MSU, Rural Development Papers
MSU Rural Development Papers
MSU Rural Development
Paper No 5
The Farming Systems
Approach: Relevancy for
the Small Farmer
David W. Norman
Department of Agricultural Economics
Michigan State University
East Lansing, Michigan 48824
MSU RURAL DEVELOPMENT PAPERS
Carl K. Eicher and Carl Liedholm, Co-editors
The MSU Rural Development Paper series is designed to further the
comparative analysis of rural development in Africa, Latin America, Asia
and the Near East. The papers will report research findings on community
development and rural development in historical perspective as well as on
contemporary rural development programs. The series will include papers
on a wide range of topics such as alternative rural development strategies;
non-farm employment and small-scale industry; marketing problems of small
far'iers; agricultural extension; interrelationships between technology,
employment, and income distribution; and evaluation of rural development
projects. While the papers will convey the research findings of MSU
faculty and visiting scholars, a few papers will be published by re-
searchers and policy-makers working with MSU scholars on cooperative
research and action programs in the field.
The papers are aimed at teachers, researchers, policy-makers, donor
agencies, and rural development practitioners. Selected papers will be
translated into French, Spanish, and Arabic. Libraries, individuals,
and institutions may obtain single copies of the MSU papers free of charge
and may request their names be placed on a mailing list for periodic
notifications of published papers by writing to:
MSU Rural Development Papers
Department of Agricultural Economics
206 International Center
Michigan State University
East Lansing, Michigan 48824
THE FARMING SYSTEMS APPROACH:
RELEVANCY FOR THE SMALL FARMER*
David W. Norman
Professor of Agricultural Economics
Kansas State University
*This paper is published as part of the farming systems research program
presently carried out by Michigan State University under the "Alternative
Rural Development Strategies" contract AID/ta-CA-3, U.S. Agency for
International Development, Development Support Bureau, Office of Rural
Development and Development Administration. The paper is a revised version
of a paper originally included in the proceedings of "The CENTO Seminar on
Increasing the Productive Capacity of Small Farmers," Lahore, Pakistan,
17-21 December, 1978.
MSU is an Affirmative Action/Equal Opportunity Institution.
Over the past few years Farming Systems Research (FSR) has gained a
"ground swell" of support among donors and international research institutes.
The thrust of FSR is to carry out farm level research through a farmer/
researcher partnership in problem identification and farm level testing of
improved technologies. The proponents of FSR contend that the farmer/
researcher partnership is needed because much of the "top down" research in
experiment stations has not given sufficient attention to the relevance of a
technology in terms of the goals and resources of small farmers.
FSR starts with the premise that there are substantial institutional,
political, informational, and attitudinal barriers which inhibit the "voice"
of small farmers in shaping research priorities in commodity and disciplinary
research programs in the Third World. Farming Systems Research is advocated
as a more immediate and systematic way to provide a voice and research assis-
tance to small farmers. But while FSR can theoretically give "voice" to
small farmers it does not follow that small farmers will have political power
to press for the reform of other institutions and policies which limit their
access to extension, credit and reliable markets.
There is a great deal of confusion over whether FSR is a philosophy of
research (farmer/researcher partnership) or whether it is unique and differ-
ent from commodity and disciplinary research. If it is unique and different
then FSR units should be established within research institutes, FSR projects
prepared for donors, and special training programs established to train FSR
researchers. Many observers question whether FSR is new or simply farm
management research under a new label. Finally there remains the question
of why it has been so difficult to move FSR beyond the International Agri-
cultural Research Centers and into national research systems.
To understand these issues it will be helpful to examine FSR within a
historical perspective. First, on the question of whether FSR is the same
as farm management research, we note that in the 1920s and 1930s farm
management research in the United States emerged, and emphasis was placed on
a holistic approach to examine alternative farm enterprises. At that time
farms were primarily subsistence in orientation and the rural household was
engaged in a wide range of production and consumption activities. The
record shows that farm level research was underway in the United States by
Tennessee Valley Authority (TVA) researchers in the 1930s. TVA researchers
carried out a wide range of fertilizer trials on farmers' fields throughout
the Southeastern states. The "balanced farming" program in Missouri in the
1940s is another example of farm level research. If FSR is defined to mean
an active farmer/researcher partnership, then FSR is not a new approach.
But most current FSR in the Third World is focused on only one enterprise--
crops. Much work remains to be done to integrate research on cropping and
livestock systems in the developing world before FSR can claim to be as
broad as farm management research.
Second, on the issue of the uniqueness of the FSR approach, there is a
striking parallel between the conflicting views on Community Development (CD)
and agricultural development in the Third World in the 1950s, and whether FSR is
a philosophy of research which can be carried out by commodity research programs
or a unique system which requires separate FSR units and specially trained farming
systems researchers. In the 1950s many experts argued that community develop-
ment was a "philosophy" of helping farmers and rural people to express their
"felt needs", to increase their participation in the development process,
etc., and that this philosophy could be incorporated into ongoing agricultural
programs. Other experts argued that CD was unique and different from agri-
cultural development and that a Ministry of Community Development should be
established, along with CD projects, for donors to fund. Another issue was
whether CD agents were needed in addition to agricultural extension agents.
The Ministries of Agriculture and Community Development fought over these
issues in many countries in the 1950s. However, as Akhter Hameed Khan1 and
Lane Holdcroft2 have pointed out in earlier papers in this series, after a
few years the Ministry of Agriculture invariably won the battle. Currently
1Akhter Hameed Khan, "Ten Decades of Rural Development: Lessons from
India," MSU Rural Development Paper No. 1, 1978.
2Lane E. Holdcroft, "The Rise and Fall of Community Development in
Developing Countries, 1950-1965: A Critical Analysis and an Annotated
Bibliography," MSU Rural Development Paper No. 2, 1978.
many people argue that FSR should be given a separate identity, just as
CD was given in the 1950s. Others argue that FSR is a philosophy of
research which can be carried out within established research programs.
