its use by farmers.
If the wrong technology base is chosen for the programme, not only
has the research effort in producing the technology been wasted,
but a whole range of development resources have been misapplied.
Even if the sheer weight of effort pushes the new techniques to some
farmers,.diffusion through the farming community will be poor and
adoption is unlikely to be sustained once the special push is
The more appropriate the new technology is to the needs and conditions
of the majority of farmers in the community, the more rapidly will it
spread and the more farmers it will eventually reach. A comparison
of the effects of introducing (relatively) inappropriate (I) and
appropriate (A) technology into a farming community is shown in
60 (A) /
of farmers i
using the 40 ,
innovation / Benefits to (A)
over 5 years.
20 // /
(I) .Benefits to (I)
over 5 years.
1 2 3 4 5
The appropriate technology (A) reaches 75 of the community in a three-
year period. The (relatively) inappropriate technology reaches only
20,I, of the community over a five-year period. The incremental economic
benefits to A over the five years, represented by the shaded area, are
several times the benefits to I, possibly from a similar level of
Government investment in research, enabling facilities and services.
The social benefits of reaching 75% of the community in three years
compared to 20% in 5 years are equally clear.
"Increasing the speed and extent of diffusion of improved technology
through a fanning community is a major thrust of CIITIYT's efforts.
CIInIYT has thus placed priority on improving procedures for formulating
technologies. Economics has a significant, if little understood, role
to play. Developing that role, demonstrating its utility, and promo-
ting its use are the goals of the East African program in economics.
All of this will, of course, add to the dimensions of agricultural
economics research in East Africa, strengthening one facet of the dis-
ciplines contribution to development".
TIIE ECONOMISTS CONTRIBUTION TO RESEARCH PLAhNNING
SThe key to these increased benefits in agricultural development efforts
is in identifying and designing appropriate technology through problem
oriented research efforts.
The desirability of a contribution from economists is based on the fact
that economics whether of family security or profit plays the
dominant role in farmers' decisions. Farmers decide what they will
produce, how much of it they will produce and how they will produce it
in the light of the full set of circumstances in which they find them-
selves. This set of circumstances includes both natural conditions of
climate and soil, important to the farmer for their economic implications,
. /3. .
and availability of resources including capital and technology. In
order to design appropriate improved technology, the orientation of
research work and the content of experiments must reflect as closely
as possible the set of circumstances which the farmer faces in taking
his production decisions. The work of the economist is in describing
the set of circumstances in which the farmer operates, in evaluating
how these influence his production decisions, how he manifests these
decisions in his farming system, and how this will colour the accepta-
bility of the many potential facets of a new technology.
The acceptability of new techniques is coloured by the contribution
they make to the farmer's basic production objectives: family security,
particularly important to the subsistence and semi-subsistence small-
holder, and profitability. iany breeders and agronomists now accept
that yield, in terms of physical output per unit area, is an inadequate
criteria to measure the potential impact of a change in management
,practice on farmers' production objectives. Experimental yields,
unlikely to be reproduced under farmers' conditions, are particularly
problematical, only when representative of expected results in farmers'
yields do they become a starting point for evaluation in economic terms.
For evaluation, in addition to valuation of the physical output in
market terms, yield data need three major qualifications to reflect
farmers' production objectives adequately:
(1) By the costs to the farmer of achieving the yield;
(2) By the year to year variability; of the yield, and of
the costs of achieving it;
(5) By the complexity of the management re-organisation
required to implement the practices needed to achieve
Each set of qualifications creates 'costs of change' for the farmer,
and deserves further comment:
(1) Farmers costs.
Costs to the farmer as a result of adopting new technology can be put
into three categories:
(a) Direct costs; items to be purchased such as fertiliser,
insecticides, sprayer, etc. in order to implement the
(b) Increased labour costs when the methods involved in impleme-
nting the new technology and harvesting the extra output
take more labour than under existing management.
(c)- Opportunity costs, these are particularly relevant to the
small farmer who has no access to, or the means for, hiring
extra resources. Modifying the timing of or increasing his
labour input in implementing the new technology may be at
the expense of another crop which may not get planted, or
may not get wooded so well, so that its yield falls. These
losses are properly a cost of the new technology to the
farmer. These costs may be high when the losses are of crops
which contribute to security of the farm family food supply,
or which are highly preferred as food by him and his family.
Opportunity costs are often difficult to measure in money
terms; nevertheless their identification, if not precise
measurement, is vital as they are often decisive to the farmers'
reaction to now technology.
