Farming systems R&D methodology

Material Information

Farming systems R&D methodology
Alternate title:
Farming systems research and development methodology
McDermott, J. K ( James Kenneth ), 1922-
United States -- Agency for International Development. -- Office of Agriculture
Place of Publication:
Washington, D.C.
Office of Agriculture, Technical Assistance Bureau, United States Department of Agriculture
Publication Date:
Copyright Date:
Physical Description:
iii, 41 leaves : maps ; 28 cm.


Subjects / Keywords:
Agricultural systems -- Research ( lcsh )
Agricultural extension work -- Research ( lcsh )
Farming systems ( jstor )
Farms ( jstor )
Developing countries ( jstor )
bibliography ( marcgt )
non-fiction ( marcgt )


General Note:
"May 17, 1977."
General Note:
At head of title: "Project Paper."
General Note:
Statement of Responsibility:
McDermott ; prepared by Office of Agriculture, Tehnical Assistance Bureau.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact Digital Services ( with any additional information they can provide.
Resource Identifier:
436460665 ( OCLC )


This item has the following downloads:

Full Text

"rroj ectizaperf'



Office of Agriculture
Technical Assistance .Bureau
*Ma L7, 1977

Project Paper


Page N.o.

Part 1: Summary and Recommendations . . *

A. Face Sheet
PID Face Sheet

B. Summary Description of Project

C. Recommendations
Environmental Determination

D. Summary of Findings

Part 2: Background and Detailed Description . 3

A. Background

The Problem
The Rationale
Definitions and Concepts
State of the Art

B. Detailed Description of Project

Substantive Output

Part 3: Project Analysis . . . . . . . 16

A. Review, Technical
Literature Review

B. Conceptualization and Methodological Notes

Methodologies for Evaluating Farming Systems
Linear Programming
Budgeting and Cash Flow Analysis

Transect Methodology
Adaptation of the Methodology
Selected Bibliography

Page No.

C. Project Design and Work Plan

Specific Objectives
Plan of Work
Project Activity Schedule, Phase 1
Provisional Activity Schedule, Phase 2

D. Environmental Impact

E. Role of Women

F. Estimated Budget

Part 4: Implementation ... . . . . . .40

Contractor Coordination
Test Site Selection
Selecting Contractors
Relationship with Other Projects


1. Project Matrix

2. Development and Application of Systematic
Methodology for Farming Systems Research

Appendix 6A, Ch 6
Page 4 TM 3:1)

a. Symbol b. Code
TA I FY1812 1I

Farming System R&D Methodology J a. INITIAL I 8|177 b. FINAL FY 18 I

9. ESTIMATED TOTAL COST S000 or equivalent, S1 = )
*. FUNDING SOURCE ------ I ----- ~ ------------- - -,------
b. FX c. L/C d. Total e. FX f. L/C g. Tota!
(Grant) ( s ) ( ) ( 650 ) 1 ( )J,3,750
(Loan) ) (( ) ( ) ( ) ( ) ( 3
Other 1.
U.S. 2.
a. Appro- b.Prlmny c. Primary FY..j FY FY ALL YEARS
priation Purpoe Teen.
(Alph londi. ) u ..i. C deS C...n i f. Grant I 9. Loan h. Grant 1. Loan j. Grant I k. Loan
650 1Onn0 11100 ~ 3,750 7

12. PROJECT PURPOSE(S) (stay within brackets) [X Check if different from PID/PRP

To improve the capability of agricultural R&D systems in LDCs to -
generate and diffuse technology innovations in intensive production
systems for small farm agriculture that increase food production and
farm income.



] Yes


14. ORIGINATING OFFICE CLEARANCE 15. Date Received in AJD/W, or For
Signature AIO/W Documents. Date of
Ina-t-u- - I7 .^7Distribution

Title Date Signed
mo. day yr. mo. day yr.
Deputy Director, TA/AGR0 '7
AIDO 1330-4 (7-75)


TA/AGR Interregional
e/-?a. symbol t.