For example, although CIMMYT does not have a FSR program, substantial
farming systems research (farm level research) is being carried out as an
integral part of CIMMYT's three major research programs--wheat, maize and
economics programs. The debate over whether FSR is a philosophy of research
or a new approach requiring separate organizational arrangements will con-
tinue in the 1980s.
The historical insights suggest a need for not only debate and dialogue
on FSR, but also the publication of papers on the FSR experience in different
regions of the world. We are pleased to announce that four papers on FSR are
being prepared for our MSU Rural Development Paper series. These papers are
being published as part of an AID financed contract "Alternative Rural
Development Strategies" with Michigan State University; some will be trans-
lated into French and Spanish. These papers are designed to provide insights
from FSR experiences which can be used by researchers, policymakers, rural
development practitioners and donors in designing research and action pro-
grams relevant to small farmers.
The author of this FSR paper--Dr. David Norman--is writing from first-
hand experience in helping organize and carry out multi-disciplinary research
on problems of small farmers in Nigeria for 11 years (1965-1976).
Dr. Norman's analysis is an honest assessment of the strengths and limitations
of FSR. He rightly points out that FSR is neither a panacea nor a substitute
for, but a complement to strong commodity and disciplinary research programs.
The second paper in our FSR series is a state of the arts paper by
Elon Gilbert, David Norman and Fred Winch. This May 1980 paper will provide
a worldwide assessment of FSR research, including examples from international
and national research systems in the Third World. The third paper is an
annotated bibliography of FSR by Doyle Baker of MSU. Since there is consid-
erable regional variation in FSR approaches we shall publish several papers
on farming systems research in different regions of the Third World. The
fourth paper, by Mike Collinson, will assess CIMMYT's multi-disciplinary
farming systems research program in Eastern Africa. Regional papers will be
published for Central America and Southeast Asia.
Carl K. Eicher, Director
Alternative Rural Development
TABLE OF CONTENTS
The Farming Systems Approach: Relevancy for the Small
Evolution in Thinking About Agriculture and Technology
Defining a Farm System . . . . . . .
Common Elements in Farming Systems Research (FSR) .
FSR Gives "Voice" to Small Farmers . . . . .
Empirical Results of Farming Systems Research . .
Sole Versus Crop Mixtures . . . . . .
Traditional Versus Improved Cotton . . . .
Increasing Cropping Intensity . . . . .
Increasing Small Farmer Income in Guatemala . .
Implementing Farmer Based Farming Systems Research .
Creating the Proper Working Environment . . .
Implementation Problems . . . . . .
Conclusion . . . . . . . . . .
References . . . . . . . . . .
. . . .
LIST OF TABLES
1 Comparison of Sole and Mixed Crops Grown by Farmers on
Rainfed Land in Three Areas of Northern Nigeria,
1966-68 . . . . . . . . . . . . . 11
2 Comparison of Results of Traditional and Improved Technology
for Sole Crop Cotton, Daudawa Village, Northern Nigeria,
1971-1974 . . . . . . . . . . . . 14
3 Percentage of Rainfed Cropland in Various Cropping Patterns,
Iloilo Outreach Site, Philippines, 1974-1979 . . . 19
LIST OF FIGURES
1 Schematic Representation of Some Determinants of the
Farming System . . . . . . . ... . . . 4
2 Schematic Framework for Farming Systems Research at the
Farm Level (Downstream Farming Systems Research) . . 7
3 Monthly Labor Distribution of Total Annual Labor Require-
ment for Cotton, Daudawa Village, Northern Nigeria,
1971-1974 . . . . . . . . . . 16
THE FARMING SYSTEMS APPROACH:
RELEVANCY FOR THE SMALL FARMER
Two of the most common development slogans of the 1980s are "basic
human needs" and "growth with equity". Both describe strategies for help-
ing specific groups--especially small farmers in less developed countries
(LDC's). These strategies imply that instead of taking a welfare approach,
development programs should help the poor to increase their income earning
opportunities, and crucial to creating such opportunities is providing
small farmers with relevant and improved technology to meet their needs.
It is from the quest for relevant technology that the farming systems
research (FSR) approach has emerged.
The aims of this paper are to:
(a) Briefly review the evolution in thinking about agriculture
and technology development in the LDC's;
(b) Define a farming system and the general characteristics of
the FSR approach;
(c) Discuss the role of the FSR approach in designing and imple-
menting projects to help small farmers;
(d) Discuss some of the problems of implementing a FSR program.
EVOLUTION IN THINKING ABOUT AGRICULTURE
AND TECHNOLOGY DEVELOPMENT
People originating from or trained in high-income countries have dis-
played strong biases, sometimes unknowingly, about how to achieve develop-
ment in the LDC's. Over the last two or three decades thinking has evolved
through four successive strategies: (a) taxing agriculture to finance
industrial/urban development; (b) transferring technology from the high-
income countries to the LDC's; (c) developing technology within the LDC's
by drawing on elements of technological packages in high-income countries;
and recently, (d) supplementing the selective importation of technology with
a "bottom-up" approach to technology development, or what is now commonly
called the farming systems approach.
1"Bottom up" refers to the strategy of starting the research process at
the farmers' level by first ascertaining their needs, and then using these
needs to determine research priorities. This contrasts with earlier "top-
A number of factors explain the move to farming systems research and
local technology development. First, previous strategies to improve the
livelihood of small farmers have repeatedly failed. Second, many agricul-
tural programs have led to an unequal distribution of benefits. While the
success of the Green Revolution should not be underestimated, numerous
equity problems arose in the process of increasing agricultural production
(Saint and Coward, 1977). Despite claims that Green Revolution technologies
were intrinsically neutral to scale, for instance, many small farmers and
the landless found it difficult to gain access to land and the technological
packages (Khan, 1978; Poleman and Freebairn, 1973). A third reason for the
shift to farming systems research has been the rising cost of fossil energy
which is embodied in much of the Green Revolution technology. The fourth
reason is the increased realization, supported by empirical evidence, that
many traditional practices used by small farmers for generations are sound
and should be preserved (Jodha, 1978; Navarro, 1977). These and other
factors have contributed to the emergence of a "bottom-up" or farming
systems approach to the development of small farmer technology. The FSR
approach, however, is not easy to define.