(2) Variability, year to year, in yield, _rice and costs
Stability in the results to be expected from new technology is also of
particular importance to the small farmer, especially, in the case of
the subsistence or semi-subsistence farmers, with new technology for
their food crops. Such farmers live relatively close to subsistence
level. Significant year to year variations in output, which to a large
grower may make the difference in the size of car he can afford to run,
may mean semi-starvation to the small farmer. Thus, although on a year
to year average, results may be better from the new technology, with
family security as a dominant objective of small farmers, wide variations
in yields are unacceptable. Interactions between crops in the farming
system are inevitable. ITevw technology on cash crops often demands
changes in management on basic food crops. This also may increase the
year to year variability of food production, reducing farmers' security.
Similarly security may be reduced by the uncertainty created by wide
fluctuations in the costs of inputs required to implement the new
(3) management complexity
Absorbing new technology into the farming system demands 'a degree of
management reorganization. The more complex the repercussions it creates
within the system, the more complex the reorganisation required. Hana-
gerial ability is an attribute distributed approximately normally in
Sa farming population. The more complex the reorganization required to
absorb new technology, the fewer the farmers with the required ability
and the poorer the diffusion of the technology over the population.
Each component of a technological package (improved seed, correct spacing,
pure stand, correct time of planting, timely weeding, use of fertilizer,
etc.) contributes both to added profitability and to added management
complexity. Dropping components which make large contributions to mana-
gement complexity increases the degree and speed of diffusion over the
farm population. More farmers are capable of handling the reorganization
required by simpler.new technology. At the same time, dropping such
components entails some loss of profitability. Using hypothetical examples
based on Fig. 1; if (I) is a complex technological package and (A) a rela-
tively simple one, the comparison of benefits in a 100,000 farm population
1) (A) (a) Profitability per farm of 3.90, spreads to 75, of
farms over 3 years at 25/ per year.
(b) Aggregated benefits of s.27 million over a 5-year
2) (I) (a) Profitability per farm of ;.120, spreads to 20% of
farms over 5 years at 4% per year.
(b) Aggregated benefits of ".7 million over the 5-year
From a similar investment in government services, gross economic
benefits to the country would be three times as high from A. and would
be shared over 75% of the farming community, against 20% for benefits
from I. Such a comparison, although based on hypothetical examples,
cloarly demonstrates the importance of management complexity as a criterion
iu identifying appropriate new technology.
- 5 -
To illustrate the variety of facets which can be involved, a hypothetical
example of an existing and recommended maize growing practice, showing
what each offers the farmer and detailing the costs of changing to the
recommended practice from the farmers point of view is set out as an
annex to this programme description,
The critical contribution of the economist is a description of the
existing system in farming populations to allow advance diagnosis of the
likely importance of these three sets, or sources, of costs of change.
.7ith such a diagnosis, agricultural researchers can guide their programmes
into those components of new technology most compatible with farmers'
production objectives and therefore most appropriate to the local situation.
In short, the aim is to identify the line of least resistance to change in
the existing farming system.
PROCEDURESS POR A CO-OPERATIVE EFFORT BET'EEIN NATURAL SCIENTISTS AND
ECONO I STS IN PLANNING ADAPTIVE AGRICULTURAL RESEARCH
Considerable space has been used in setting out the role of the economist
because of the newness of the idea of his contributing to research design.
As an agricultural research centre, CIHI.IYT is by no means suggesting a
take over by economists of research planning, almost certainly a disast-
rous step. The success of the approach recommended depends on inter-
action between natural scientists, as experts in crop and livestock
improvement, and farm economists, as experts in the economic circums-
tancen facing the farmer. There has been considerable reference in the
literature to the need for an interdisciplinary approach to research
problems in developing agriculture, but little practical guidance as to
how such an approach might be organised. Having noted that the economist
has been playing an exposte role, usually in criticising the inappropria-
toness of the results of adaptive research programmes, CIITiYT economists
have sought procedures for a positive approach by bringing the economics
input into research planning. To do this, a rapid turnover of data is
needed which precludes a protracted full farm economic survey. The
emphasis is on descriptive information and an intuitive evaluation of the
relationships it reveals, the whole effort demanding months rather than
years. The sequence of procedures which follows is essentially an inter-
action between natural scientists and economists to produce a more
) relevant orientation of research programmes. The sequence would be
appropriate for scientists working from a centre with responsibility for
adaptive agricultural research in a region with a farming population
operating under a variety of natural and economic conditions. The sequence
ban cover the orientation of as many crop or livestock research programmes
as are mounted from the centre. The economist will liaise with each of the
specialist groups involved. The sequence is outlined in relation to a
single crop programme.