7. PROJECT TITLE - i sy ithu-in Cracketts)

ESmall Farm Intensive CroDping Syste2


a. INITIAL FV 77 b F'NAL FY 17 9

mo. yr.
a. [ PRP [j PP b. DATE p 1 77 j
9. ESTIMATED COST (life of project)
(SOO or equiv-lent. S1 -= ) $1,500,000
a. AID APPRO;RIA.-TED "' 9C.0 (-O


&. Afproprl.- i. PrIma'y C. r-.,- ry I r- ST I EA.r FY__ ALL YEARS a. FUNDING b. FIRST c. ALL
ion "s Pur.eC1 )h. I
(Alpha Code) Cd CC" C3. Z.. t I e. .oajn I f. Cant i Loan SOURCE YEAR YEARS
-_.TM.o i io-__ ~"
3 0 I 1,500 I

13. SPECIAL CONCERNS COCE (mam.m um six cons of four positons cach) 14. SECONDARY

15. PROJECT GOAL (sto.- uwthin rao lrts)

O increase income and welfare of the poor majority in LDC's. j

16. PROJECT PURPOSE(S) stay within braciels)

Increase profitability of farming operations of the small landholder
in the LDC' s.


I .
Ten months' direct hire plus approximately six months of either consultants
or potential contractor input.

Sin t r 161%fl flli -. L .

-In 0~. 4"ay yr.

ma. day yr.
I ~IiI~I

Director TA/AGR

1 Data Sq,.zJ
'no* yr



Part 1, B. Summary Description of Project.

Purpose of this project is to improve the capability of agricultural
research and development institutions of the lesser developed countries
(LDCs) to generate and diffuse technology innovations in intensive pro-
duction systems for small farm agriculture that increase food production
and farm income.

Primary means to that end is the testing and adaptation of R&D metho-
dologies to the manpower and financial constraints of the LDC institutions.
The project will be implemented by U. S. institutions working in close
collaboration among themselves and with LDC institutions.

Methodologies will pertain to evaluation of farming systems, collection
and analysis of data on the small farmer and his resources, efficiency of
collection of technical data, generation of alternative systems, and
diffusion of farming system technology.

The project also provides for stimulating the utilization of the
methodologies by means of publications, training sessions, and consultant

While the project aims primarily to test, adapt, publish, and teach
improved methodologies, it will test them in on-going, operational farming
system programs in LDCs; i.e., it will work on substance. For this reason
there will also be a substantive output to the program such as improved
systems adapted to the sites in which it is working, some system component
technologies with rather wide adaptation, and some problems identified for
conventional R&D activities.

The project is designed in two phases. #rhe first phase, lasting about
two years, will 1 nvtory aml futb mth1 Anlngies now in use in farming
system development. .The second phase, for about three years, will test
methodologies identified in Phase 1 and adapt them to various levels of
competency of LDC institutions. Each has wm ,-rSlt in a set of how-
to-do-it manuals and one or more training sessions,

At least three U. S. institutions will be involved in the project,
each one taking responsibility for a major geographic area and providing
leadership in a methodological area,

The testing and adaptation phase will be implemented in LDCs with
as much collaboration with LDC institutions as can be achieved, This will
necessitate the identification of LDCs which either have ongoing farming
systems projects or are planning them in their own agricultural develop-
ment strategy.

Part 1, C, Recommendations.

It is recommended that this project be approved for a period of five
years for a total expenditure of $3,750,000. Fund authorization is being

requested for Phase 1, which will be completed within two years, in the
amount of $750,000. It is recommended that the remaining funds $3,000,000
be authorized on the basis of an evaluation of performance in the first
phase and a general work plan for the remaining three years.

Environmental Determination.