DEFINING A FARMING SYSTEM
A system can be defined conceptually as any set of elements or com-
ponents that are interrelated and interact among themselves. Thus, a farming
system is the result of a complex interaction of a number of interdependent
components. At the center of this interaction is the farmer himself; he is
the central figure in FSR. Moreover, both farm production and household
decisions of small farmers are intimately linked and should be analyzed in
farming systems research. A specific farming system arises from the decisions
taken by a small farmer or farming family with respect to allocating different
quantities and qualities of land, labor, capital, and management to crop,
livestock, and off-farm enterprises in a manner which, given the knowledge
the household possesses, will maximize the attainment of the family goalss.
down" approaches where research priorities, determined at the experiment
station level, are transmitted "down" to the farmers, who are not directly
consulted in the research process.
Figure 1 illustrates some of the underlying determinants of the farming
system. The total environment can be divided into two elements: technical
and human (Institut d'Economie Rurale, 1976). The technical element deter-
mines the types and physical potential of livestock and crop enterprises,
and includes physical and biological factors that have been modified to some
extent by man--often through technology development. Man has developed, for
example, mechanical techniques to improve the availability of water through
irrigation, and chemical techniques to improve soil quality, etc. The farm-
ing system that actually evolves, however, is a subset of what is potentially
possible as defined by the technical element.
The human element is characterized by two types of factors: exogenous
and endogenous. Exogenous factors (i.e., the social environment), which are
largely outside the control of the individual farmer, influence what he will
and/or is able to do. They can be divided into three broad groups:
(a) Community structures, norms and beliefs.
(b) External institutions. These can be subdivided into two main
groups: inputs and outputs. On the input side, extension,
credit and input distribution systems are often financed and
managed by government agencies. On the output side, the
government may directly (e.g., marketing boards) or indirectly
(e.g., improved evacuation routes, transportation systems, etc.)
influence the prices farmers receive.
(c) Miscellaneous influences, such as population density and location.
Unlike the exogenous factors, the endogenous factors are controlled by
the farmer himself, who ultimately decides on the farming system that will
emerge, given the constraints imposed by the technical element and exogenous
The farming system as defined above highlights the complex nature of the
underlying determinants. An appreciation of these determinants can provide
insights as to why small farmers have failed to adopt improved technology.
Specifically, most conventional approaches to technology development,
utilizing a "top-down" approach, tend to modify the technical element to
fit crops or animals and to ignore the human element. The farming systems
approach, on the other hand, potentially imparts greater reality to tech-
nology development by making technology a variable instead of a parameter
Broken lines represent results of farming system.
I I Mechanical
Farming System I
( _ _ _I
Figure 1 Schematic Representation of Some Determinants of the Farming System
(Saint and Coward, 1977). FSR increases the potential for fitting the
animal or crop to the environment rather than vice-versa (Van Schilfgaard,
COMMON ELEMENTS IN FARMING SYSTEMS RESEARCH (FSR)
FSR recognizes and focuses on the interdependencies and interrelation-
ships between the technical and human elements in the farming system. As
such it is more holistic in orientation than the reductionist approach
traditionally used by technical agricultural scientists--an approach that
requires studying one or two factors at a time while attempting to control
all others (Dillon, 1976). The primary aim of the FSR approach is to in-
crease the overall efficiency of the farming system; this can be interpreted
as developing technology that increases productivity in a way that is use-
ful and acceptable to the farming family, given its goalss, resources and
Research on farming systems in the LDC's has developed mainly in the
last decade and is now being pursued in Africa, Asia, and Latin America at
national institutes (e.g., ISRA in Senegal, ICTA in Guatemala, etc.), regional
institutes (e.g., CATIE in Costa Rica, GERDAT in France--which serves
Francophone countries in Africa, etc.), and international institutes (e.g.,
IRRI, ICRISAT, ICARDA, IITA, CIAT, etc.)
There are two basic types of FSR programs--"upstream" and "downstream"
(Technical Advisory Committee, 1978). Upstream FSR uses research from
experiment stations to find prototype solutions to the major constraints
on agricultural improvement in a relatively large region or area (e.g.,
the semi-arid tropics). Downstream farming systems research is a farm
level research approach whereby farmers and a multi-disciplinary research
team work together to diagnose, design, modify and improve farming systems
in a local area. Downstream FSR uses information from upstream FSR, experi-
ment stations and commodity research programs in order to design improve-
1The use of the FSR approach for improving rural development strategies
is currently being tested by Michigan State University in the Eastern ORD of
ments in particular farming systems, but only after the constraints of
that system have been analyzed. The schematic framework in Figure 2 out-
lines the downstream or farm level FSR approach at various institutions.
Upstream and downstream farming systems research are mutually supporting
and reinforcing. Also, FSR is a complement to rather than a substitute for
commodity research programs.
The following should be taken into account in carrying out farming
(a) There are four successive research stages: descriptive
(diagnostic), design, testing and extension.1 The de-
scriptive stage identifies the constraints and flexibility
in the current farming systems. Based on interviews with
farmers, this information is used to design, test and
extend programs for improving farming systems. These pro-
grams are then assessed by applying evaluation criteria
derived from farmer interviews.
(b) Because the farm household is central to the research pro-
cess, the farming system research approach could aptly be
called farmer's system research (Gotsch, 1977). This con-
cept emphasizes the important role that a farmer's know-
ledge, derived from experience (Swift, 1978) and traditional
experimentation (Johnson, 1972; Jodha et al., 1977; Vermeer,
1979), can play in improving his farming system. Moreover,
the farmer's involvement in the research process increases
the possibility that improved systems will address farm level
problems. Ultimately, a new system arises which combines
the best of the system he already uses with the results of
the research process (Harwood and Price, 1976). Thus, many
changes envisioned in FSR involve small adjustments rather
that complete changes in the system. The role of the farmer
is maximized and reality in the research process is ensured
These stages were specified at a seminar in Mali on improved agricul-
tural production systems (Institut d'Economie Rurale, 1976). The stages will
be discussed in more detail in Gilbert, Norman and Winch, (1980).