(1) Crop specialists, breeders, agronomists, pathologists, entomologists
and soil scientists relate critical aspects of crop physiology to natural
conditions as they occur in the region. Working from basic principles,
they attempt to establish the relative importance of possible package
components to potential yield, given local conditions of climate and soil,
and the prevalent pathogens.
(2) The economist outlines critical aspects of crop economics, present
end uses, product specifications, cropping patterns and existing production
technology for all areas of the region.
(3) From these two sets of information
(a) The region is divided into 'Recommendation Domains' or
'Customer Zones'. This is an initial agro-economic zoning of the
target population into domains which, because of the variability
among them in natural potential or economic conditions, are likely
to require different technologies and for which different experi-
mental programmes may be justified. Zoning also creates a frame-
work for deciding research priorities. If the criterion is the
greatest benefit for the greatest number, other things being equal,
zones with the largest farmer population would take priority in
experimental work. However, crop scientists, economists and the
policy planners would weigh factors such as the number of growers
of the crop in each zone, the importance of the crop to these
growers, and the potential performance of the crop in each zone
in allocating priorities.
(b) The economist builds a schedule of descriptive information
required for each zone as a whole and for the on farm situation in
each zone. The schedule for the on farm situation seeks a descri-
ption of how farmers currently manage the crop, and how this
management interacts with management practices on other crops in
the farming system. It emphasises how farmers presently manage
those aspects of the crop which the specialists have identified
as probably important to yield potential.
(4) The economist makes a preliminary, informal pre-survey of the zone.
He discusses the scheduled information with officials and farmers
in the area over a period of one or two weeks. This pre-survey
verifies the homogeneity of farming within the zone and provides
a basis for the organisation of a farmer survey.
(5) The economist mounts a farmer survey within the zone. The main
objective of the survey will be to verify the incidence of features of
existing farm management and identified farmer problems among the farming
population. Close coordination will be required with crop scientists in
drawing up the survey content and training the enumerators. The survey
will be used to establish the incidence of pests, diseases, present
farming, as well as economic features of the farming population. The
information is analysed and used, together with that collected less formally
during the pre-survey, to give a detailed description of existing farm
management practices. It will show how each facet contributes to the
achievement of farmers' production objectives. The economist will identify
the 'costs of change' related to each potential component of an improved
(6) Crop scientists discuss with the economist the weighting of these
costs and the implications for experimental programmes aiming to develop
technology to take advantage of 'lines of least resistance' into the
existing system for easy absorption by the farming population. There will
be four areas of discussion:
(a) Programme framework. It is important to breeding, agronomy and
crop protection work that a general context is fixed for the
research programme. Ideally existing farm practice should form
a basic framework; any step away from it may contribute to
profitability and to 'costs of change' and need assessment in
the programme itself. Mixed cropping is a particularly pertinent
S. ./7 ....
and critical example. For zones in which.farmers grow .(say) 80 of the
crop in question mixed with other crops, an explicit decision is required
whether improved varieties and management .practices should be sought
within the framework of a crop mixture. Crop scientists will evaluate
the physiological inter-relationships between plant components and between
plants and the local soil and water conditions. The economist will
evaluate the importance of the practice to the farmers' production
objectives and the management complexities consequent on reverting to
pure stands. It may be decided .that mixtures are crucial to water r utili-
sation or soil conservation, and to the subsistence farmers' preferred
food supply patter.nsor that the managerial consequences of changing to
pure stands would be too complex for most farmers to cope with In this
case, the best strategy would be research to improve the productivity of
the mixtures; breeding and agro:'omy within the context of mixtures already
present, On the other hand it may be decided that there are no physiolo-
gical or crop protection or soil conservation advantages to mixtures, nor
are the managerial consequences of changing to pure stands too complex
In this case the experimental work would be done in the context the crop
scientists believe would offer the greatest potential for the improvement
of crop productivity. The important point is that the framework for the
experimental programme is given explicit consideration and not settled
by default. An important part of the framework for consideration would
usually be whether experimentation should be done within the rotational
sequence followed by'the farmer
(b) Specification of supplementary breeding criteria Present cultural
practices as well as harvesting, processing and storage techniques will
have implications for desirable features in any new varieties An
economic assessment can be made of the costs of changing present practices
or, alternatively, of the losses i.xvolved in putting a poorly adapted
variety through the existing techniques.
(c) Locational characteristics of experimental fields The descriptive
survey data will allow the location of experimental work on the soil
types and topographical situations characteristic of farmers fields
(d) The content of agronomic trials. Perhaps the most detailed area
of discussion is the content of agronomy trials. Having emphasised the
need to document the profitability and complexity of each step away from
existing farmer practices to allow the compiling of appropriate techno-
logical packages, in practical terms few research programmes have the
capacity to treat all management practices as experimental variables.