Environmental aspects of this project have been considered. Projects
of this type normally do not require the filing of an Environmental Impact
Statement (22CFR 216.2(b)). This activity is not deemed a major Federal
Action (Section 1500.6, CEQ guidelines) since the project will have no
significant effects which adversely affect such aspects of the human environ-
ment as air, water, land, flora, or socioeconomic conditions. Therefore, it
is recommended that the Threshold Decision be deemed negative, constituting
a Negative Determination since the project will not result in activities
affecting environment.

Part 1, D. Summary of Findings.

1. Interest is high in farming systems, in both A.I.D. and other
programs. A large number of mission projects treat the problem directly or
problems closely related. A.I.D. conferences have also identified the
interest. This project, in fact, originated in a joint conference of Latin
American mission directors and agricultural officers and Technical Assistance
personnel. A similar conference in Asia reflected the interest, which was
confirmed in mission visits by ASIA and Technical Assistance Bureau personnel.
At least four of the international agricultural research centers have active
programs in farming systems.

2# Conventional technology innovations tend to favor the relatively
well-endowed subsector, both because of the bias of delivery systems and
because the innovations tend to be keyed to high resource situations, In
part the interest in farming systems is based on the judgment that in-
novations in farming systems would be biased in favor of the small farm

3. Much of the work being done both in the international agricultural
research centers and in national systems is highly empirical in nature, and
it addresses substance much more than methodology. Few underlying con-
ceptualizations of the R&D process in farming systems were found. Terms
are not standardized. The original term "multiple cropping" was found in
three distinct usages.

4. Methodology can be made much more generalizable and thus trans-
ferable than can the substance of farming system technology.

5. Two characteristics of technology innovation are relevant to
this project. Most standardized methodologies have been developed in

the MDCs (More Developed Countries) and in keeping with their resource
endowments of well-trained personnel, adequate to liberal budgets, and
modern equipment and maintenance. The codified methodology is found to a
much greater extent in the field of scientific research than it is in
technology development, testing, and adaptation. Recently, there has
been some innovation of these MDC technologies for use in LDCs, in rural
development projects and in some national systems, but no significant
methodology testing, adaptation. codification, and publication have been

6. Most MDC technology innovating methodology (genetic, cultural,
n~nn oomic) has been or ented to single commodities, either crop
or livestock..

The conclusions drawn from these findings are (1) that a worldwide project
would add little if it simply becomes another empirical project oriented
directly to the farming system per se, but (2) that a project aimed at
adapting and diffusing methodologies fitted to LDCs is badly needed and
is appropriate for a worldwide project.

Part 2. Project Background and Detailed Description

A. Background.

During the past few years, technical assistance agencies and
national governments have come to realize that small farms and small farmers
are important in the battle against world hunger and poverty. Increasingly,
efforts are being made to find ways to serve the small farmer or herdsman
and to help him receive the benefits of new agricultural technology. To
date, such efforts have been less successful or predictable than desired
Concern has been expressed about the ability of the scientific and tech-
nological community to come to grips with the small farmer and his pro-
duction and income problems. Indeed, there is cause for such concern, for
at present there is little opportunity for the scientific community to
become familiar with the small farmer and with potential innovations needed
to improve intervention and technology transfer opportunities for small

The Problem.

The general problem complex to which this project is relevant is
that characteristic of small farm agriculture--a very high ratio of man
to resources exacerbated by the low productivity of those scarce resources.
There are many problems in that complex. This project focuses on one,
the scarcity of improved technology that is relevant to the small farm
operator. It addresses that scarcity indirectly, not directly. That is
to say it does rt* im to develop improved r '--nIQolo7v on a significant
fcale, but rather to provide to t rsriealy mion institutions
-e --e 1LDC T an improved R&D methodology that will improve thi-r ranability
for developing their own better intensive farming systems.
i y

This project was conceived as an attempt to stimulate development
of technology biased toward the small farmer. While most modern agricultural
technology j r _Q2 zcas-le neutral, it is de facto biased toward the farmer
relatively well-endowed with resources. There are two explanaTTons fort-
tbPo fazto iag- .f *eost obvious is that the institutional mechanism that
delivers information, -inputs, and services LU Ll -the-predeee- i + asked ward
the larger, better-endowed farmer. Even though seed, tertilizer~and
insecticide are divisible inputs that can be applied on the smallest farm,
e institutions simpl ont deliver them adequately to the small farmer