1. Description or
diagnosis of present
2. Design of improved
3. Testing of improved
4. Extension of improved
Experiment Station Trials ----------------> BODY OF
--------------- -------------------------- >
Trials at Farm Level --------------------->
Farmers' Testing --------------
MODIFIED FARMING SYSTEM '< -----------------------
Figure 2 SCHEMATIC FRAMEWORK FOR FARMING SYSTEMS RESEARCH AT THE FARM LEVEL
(Downstream Farming Systems Research)
by minimizing research on experiment station fields and maxi-
mizing it on the farmer's fields. Initially, the researcher
manages field trials (Figure 2); later the farmer provides
this input through farmer testing.1 And the value of on-
farm research can also be enhanced by involving extension
(c) A multidisciplinary team is required to understand the inter-
action of the technical and human elements. With the aid of
a social scientist who plays an ex ante rather than the tra-
ditional ex post role, this team would work in an interdisci-
plinary manner at the first three stages of the research process
and, possibly, at the fourth.
(d) There is recognition of the locational specificity or heterogen-
eity of the technical, exogenous, and endogenous factors. Dis-
aggregating such heterogeneity into homogenous subgroups and
developing improved technologies appropriate to each are central
to the farming systems research approach. The disaggregation is
done first in terms of ecological systems or differences in the
technical element and then, if further disaggregation is neces-
sary, in terms of differences in the human element.2 Its pur-
pose is to maximize the variance between farm systems in the
subgroups and minimize the variance within subgroups; the goal
is to produce useful classifications for developing relevant
improved technologies and for implementing programs. The most
limiting constraints found in the farming systems of each sub-
group then become the focal point for developing technologies
either to overcome them or to avoid them by exploiting the flexi-
bility that exists in the current farming systems. The pro-
On-farm research involves an analysis of the actual system instead of
simply attempting to simulate actual conditions through models such as unit
farms, linear programming, simulation, etc. (Technical Advisory Committee,
2For example, subgroups could be disaggregated by ethnic origin, differ-
ing access to the external institutions, size of farm, land per worker ratio,
posed technologies must, however, be compatible with the exo-
(e) In evaluating a farming system, researchers must understand the
multi-utilization of resources and the rural household as a
production and consumption unit, in order to ensure that evalu-
ation criteria will be relevant to the rural household. Returns
per man-hour of labor, for instance, may replace the traditional
net return per unit of land in land surplus economies.2
(f) The research process is recognized as dynamic and interactive
and emphasizes linkages between the farmer and research worker.
(g) The FSR approach provides a feedback mechanism for shaping
priorities for basic and commodity research programs.
FSR GIVES "VOICE" TO SMALL FARMERS
The priorities of public-financed agricultural research are often
based on: (a) expressed needs of more influential farmers, who also may
hold influential non-agricultural jobs; (b) types of research which appeal
to professional "peer groups"; or (c) the types of technology that have been
developed and adopted in high income countries. In contrast, the FSR
approach gives the small farmer, often for the first time, a "voice" in
tailoring research priorities, both in technology development and evalu-
ation, to his needs. The small farmer becomes the central figure in the
In developing strategies to overcome the most limiting factor or factors,
new technology may not always be necessary; other approaches might be appro-
priate, such as group action in irrigation, (Binswanger and Ryan, 1977). How-
ever, as mentioned in footnote 1, page 5, the FSR approach is currently being
tested and has not yet established itself in solving such problems.
2The value of a proposed technology will be determined by whether or not
it satisfies the relevant evaluation criteria. In general, all proposed tech-
nologies must be compatible with the technical element and with exogenous
factors (e.g., community structures, norms, beliefs, external institutions
such as extension, credit and input distribution system, and markets for
products produced, etc.) However, the technical feasibility and social
acceptability that this implies is not sufficient. Specific evaluation
criteria relating to endogenous factors will tend to be farmer and farming
system specific. In general terms, it must be economically feasible, de-
pendable and compatible with the farming system used by the farmer (Norman
and Hays, 1979).
research process, particularly at the descriptive and testing stages when
dialogue with him is so important. In LDC's to date there has been little
communication between the small farmer and the researcher. Ideally communi-
cation should be possible via the extension worker, but for a number of
reasons this has not often worked.
Small farmers draw on traditional skills and experience in shaping
their farming systems. By ignoring these skills, researchers in experi-
ment stations have often cut themselves off from valuable sources of know-
ledge and wisdom. As a result, considerable time is spent in experiment
stations in "rediscovering the wheel" rather than building on knowledge
the small farmer already possesses. For example, for many years agricul-
tural scientists, and even officials in ministries of agriculture in LDC's,
regarded the traditional practice of growing crops in mixtures as "primi-
tive" and not compatible with "modern" agriculture. Hence mixed cropping
was not considered worthy of serious research endeavor. Yet farmers
resisted growing crops in sole stands and as the next section indicates,
their reluctance is understandable.
EMPIRICAL RESULTS OF FARMING SYSTEMS RESEARCH
FSR has already contributed to the development of improved technolog-
ical packages for small farmers.