The selection of experimental variables requires a detailed dialogue
between the agronomist and the economist.
(i) The description of present management practice from the survey
provides a basis for control treotnents in the agronomy programme,
Similarly it provides a basis for the levels of all non-experimental
(ii) All potential experimental variables facets of management which
the agronomist believes will improve the yield of the crop should
be discussed by the agronomist and the economist The agronomist
will be concerned to estimate, from first principles, their likely
impact on yield, The economist will be concerned to establish
the 'costs of change', including management complexity, for the
farmer in absorbing each facet Those facets which will -have no
or low costs of change but have a significant potential impact on
yield may be readily incorporated as blanket, non-experimental
treatments for all plots except the control under farmers' practice.
(iii) Out of the remainder, candidates for the experimental variables,
a balance will be struck between the -:eighting given by the
agronomist in terms of possible contribution to yield and that
given by the economist in terms of 'costs of changCe'. Initial
factorials should incorporate as experimental variables those
facets of management with the highest possible contribution to
yield and the lowest 'costs of change' to the farmer
This sequence of interaction between crop scientists and the economist
will be extended to lower priority customer zones as manpower and funds
permit expansion of the research programmes There more quantities
information on existing resource use and productivity is required to
crystallise the costs of change to the farmer, a full scale farm economic
s survey may be mounted as a follow-up to the initial survey aimed prima-
:r;: rily at identifying the nature of these costs, In most cases the initial,
rcn-quantative but rapid survey will be adequate to provide clear research
Guidelines. The diagram attached shows the sequence.
THE OPERATION OF CIMILYTT'S EAST AFRICAH REGIONAL ECONTOIIC PROGRAMIE
CItIMYT fully appreciates the scarcity of research personnel and funds.
The initial and very limited projects are as a result of an interest
expressed by research administrators, and are in association with on-
going research programmes, The aim, as already stated, is to convince
workers in these programmes and particularly the administrators
responsible for research policy, that this interdisciplinary approach
has considerable potential benefits. Benefits measured in terms of
the effectiveness of the research programmes in producing new techno-
logy which is appropriate for and therefore rapidly accepted by farmers
Given interest by research administrators, CIIMYT's Regional economist
f works to bring together local professional staff in small demonstrations
of the approach. CIMIYT provides professional advice and supplementary
funding if required.. By co-operating with interested local professionals
rather than using its own staff, CIIUIYT seeks to create an awareness of
the approach, and the capacity to implement it, within national research
The belief is that if the demonstration is convincing, a demand for this
approach will arise amongst research workers and a willingness to supply
funds and manpower will arise among research administrators. At this
stage CIM!.YT will cooperate with research administrators in identifying
ways for incorporating the approach into the national agricultural
research planning process.
. 0. .* .0 0. 1 ..
INTERACTIONS BETVEEM CROP SCIENTISTS A'.D ECONOMISTS IN TIHE
ORIENTATION AND PLAT"!ING OF A PROGRAMTIE OF AGRICULTURAL RESEARCH
Identify vital crop characteristics
to guide initial zoning and focus a
IdentifAy market and production
characteristics of the crop to
guide initial zoning and focue
a preliminary survey /
INITIAL ZONING AND DOIIAIN
DEFINITION SELECTION OF
Assess possible improvements in crop Assess" costs of cK'/"'e'
yield from all potential revisions in associated with:.v'ri'ofty
management practice and breeding characteristics and potent al
criteria revisions in management
CLOSELY FOCUSED TRIALS
PARIER rEFCC: -IE!1DATIONT
S ANNEX: AN EXAMPLE OP COSTS OF CHANGE
Present practice: Maize production is wholly for subsistence.
Mlaize is grown in mixtures with groundnuts and sweet potatoes.
Maize is planted into the prepared field spaced approximately 75 cms
between rows and 75 cms in the row. A weeding is done at three weeks
when grounds and sweet potatoes are interplanted and rapidly cover
the ground. No further weeding is done. Three or four fields are
planted with this mixture at intervals, as the rain falls, between
mid-November gnd mid-January. A short term variety of maize is used
for the first and final, plantings.
This enterprise'gives the farmer the following returns:
(a) Dry maize, usually sufficient to feed his family until the
following harvest, mainly from the middle plantings of a longer term
variety which has good storage qualities under traditional storage
(b) A complementary, high flavour food in the form of groundnuts
used to garnish the maize staple as the basic dish in the local menu.