The second explanation lies in the technology itself. A central
function of improved technology in agricultural development is to provide
means (techniques and commodities) by which the producer can use relatively
abundant and cheap resources to substitute for those resources which are
relatively scarce and costly. The relationship of scarce resources to
relatively abundant resources varies among farmers, and much variation is
related to farm size. A new technology always has differential impacts on
different groups. The early high yielding varieties, for example, were
specifically designed to make efficient use of chemical inputs and irrigation.
Once the capital investment is made in an irrigation system, the opportunity
cost of water is low, and for years chemicals and especially fertilizers
were low-cost inputs to relatively commercialized farmers. However,
there is a large group of farmers for whom water is a scarce item and
fertilizer is costly, and it is to this group that this project is addressed.

The consequence of these biases are serious. Unless demand presses
exceptionally heavy on supply,, the producer by-passed by a technology in-
novation is worse off because of it. His competitor has a new advantage..

The Rationale.

This project is based on the rationale that an intensive system of
production (manifest in a variety of ways) would be biased in favor of
that farmer who hires relatively small amounts of labor and who operates
with a relatively low level of mechanization and other forms of capita-
lization. This rationale is supported by the fact that machine tillage
and harvesting of crops grown in association (a common technique of in-
tensive farming) is not feasible to a very great degree and by the assumption
that hired labor will not give the timely care and attention r1iuired by
intensive production systems, unless under careful supervision.?.

The conventional approach to increasing farm profitability is to
increase crop yields per acre, almost always commodity by commodity and with
a heavy dependence on purchased inputs. Many farm operations in the LDCs
are beyond the reach of this approach. In the basic food crops of corn,
beans, and sorghum, small farm crops in many countries, yields approaching

today's biological maximum would not produce adequate levels of income
on small holdings. further, the conventional sinzgjr "p approach wouln
place a hL ym stress for short periods ofthe year on the family labor
whil ..i i4 41a most of the timely.

Definitions and Concepts.

The so-called small farmer is commonly defined as one with a
shortage of the land resource. In fact, he has a shortage of all resources.
Consider the shortage of labor in breaking land with no, or limited, outside
source of energy. This heavy work must be done in a period curtailed often
by the beginning of the rains and the crop planting deadline. If broken
land can be used more intensively, then labor productivity would be increased.
Increasing employment opportunity will result not from simply increasing
total amount of labor needed but from a more uniform distribution of labor
requirements throughout the year. Conventionally, a seasonal labor bottle-
neck is broken by mechanization, a substitution of capital for labor, an
alternative of limited value if the capital constraint is as severe as
the labor constraint, Another alternative is to distribute labor require-
ments to other seasons of the year, which would require modification of the

This project deals with systems, and the system concept is a
treacherous one. Since "everything is related to everything else", systems)
form and reform as a function of analytical or other needs and must be
defined to serve the convenience of the manager.

Three target (or object) systems are relevant the farm household -
firm system, the farm production system and the multifarm system, These
are the systems for which action is intended* the "object' of the action.

The LDC small farm household-firm is the system to which the pro-
ject will be addressed; i.e., it will be the beneficiary of the project.
The household-firm system is defined as the farm family, with its values
and aspirations, its supply of the human resource, its stock of other
resources (land and non-land), and its variety of activities (economic
farm, economic nonfarm, and noneconomic). The household-firm is analogous
to the human being as a system,

Within the household-firm as a system, there is a production
system, just as within the human being there is a circulatory system. The
production system will be the principal dependent variables

The production system is defined as that part of the household-
firm combining land, capital, technology, and the- human resource (labor
and management) for the purpose of producing crop and animal products
for home use and sale. The analysis will take the values and aspirations
of the farm family as well as off-farm employment opportunities as given,

just as it does market forces and ecological and locational factors.