Sole Versus Crop Mixtures
In many parts of the Third World researchers and extension workers
have often failed in their attempts to encourage farmers to plant improved
crop varieties in sole stands. Why? The results of research on farming
systems in northern Nigeria help answer this question (Norman, Pryor and
Gibbs, 1979). In this region agriculture is primarily rainfed, with
rainfall varying in the areas studied (Table 1) from 752 to 1102 mms. There
is marked seasonality in rainfall distribution and the growing season ranges
from 150 to 190 days. Hand cultivation systems are the rule. There is also
a marked seasonality in the agricultural cycle, with labor demands peaking
COMPARISON OF SOLE AND MIXED CROPS GROWN BY
FARMERS ON RAINFED LAND IN THREE AREAS
OF NORTHERN NIGERIA, 1966-68a
Annual Rainfall (mm)
Days in Growing Season
Percent Cultivated Area
Devoted to Sole Crops
Average percent change
Sole or Crop Mixture Sole Mixture Sole Mixture Sole Mixture from sole to crop
Annual b 425.8 485.6 362.3 586.4 564.9 597.5 27.2
Labor peak period 232.5 237.9 122.3 157.9 247.3 247.3 10.5
Millet 736.4 686.0 -- 366.5 727.4 393.4 -26.4
Sorghum 652.3 122.2 785.7 644.5 839.5 728.6 -37.5
Groundnuts 429.3 188.3 587.3 412.5 392.3 217.5 -43.5
Cowpeas -- 56.0 -- 132.3 -- 51.6 --
Value of production:
(N)/hac 31.65 40.80 37.96 61.36 29.50 33.73 34.9
Annual man-hour 0.06 0.12 0.13 0.11 0.08 0.08 28.2
Man-hour put in
during peak period 0.13 0.32 0.35 0.42 0.24 0.25 56.8
Net return (N/ha):
Labor not valued 30.74 38.94 36.79 59.48 30.74 35.76 34.9
Costing hired labor 28.27 36.13 33.41 54.02 28.64 31.18 32.8
All labor costed 17.96 24.36 18.31 29.60 14.80 18.68 41.2
aThe weighting system
size of farmers = 330.
used to derive the figures used for comparison in the table is discussed elsewhere (Norman 1974). Sample
bPeak periods were: June August in Sokoto; June and July in Zaria; and July September in Bauchi.
COne Naira (N) = $1.50.
during weeding between June and September. In many areas in northern
Nigeria a seasonal labor shortage, rather than land, is the major con-
straint on expanding farm output. Farmers have traditionally grown crops
in mixtures of two or more crops together on the same field.
The following results (Table 1) were obtained by comparing sole stands
and the more common crop mixtures on farms.
(a) The annual labor input per hectare from crops grown in mixtures
was 27% higher than crops grown in sole stands. However, this
differential was reduced to 10% when only labor committed during
the peak farming period was considered. Labor is truly limiting
only at this time.
(b) On farms where crops were grown in mixtures the average yield
of individual crops was 26% to 43% lower than yields of sole
(c) When the yields of individual crops were expressed in monetary
terms, however, the average value per hectare of crop mixtures
was 35% higher than sole stands, indicating that the reduced
yield of some crops in crop mixtures was more than offset by
yields of other crops in the mixture.
(d) The return from crops grown in mixtures per annual man-hour was
28% higher than from growing crops in sole stands. This return
was even greater when labor applied during the bottleneck period
of weeding was considered separately: the average increase in
return per man-hour was 57% higher for crop mixtures. It appears,
therefore, that mixed cropping helps alleviate the labor bottle-
neck problem. Linear programming studies provide additional
empirical support for mixed cropping (Ogunfowora and Norman,
(e) The level of profitability or net return per hectare also reveals
the superiority of crop mixtures over sole crops; it ranged from
32% to 41% higher, depending on how labor was costed. Finally,
results given elsewhere (Norman, 1974) indicate that growing crops
in mixtures gave a more dependable return, which is very important
to farmers pursuing risk aversion strategies.
These research results have demonstrated that mixed cropping in tra-
ditional farming systems in northern Nigeria is compatible with both the
technical and human elements. Hence, it is not surprising that farmers
in northern Nigeria have been reluctant to follow the advice of technical
researchers and change to sole crops. Mixed cropping is a rational strategy
for farm families faced with a land or labor constraint and high risk
associated with uncertain weather.
In the last few years technical scientists in Nigeria and other coun-
tries have expressed considerable interest in developing improved tech-
nology for mixed cropping. The FSR approach can be helpful in applying the
total sum of knowledge about agriculture, including the practices of tra-
ditional farming, in developing relevant and improved technology.
Traditional Versus Improved Cotton
Another example from northern Nigeria illustrates the potential advan-
tages of the FSR approach when developing improved technology for small
cotton farmers. Traditionally, cotton is planted after food crops have
been planted and partially weeded. Researchers at Ahmadu Bello University
developed an improved cotton technology package with emphasis on higher
yields. This package required not only planting earlier and in sole stands,
but also the application of fertilizer and six sprayings with a knapsack
sprayer that used 225 litres of water per spray per hectare. The cotton
package was developed in the experiment station and, in retrospect, over-
looked the human element of small farmers.
The following conclusions were derived from an ex post farming systems
survey of farmers who used the improved cotton practices over a four year
period (Beeden, et al., 1976):
(a) The results in Table 2 indicate that the net return per hectare
of improved cotton was considerably higher, except in the
drought year (1973), than the returns from traditional cotton.
(b) Yet even though the net returns per hectare of improved cotton
were higher, virtually no farmers adopted the improved cotton
recommendations in their entirety. Reasons were numerous.
COMPARISON OF TRADITIONAL AND IMPROVED TECHNOLOGY
FOR SOLE CROP COTTON, DAUDAWA VILLAGE,
NORTHERN NIGERIA, 1971-1974a
Net Return (N)b
Per ha Annual June/July
31.89 0.07 0.20
48.87 0.09 0.09
28.03 0.07 0.07
89.96 0.19 0.19
49.69 0.11 0.14
Percent Change From Traditional to Improved Cotton
Net Return (N)b
Man-hours/ha Per man-hour
(kg/ha) Annual June/July Per ha Annual June/July
177.5 50.5 101.4 90.6 40.0 4.8
124.1 64.2 124.2 94.0 28.6 -13.9
44.9 22.1 103.3 -13.4 -22.3 -57.4
115.4 100.9 71.9 111.3 5.5 22.7
115.5 59.4 100.2 70.6 13.0 -13.3
aSome operations were undertaken using oxen.
bOne Naira (N) = $1.50.
First, the results in Table 2 indicate that the average labor
inputs required for growing improved cotton were 59% higher
than those for producing traditional cotton. Although the
higher yields compensated for higher labor costs and increased
the annual return in Naira per total man-hour by 13%, the
return per man-hour during the June-July labor bottleneck
was 13% less for improved cotton than for the traditional cotton.