(c) The staggered planting times give a prolonged flow of fresh foods;
initially green maize from the short term variety planted early, then
green maize from subsequent plantings followed by the sequence of
sweet potatoes.1 In a good year the family will have fresh foods
available as part of their diet the season round, as the sweet potatoes
can be kept in the ground and used, fresh, as required.
(d) The staggered planting reduces the risk of crop failures maize
will be at varying stages, with varying water requirements,' over dry
(e) The short term variety used in the first planting gives early
food availability in seasons following poor harvests. The short term
variety used in the final plantings fits into the water availability
pattern where a longer term variety would come under sever.ei~tress at
tasselling from the same planting date.
Recommended practice for maize growing: PTaize should be planted in a
pure stand in the first week of December. Fresh hybrid seed ( along
term variety) should be purchased each year. The seeds should be
spaced 75 cms between rows and 30 ems between plants in the row.
VWeeding should be done at 2, 5 and 8 weeks. 50kg/ha P should be applied
in the seedbed and 75 kg/ha N top dressed after the second weeding.
At knee high the crop should be protected from stalk-borer by the
application of DDT dush into the funnels of the plants.
This gives the farmer (if planted on the same area as his traditional
maize mixture enterprise) more dry maize than he needs to feed his
family in most years. A relatively short period of green maize before
the crop dries off. It can be questioned whether the new maize
enterprise is really a substitute for the old one. Costs of changing,
as they will appear to the farmer, are listed under the three sources:
Direct costs, Opportunity costs and Complexity.
(1) Direct costs Direct costs of the change will be for hybrid seed,
P and N fertilisers and DDT dust.
Where hired labour is used, the costs of extra labour for planting,
increased weeding, fertilizer application, thinning and harvesting the
larger crop will all be direct, additional costs to be set against the
value of the increased output. The changes in the timing of operations,
when it creates peaks, may require an additional work force to adhere
to the recommended crop calendar. These would be additional direct
(2) Opportunity costs In farming situations where additional resources
are not available for one reason or another, opportunity costs are
incurred when resources are re-allocated from other uses to implement
the change and are represented'.by the lost output from other uses or
opportunities. In situations.where there is a full knowledge of the
existing farming system, of 8he resource allocations and their produ-
ctivities, these costs canbe quantified. In situations where there
is little or no knowledge of'lthe farming system, quantification becomes
difficult and even impossible. Even where there is knowledge of the
system, quantification is difficult where risk is a large factor
weighting farmers' decisions, and when a significant part of the farm
operations are in subsistence production. The opportunity costs in
this example can be divided into: (a) Possible quantifiable; (b) not
quantifiable in practice.
(a) Possible quantifiable: ; here there is no labour market and
existing labour resources must be re-organised to implement the change:
(i) Extra labour required may mean reducing the production of
another existing enterprise.
(ii) The concentration of planting date, or indeed any change in
the timing of operations, may create or increase labour peaks,
at planting time, weeding time or harvest time. This also
may enforce a reduction in the si.ee of other farm enterprises,
the losses from which would represent a cost of the change.
(b) Not usually quantifiable in practice.
(i) Concentration of planting time will increase the risk of
crop failure in areas of uncertain rainfall.
(ii) Ve can say that the recommended hybrid is a relatively poor
keeper under existing storage and processing practices.
(iii) The single planting time reduced the length of the period
when green maize as a preferred food is available. Together
with the fact it is a long term variety, this also sacrifices
early staple food in years when stocks are poor and sacri-
fices early green maize from the early planted short term
variety in all years.
(iv) The pure stand involves: the sacrifice of the legume inter-
crops grown as complementary food; the sacrifice of sweet
potatoes which gave a fresh food to follow the period of
(3) Complexities Several facets of this change create complexities in
the management re-organisation required to absorb it successfully.
(a) Growing maize in a pure stand raises the problem of how to obtain
the legumes and sweet potatoes valued for their dietary contribution.
(b) Shifting the time of planting, and consequently of most subsequent
and preceding operations, involves re-scheduling of long-established
routine timetables both on the maize mixture enterprise and other
enterprises. Where there are periods of labour shortage or capital
scarcity, re-allocation problems may present difficult management
(o) The purchase and application of inputs demands a new management
function. It may also aggravate the capital allocation problem.
'7hich of these costs and complexities are relevant, and whether
the extra labour requirements will be direct costs or will create
opportunity costs, depends on a detailed description of the local
farming system for which the adaptive research programme is
being planned. But it is clear that evaluating the true profi-
tability of such a change to the farmer requires consideration
of all these aspects.