The project will also involve a multifarm system, a group of
farm household-firms, analagous to a baseball team which is a system con-
stituted by a collection of human beings. A multifarm system is defined
as a group of household-firms with similar production systems plus selected
nonfarm entities, the services of which must be shared by the household-
firms and are important to the success of each member enterprise.

There must be a considerable interrelation between the multi-
farm system and the production system, and this interrelation has to
be accommodated in the methodology to be developed. For example, a pro-
duction system, perfectly logical per se, may be viable only if a certain
facility can be supplied, or the technicians may have to search for pro-
duction systems under specified constraints of what can be made available
in the multifarm system. Reasonable expectations of the facilities of the
multifarm system will in most cases be "givens" in the selection of pro-
duction systems technology. In turn, the individual production system will
be a given in analysis pertaining to the multifarm system.

Several distinct "implementing systems" (as contrasted to "object
systems") also can be identified. From one point of view these systems
can be thought of as subject (or actor) systems acting on the target (or
object) systems.

The first of these is the national technology innovation system.
It will have several component systems. (Use of the word subsystem is
useful only in specific situations.) One of these systems is the R&D
system, commonly identified as the research service. This is the system
that generates technology innovations and must assume responsibility for
delivery to the diffusion systems. The R&D system must maintain close
contact with the farmer, both in designing R&D work and in tes'n 1-
novations. In most countries there will be several diffusion systems -
each of several agencies, periodicals, radio stations, private organizations,
and commercial entities. The A.I.D. contractor or group of contractors
constitute another system. From the project point of view, this system
(or systems) will be the implementing (or actor) system. From the project
point of view, the national R&D system and the national technology
diffusion system are target systems. The client of this project is the
national system o just as the producer is in turn the client of the
national systems

Still another system is relevant, the product or output of this
project, which is a collection of technologies ,-economic, agronomic,
ecological integrated into a system of technology innovation especially
adapted to some specific types of situations.

.A "small farm" is one which employs less labor than is supplied
by the family of the farm operator, either owner or renter of the land,
assuming a low level of investment in machinery typical of LDCs.

Production system technology per se is the technology of putting
pieces together production technology, labor, land, purchased inputs,
and capital under known and expected characteristics of weather, eco-
logical factors, price, supply, and demand, Systems technology is a
jmanagM-pap- -Qa"-nlgyg in contrast to production technology*

Production technology is defined as a set of techniques some
embodied in commodities pertaining to the manipulation of physical and
biological factors for the express purpose of production. Conceptually,
management technology and production technology can be distinguished,
but in practice they are often so closely interdependent that distinction
will not always be easy. Further, standardization of systems among
household-firms may be possible to such a degree that the management
function will be obscured. However, sight must be maintained of the fact
that the central technology involved in production systems is that of
"putting pieces together".

Placing heavy emphasis on the individual farm in a central A.I.D,
R&D project, which simply cannot operate at retail, is the chief reason
for the central role of methodology in this project.


Certainly, there is great interest in farming systems research,
Several intern-alaonal centers i. i, _LLiSAT T IITA- and ILCA- have
ormal tarmlnmn T had a farming systems program, but
dropped it in order 1o cUlt7;Le on detailed component research. Several
LDC institutions have formal research programs in farming systems, CATIE
in Central America is mounting a regional program and providing leadership
to programs in Central America. Countries with significant research programs
include the Philippines, Indonesia, Thailand, Nepal, Sri Lanka (Coconut
Research Board), Nigeria, Guatemala, and El Salvador.

Most of the existing research programs are based in experiment
stations or research centers. Some involve farmers directly in the research;
others deal mostly with technological or biological opportunities which
revolve around one or two major commodities, and relatively little effort
is made to understand the system being used by farmers in the region. Adoption
by the small farmer of res~parh information gained is often poor, h...... the
farmer is the last person to be considered in the.system,

There is real need to look at the methods used in research on farm-
ing systems. Methodology being used ranges from a conventional, experimental
plot approach on research stations to some form of farm management surveys.