Second, because the improved cotton had to be planted earlier
than traditional cotton (Figure 3), a labor conflict emerged
and the farmer had to choose between weeding his food crops or
planting the improved cotton. Whereas the cotton researchers
had compared traditional and improved cotton yields on research
plots, the farmers had analyzed improved cotton as part of their
total farming system. The ex post farming systems research
revealed that the farmers had not compared improved cotton tech-
nology with the traditional cotton technology but instead with
labor requirements for food crops. Thus, one of the major reasons
for rejecting improved cotton was the incompatibility of the new
technology with endogenous factors such as family labor bottle-
necks and labor availability for food production.
(c) Other reasons for non-adoption related to the difficulty of trans-
porting large amounts of water required for spraying and the lack
of adequate extension, fertilizer, etc.
In the light of the above results, obtained through ex post farming
systems research, it was recommended that plant breeders develop cotton
varieties which could be planted later. Even though yields would be po-
tentially lower, the later varieties could be accommodated in a farming
system which gives first priority to family food production and second pri-
ority to cash crops such as cotton. Also recommended was the replacement
of a water-based insecticide with an oil-based one; this would be applied
with an ultra low-volume sprayer and its use would decrease labor inputs
for carrying water. If a FSR approach had been applied much earlier in
IRecently these deficiencies have been largely overcome as a result
of an IBRD integrated agricultural development project in the area.
* * Improved practices
- Traditional practices
*---- Composite farm operations
251 (b) 1972
Figure 3 Monthly Labor Distribution of Total Annual Labor Requirement for Cotton,
Daudawa Village, Northern Nigeria, 1971-1974
the development of improved cotton packages, the efficiency and the credi-
bility of the research process would no doubt have been increased.
Increasing Cropping Intensity1
The development of high-yielding and early-maturing rice varieties
has opened up possibilities for increased cropping intensity in land-
scarce Southeast Asia. The Cropping Systems Program at the International
Rice Research Institute has focused on rice-based cropping systems of small
farmers located in rainfed areas. Drawing on the knowledge that exists
for facilitating the introduction of additional crops in a crop sequence,
researchers are able to select techniques appropriate to the agro-climatic
and socio-economic conditions prevalent in areas under investigation.
Growing seasons of intermediate length2 are potentially adaptable for
intensified cropping systems. In such production situations, researchers
seek to lengthen the effective growing season by a variety of methods,
alone or in combination. These include (Zandstra and Carangal, 1978):
(a) The use of shorter duration varieties,
(b) The use of techniques which allow earlier planting at the
beginning of the rainy season,
(c) The overlapping of growing periods by relay cropping and
(d) The extension of the growing season into the dry season,
by using drought-tolerant crops,
I wish to thank Jim Chapman of Michigan State University for this
2A qrowinq season of intermediate length is characterized by 5-6
months of rainfall above 200 mm per month plus at least 3 months with
rainfall between 100 and 200 mm.
(e) Improving soil moisture utilization,
(f) The use of supplementary irrigation.
An important example of these methods in application is the use of
new short duration varieties in combination with direct planting techniques.
When IRRI established a Cropping Systems Outreach Site in Iloilo, Philip-
pines in 1975, 82 percent of the rainfed land was planted to a rice-fallow
pattern (Table 3). By modifying their existing farm systems to incorporate
new technologies, Iloilo farmers have been able to plant upland crops be-
fore or after rice and, in lower lying areas, harvest two rice crops in a
single season. In the years since 1975, when rainfall patterns are normal,
farmers have planted two or more crops on roughly 75 percent of their crop-
land. For example, Table 3 shows that in 1978-79 farmers planted two or
more crops as follows: two or more rice crops, 24 percent; one rice and
one or more upland crops, 40 percent; two or more upland crops, 11 percent;
or a total of 75 percent. Even in years of low rainfall, such as occurred
in 1977-78, cropping intensity still greatly increased over previous levels.
This example illustrates how cropping systems research has led to a rapid
increase in cropping intensity in the short span of 4 years.
Increasing Small Farmer Income in Guatemala
An example in eastern Guatemala (Hildebrand 1977) demonstrates the
potential of the FSR approach for improving small farmer productivity and
income. Here, farming systems research revealed that the two controllable
factors most responsible for limiting traditional farm production on the
steep hillsides were the short planting season and the limited amounts of
bean seed available for planting. Traditionally corn, beans, and sorghum
were planted simultaneously, and land was not a limiting factor for most
farmers in the area. Research indicated that if farmers planted twin or
double rows of corn and sorghum and concurrently planted fewer beans
(which require the most time to plant), labor productivity would increase
because each farmer could plant more land than under the traditional cropping
system. That is, with the same amount of planting labor and somewhat less
bean seed farmers could plant 40% more land, produce 75% more corn, 40%
PERCENTAGE OF RAINFED CROPLAND IN VARIOUS
CROPPING PATTERNS, ILOILO OUTREACH
SITE, PHILIPPINES, 1974-79a
1974- 1975- 1976- 1977- 1978-
Cropping Pattern 1975 1976 1977 1978 1979
Two or more rice crops 5 20 27 16 24
One rice + 1 or more upland 11 28 47 31 40
Two or more upland 2 5 4 10 11
One rice + fallow 82 47 19 41 22
One upland + fallow 3 2 3
Source: Derived from Genesila, Servano and Price, 1979.
aThe 1974-75 data represent average results of a 205 farm baseline survey conducted in January
1975. Data from 1975-79 came from a farm record-keeping study of 45 farmers selected randomly from
the baseline list.
more sorghum, and the same quantity of beans. Farmers received 33%
more income from the revised planting system.
IMPLEMENTING FARM LEVEL FARMING SYSTEMS RESEARCH
While the FSR program approach may lead to a more efficient develop-
ment of improved technology for the small farmer, numerous problems exist
in mobilizing multidisciplinary teams and in implementing the research.
Creating the Proper Working Environment
An efficient FSR program often requires substantial changes in both
administrative arrangements and philosophy in agricultural research
institutions. For example:
(a) Research programs in agricultural institutes are usually
organized along disciplinary lines. Some institutes have
moved recently to commodity based research programs, but a
FSR program will require changing to an even more holistic
approach. Attempts to do this may be frustrated by two
(i) The farming system approach requires the integration
of livestock and crop production. Research on livestock
and crops, however, is often undertaken by different
institutions, making integration virtually impossible.
(ii) A similar problem exists for social scientists (e.g.,
agricultural economists and sociologists) who are often
located in academic institutions which are separate
from government agricultural research institutes.
(b) The FSR approach requires a fundamental change in the philosophy
and research approach of scientists. The new dimensions are as
(1) The FSR approach starts at the farm level (descriptive stage)
and moves to the experiment station (design stage) and then
back to the farm (testing and extension stages). This repre-
sents a major change for scientists whose traditional
work on the experiment station was only supplemented,
perhaps, by research managed trials at the farm level
(Figure 2). The inevitable loss of controlled factors
in the experimental process (i.e., diminuation of
ceteris paribus conditions) can be a frustrating
(ii) The research worker needs to interact with the farmer,
the extension worker and the government agencies which
influence the external institutions (Figure 2). If his
work is to have relevance the researcher must listen and
take into account the comments of others when deciding
on an approach and research priorities. This will be a
fundamental change for some researchers and will require
them to be extremely sensitive to the needs of various
(c) Identifying individuals suitable for FSR programs may be a
problem. Much of the FSR is now undertaken by individuals
trained in and/or originating from high-income countries.
Their training has usually been discipline oriented and unin-
tentionally, perhaps, culturally biased. Hence, it is sometimes
difficult for such persons to appreciate and understand the
local wisdom and values, the complexities of a farmer-household
system, the role of non-economic variables, and the potentially
significant role to be played by rural sociologists or anthro-
pologists. Researchers must be able to fit in and interact
effectively with an interdisciplinary farming systems research
team. Currently those with many years of field experience are
acquiring such an appreciation--helped sometimes through short
courses at regional and international institutions. Less
experienced researchers should also be encouraged to pursue
research on farming systems and to place emphasis on building
local capacity in the less developed countries.
Even if a favorable working environment can be created with a FSR
program, there are a number of implementation problems:
(a) Presently there is no standard methodology for undertaking
FSR. Indeed, the term farming systems research is somewhat
of a misnomer. To date most FSR has been confined to crop
production processes. Yet even here methodologies for under-
taking such work need to be improved. Apart from pleas for
its desirability (Boer and Welsch, 1977), the FSR approach
has rarely been applied to livestock processes unless these
impinge directly on crop processes. There is a need to
develop a more holistic systems approach which goes beyond
agricultural and livestock production and includes the
marketing process and off-farm enterprises (Gilbert, Norman
and Winch, 1980).
(b) A time lag inevitably exists from the recognition of a problem
to the discovery of a relevant solution and its adoption by
farmers. FSR can be time consuming. A farming systems approach
is now quite rightly being advocated in places where applied
research is not well established and relevant. Funding agencies,
however, must recognize that time is required to derive results
from FSR. Otherwise problems will arise in maintaining the con-
tinuity of the research. Also, FSR results may not be visually
spectacular, even though they may be large in the aggregate.
Time between recommended solutions and farm adoption might be
shortened if the link between FSR and extension is strengthened.
Representatives of extension agencies should be integral members
of the research team.
(c) Because of the location specificity of farming systems research, it
appears to be expensive to execute. Ways must be explored to make results
more widely applicable and thereby maximize the return from such
research. For example, technological packages need to be developed
which can be adopted by a large number of farmers, even through there
is some sacrifice of both yields and relevance to the better farmers.
The farming systems research approach is consistent with current notions
of equity, participation, and employment generation in rural economic develop-
ment. Because it is largely in the developmental stage, however, the FSR
process is not yet established as an efficient way to improve the livelihood
of small farmers.1 As soon as the problems mentioned have been overcome,
the FSR approach can be of considerable help to small farmers and can comple-
ment commodity and disciplinary research.
Because of the increased concern with "growth with equity" and the
increased willingness of agricultural research workers to shed some of their
professional, and sometimes cultural, arrogance, the future for helping small
farmers in LDC's is promising. To paraphrase the words of a wise Islamic
scholar, Alhaji Junaidu (1972), sound development must build upon rather than
destroy the farmers' traditional techniques.
For a survey of the state of the art of farming systems research in
the Third World see Gilbert, Norman, and Winch (1980).
Beeden, P., Norman, D.W., Pryor, D.H., Krocker, W.J., Hays, H.M., and
Huizinga, B. 1976. "The Feasibility of Improved Sole Crop Cotton
Production Technology for the Small Farmer in the Northern Guinea
Savanna Zone of Nigeria." Miscellaneous Paper No. 61. Samaru:
Institute for Agricultural Research, Ahmadu Bello University.
Binswanger, H.P. and Ryan, J.G. 1977. "Efficiency and Equity Issues in
Ex Ante Allocation of Research Resources." Indian Journal of Agri-
cultural Economics (July-Sept.).
Boer, A.J. de and Welsch, D.F. 1977. "Constraints on Cattle and Buffalo
Production in a Northeastern Thai Village." In Tradition and Dynamics
in Small-Farm Agriculture. pp. 115-148. Edited by R.D. Stevens.
Ames: Iowa State University.
Dillon, J.L. 1976. "The Economics of Systems Research." Agricultural
Genesila, M., Servano, R., and Price, E.C. 1979. "Socio-Economic Studies
in Iloilo, 1976-78." Paper presented at the Annual Cropping Systems
Workshop, March 1979, Los Banos, Philippines.
Gilbert, E.H., Norman, D.W., and Winch, Fred. "Farming Systems Research in
the Third World: A Critial Appraisal." Forthcoming as MSU Rural
Development Paper. East Lansing, Michigan: Michigan State University,
Gotsch, C.H. 1977. "The Concept of Farming Systems in the Analysis of
Agricultural Research and Development Programs." Paper presented at
the Middle East and Africa Agricultural Seminar, 1-3 February 1977,
Harwood, R.R. and Price, E.C. 1976. "Multiple Cropping in Tropical Asia."