A few institutions IRRI and CATIE are notable examples --
have included studies Of ex.isln.i -iing ey'tom aw JU early part of
their farming system activities.

Methodologies employed in farming systems research vary widely
and lack uniformity. Some approaches hPing used are highly sophisticated
and beyond the capacities of LDC scientists and institutions, Most new
programs lack 1 uldau1e d di lu because suitable methodologies have
nf- hPen inventoriea, evaluae,teted, and codified for LDC use.

Especially needed is a systematic7 straightforward methodology
of farming systems R&D by which LDC personnel can understand the varying
characteristics of existing farming systems, and the resourcelmanagement
factors facing the farmer and his family.JThe approach to be used can
draw heavily on methods used hy pituire and range management technicians
and by ecologists.JsIt must be capable of measuring and evaluating inter-
actions of biological, physical, economic, and cultural factors under
man's management. Such an approach is necessary for LDC institutions
faced with developing more intensive production systems for their small

Technical assistance agencies and regional or international research
organizations would also benefit from a more standardized methodology.
Currently, even such a tightly knit group as the international agricultural
research centers do not have a standardized methodology. The Technical
Advisory Committee (TAC) of the Consultative Group for these centers
has organized a review in reaction to this very problem. At one TAC
meeting, a member characterized farming systems research as "proliferation
without edge." This review will terminate with a seminar in December 1977.
This "confusion" is exemplified by 4%hp .ng-q f "mlItiple
cropping" found. Dalrymple (1971) used the term to refer to two or more
single crops produced on the land in one year, but in sequence. Bradfield
used the term to describe the simultaneous production on the same land
of more than one crop. ICRISAT uses the term generically and divides it
into sub-categories of associated cropping, sequential cropping, relay
cropping, and ratoon cropping.

Calling attention to the lack of conceptualization and of standardized
methodologies, codified and presented in a teachable form, in no way con-
stitutes a criticism of current work. While farmers, in the curious
processes of folk wisdom, have been developing innovations in farming
systems for centuries, institutionalized .tsto induce innovation
vjpblic R&D entities in the Teld dates back little more than de ade
Jeginning witn Pro essor Braafiela'B work in te .nhilppines. Much more
progress has been made with the empirical approach over this time than
would have been possible had field operations been limited to the pace

which conceptualization and methodological standardization could have
attained. In fact, without the considerable empirical experience,
unsystematic though it may seem, the current project in methodology
would not have been feasible. The empirical approach has been the correct
one up until now.

Calls for a more systematic, rationale approach for farming
systems research are not new. An FAO Conference in Africa (FAO Conf. on
Establishment of Cooperative Agricultural Research Programmes Between
Countries With Similar Ecological Conditions in Africa: Guinea Zone,
FAO August 1971) recommended that: "high priority be given . . to
the study of agronomic, economic and sociological conditions existing in
farming systems in the Guinean Zone"; and, "as a primary task of this
study, the terminology and methodology of investigation, experimentation
and analysis of farming systems should be standardized to permit compara-
bility of results within the Guinea Zone". The Conference went on the
call for "a training seminar or workshop to organize and integrate
investigations and methods of information dissemination concerning farming

Geographers have long been concerned about systematizing research
methods for studying agricultural systems. Blaut (1959) presented a
careful analysis of an approach to measuring and quantifying agricultural
information through "mmrfroeooraphic sampling". Blaut's work emphasizes)
a major dilemma in farming systems research; that is, most work has been
descriptive in nature, and it has been difficult to quantify or evaluated
the dynamics of a system or to improve predictability or success in
technology transfer.

Most early farming systems work was done by farm management
personnel. Surveys were their first major tool. Two major types of surveys
were used, case studies and sample surveys. Case studies were done in
depth on one or a few farms. Sample surveys were conducted with the
aim of finding variability among farms as a substitute for experimentation,
the common tool of the colleagues in plant, soil, and animal sciences.