ASA Special Publication No. 27. pp. 11-40. Madison: American Society
Hildebrand, P.E. 1977. "Generating Small Farm Technology: An Integrated
Multi-Disciplinary System." Paper presented at 12th West Indian
Agricultural Economics Conference, 24-30 April 1977, Antigua.
Institute d'Economie Rurale. 1976. "Rapport de Synthese sur les Systemes de
Culture et d'Elevaqe dans le Contexte de Mali." Bamako: Institut
International Rice Research Institute. 1977. "Constraints to High Yields
on Asian Rice Farms." Los Banos, Philippines: IRRI.
Jodha, N.S. 1978. "Production Patterns and Dynamics of Resource Use in
Arid and Semi-Arid Tropical India." Paper presented at -CILSS Seminar,
22 February-1 March 1978, Bamako, Mali.
Jodha, N.S., Asokan, M., and Ryan, J.G. 1977. "Village Study Methodology
and Resource Endowments of the Selected Villages in ICRISAT's Village
Level Studies." Occasional Paper 16. Hyderabad, India: ICRISAT.
Johnson, A.W. 1972. "Individuality and Experimentation in Traditional
Agriculture." Human Ecology, 1:149-159.
Junaidu, Alhaji, Wazirin Sokoto. 1972. "Muhimmancin Jam'a Ga Jama'armu."
Publication No. 1. Zaria, Nigeria: Ahmadu Bello University.
Khan, H.D. 1978. "Ten Decades of Rural Development: Lessons from India."
MSU Rural Development Paper No. 1. East Lansing, Michigan: Michigan
Navarro, L.A. 1977. "Dealing with Risk and Uncertainty on Crop Production:
A Lesson from Small Farmers." Paper presented at joint AAEA-WAEA
meeting, 31 July-3 August 1977, San Diego, CA.
Norman, D.W. 1974. "The Rationalisation of a Crop Mixture Strategy Adopted
by Farmers under Indigenous Conditions: The Example of Northern
Nigeria." Journal of Development Studies. 11:3-21.
Norman, D.W. and Hays, H.M. 1979. "Developing a Suitable Technology for
Small Farmers." National Development. April:67-75.
Norman, D.W., Pryor, D., and Gibbs, C.J.N. 1979. "Technical Change and
the Small Farmer in Hausaland, Northern Nigeria." African Rural
Economy Paper No. 21. East Lansing, Michigan: Michigan State Univer-
sity, Department of Agricultural Economics.
Ogunfowora, B. and Norman, D.W. 1973. "An Optimization Model for Evaluating
the Stability of Sole Cropping and Intercropping Systems Under Changing
Resource and Technological Levels." Bulletin of Rural Economics and
Poleman, T.T. and Freebairn, D.K., eds. 1973. Food, Population and Employ-
ment. New York: Praeger.
Saint, W.S. and Coward, E.W. 1977. "Agriculture and Behavioral Science:
Emerging Orientation." Science. 197:733-737.
Swift, J. 1978. "Notes on Traditional Knowledge, Modern Knowledge and Rural
Development." Paper presented at IDS Workshop on the Use of Indigenous
Technical Knowledge in Rural Development, 13-14 April 1978, Brighton,
Technical Advisory Committee. 1978. Farming Systems Research at the Inter-
national Agricultural Research Centers. Washington, D.C.: TAC
Van Schilfgaarde, J. 1977. "Discussion:
to Increase World Food Production."
Enhancing Natural Resource Use
American Journal of Agricultural
Vermeer, D.E. 1979. "The Tradition of Experimentation in Swidden Agri-
culture Among the Tiv of Nigeria." In Applied Geography Conference,
SUNY, Binghamton. Edited by J.M. Frazier and B.L. Epstein. Vol.
21, pp. 244-257.
Zandstra, H.G. and Carangal, V.R. 1978. "Crop Intensification for the
Asian Rice Farmer." IRRI Cropping Systems Program. (Unpublished
MSU RURAL DEVELOPMENT PAPERS
RDP No. 1
RDP No. 2
RDP No. 3
RDP No. 4
RDP No. 5
David W. Norman, "The Farming Systems Approach:
the Small Farmer," 1980.
MSU RURAL DEVELOPMENT WORKING PAPERS
RDWP No. 1
RDWP No. 2*
RDWP No. 3
RDWP No. 4
RDWP No. 5
RDWP No. 6
RDWP No. 7
RDWP No. 8
Benedict Stavis, "Turning Point in China's Agricultural Policy,"
Kathryn M. Kolasa, "The Nutritional Situation in Sierra Leone,"
Benedict Stavis, "Agricultural Extension for Small Farmers,"
Steve Haggblade, Jacques Defay, and Bob Pitman, "Small Manu-
facturing and Repair Enterprises in Haiti: Survey Results,"
Peter Riley and Michael T. Weber, "Food and Agricultural
Marketing in Developing Countries: An Annotated Bibliography
of Doctoral Research in the Social Sciences, 1969-79," 1979.
Harold M. Riley and Michael T. Weber, "Marketing in Developing
Victor E. Smith, Sarah Lynch, William Whelan, John Strauss and
Doyle Baker, "Household Food Consumption in Rural Sierra
Omar Davies, Yacob Fisseha and Claremont Kirton, "The Small-
Scale Non-Farm Sector in Jamaica..Initial Survey Results,"
Single copies of the MSU Rural Development Papers and MSU Rural Develop-
ment Working Papers may be obtained free by writing to: MSU Rural Development
Program, Department of Agricultural Economics, 206 International Center,
Michigan State University, East Lansing, Michigan 48824; U.S.A.
*Out of print.
Akhter Hameed Khan, "Ten Decades of Rural Development:
Lessons from India," 1978.
Lane E. Holdcroft, "The Rise and Fall of Community Develop-
ment in Developing Countries, 1950-1965: A Critical Analysis
and an Annotated Bibliography," 1978.
James E. Kocher and Beverly Fleisher, "A Bibliography on Rural
Development in Tanzania," 1979.
Enyinna Chuta and Carl Liedholm, "Rural Non-Farm Employment:
A Review of the State of the Art," 1979.