( biological and physical scientists in MDCs did not often become
involved in farming systems research. In part, this was due to the faca
That farmers were expected to and were capable of integrating new J
I information, and that the tendency toward monoculture caused a rather r
telescopic viewpoint in looking at the farm. Hence, little farming
systems research was done outside farm management. -'

,Biologiral and physical scientists did become involved in activities
approaching farming systems research in two ma or areas land use
(su y dominate b soil scientists) and range management. Probaby

range andA pa ture management scientists come closer to doing systems
research than most other biological disciplines' Their methods -- in
soma cas~s ^ c-e. or adapted from ecological research may be of most
value to us in study of farming systems.---

In recent years, especially since the dramatic, pioneering research
of Dr. Richard Bradfield at IRRI in the Philippines in the late 1960s,
agronomists, soil scientists, economists and other disciplines have begun
to study farming systems, usually on a multi-disciplinary basis. Efforts
now began to move from the more descriptive types of studies to those
in which biological and physical measurements were being made, complicating
the situation even further.

The early work at IRRI focused on demonstrating and dramatizing
the tremendous yield potential of the tropics with its year-round growing
conditions. Centered on rice, it followed some of the multiple cropping
concepts developed by the Chinese. The studies were not related much
to the traditional rice systems of Luzon, but were geared to emphasize
the benefits possible through c ages in tillage which did not require
puddling of the soil.

Later work at IRRI, led by Dr. Richard Harwood, began to look at
conventional farmer systems of growing rice, and to an evaluation of those
crops which the farmer grows in his usual production system. This
approach soon began to yield information on the stability and wisdom of
many practices of the small farmers. There were often good reasons for
what the farmer did. Also, these studies pointed out the importance of
the need to understand whole farm systems for the "homestead" area, the
field margins, and livestock feed sources were all found to be important
factors in understanding the farm.

Recently the IRRI program has begun to relate farming systems to
agro-climatic zones and their characteristics. Also, they have found it
convenient to classify rice paddies according to the topography and level
of the water table. Another useful measurement is soil texture. Clearly,
the IRRI research program is attempting to come to grips with the natural
resource base and to ways of characterizing agricultural zones and their
common characteristics.

The ICRISAT program is focused on managing limited water resources
in dryland farming systems of the semi-arid tropics. They have emphasized
the management of farm fields as small watersheds, and are heavily involved
in tillage, land forming and development of storage capacity to allow
water harvesting. The ICRISAT program can be-described as primarily a
resource (water)-management program, coupled with socio-economic village
studies and cropping pattern work. l o 'AeOy o os1k-) A ,si0


IITA has placed emphasis on shifting cultivating in its program.
Also, IITA probably places more emphasis on indigenous food crops than
most other programs. Soil -mangmanta -a notably mulching, minimum tillage
and other cover mnana ment practices plays an important role in this
program. A large multi-disciplinary team has been assembled o luUk du
existing systems in West Africa and to evaluate potential innovations
on these systems. One very interesting feature of the IITA program is *
the effort to relate land use and cropping patterns to catenas (a succession
of soils across a geographical area which varies according to slope and
level of water table). In this system dryland crops are planted on
upper slopes while hydr- hyti crops such as rice or aroids are planted
in lower areas. This natural resource approach is similar to the IRRI
paddy classification work.

A pioneer farming system R&D program among LDC institutions is
that of CATI, which is headquartered in Turrialba, Costa Rica, but which
is working throughout Central America, This project, backstopped by
projects on agricultural information and soil fertility evaluation, is
geared toward intensifying production on the small, limited-resource farms
of 4rT-1 America. It emphasizes basic food crops, notably corn, beans,
sweet potato and cassava. Ehe methodology used at first was more or a
-ynventional ex erimpntal station asvroach, but more recently the approach
is being shifted to studies on the farm and to relating existing systems
to "gradients" in natural resources or rainfall. Efforts are also being
made to relate the systems to work on "soil analogsR1, an approach akin to
that of the benchmark soil idea It should be possible to link soil
fertility information and exp Prirp M gnnr snoil c-icassirication
system (e.g., the U.S. Soil Taxonomy) and thereby provide a basis for
technology transfer not only from one part of Central America to anroCher,
Tut also from other parts ot the world. it should be noted here that the
harn ar' research centers in the tropics into the U.S. Taxonomy so that management
information and research information can be transferred from one part ot
the world to another without costly new research.4It would appear that
the CATIE program on resource gradients and soi6 analogs ties in well
with the ecological approach to farming systems research which this
project will emphasize.

In this brief discussion it can be seen that uniformity cannot be
said to exist in either methods or philosophies of farming systems research.
It follows also that there must be great indecision and confusion on the
part of LDCs wishing to initiate farming systems research. There would
appear to be little doubt that a project on evaluating, developing, and
testing reasonable, practical methods for study of farming systems would be
beneficial. Especially useful will be approaches which are specifically
designed for LDCs conditions where trained manpower and research capacities
are limited. Increased efficiency should be a benefit of the project.

In summary, there is much research going on in farming systems.
Many approaches have been taken in this work, and there is little
uniformity in methods or procedures. A state-of-the-art analysis is
needed which will inventory, evaluate and test the more promising
approaches and methods. Methods used must be codified for broader use
by LDCs and for use in training. Methods specifically designed for
LDCs must be outlined and tested. Especially needed are approaches which
can systematically sample the variation in natural resource endowment.
as well as a t n -- s cin-coGn-mic factors. An ecological approach i
proposed which would capitalize on accepted research methods used by
ecologists in dynamic biological/physical/man-influenced situations.

Part 2, B. Detailed Description of Project.


The purpose to which this project will contribute is the improvement
of the capability of agricultural R&D institutions of the LDCs- to
generate and diffuse technology innovations in intensive production
systems for small farm agriculture that increase food production and farm
income. Although the operator of the small farm and his family are the
ultimate beneficiaries of this project, the project specifically addresses
an intermediate entity and aims to help it improve services to the small
farm operator,


Two primary outputs are expected from this project.

1. The first is a set of integrated, multidisciplinary intensive
farming system R&D methodologies adapted to manpower and financial
resources of the LDCs, packaged for easy delivery to LDC institutions.

2. The second is training in the methodologies for LDC personnel
engaged in farming systems R&D work.

These methodologies will pertain to:

a. Evaluation of farming systems, including evaluation of
the farmer's resources, identification of points of intervention, problems
identification, and specification of data needed.

b. Collection and analysis of data regarding small farm
resources, including social-cultural factors, institutional services,
ecological conditions, and economic resources.

c. Collection, generation, and analysis of data on technical
coefficients, with emphasis on efficiency of collection and generation
and adaptability and transferability of data.

d. (Generation and identification of alternative farming systems).

e. Diffusion of technology, from A&D entity to field agencies
with an extension functions and feedback from users te R&I workers

Because of the means used to accomplish the primary output, secondary
outputs or byproducts also are expected. These substantive outputs will
be produced because it is virtually impossible to test and adapt improved
methodologies without working on real-life substantive problems. These
byproducts will consist of:

1. One or a few improved farming systems for localized areas in
the countries in which testing and adaptive work is sited.

2. Some component technologies relevant to farming systems that
have a relatively wide adaptation.

3. Some problems identified that need attention in conventional
R&D programs.

-Einall.. this project will develop an expertise in three universities,
12 or 15 technical people, that will be available for consultation and
technical assistance to LDCs, during thp life of the project and beyond. .


It is feasible at this time to give the areas in which methodologies
will be defined and a discussion of the functions they will address.
Experience in the project itself may modify original expectations.

a. Evaluation of farming systems will serve several functions.
One is the conventional economics. What makes money? In conventional
procedures, accounting is done by commodity and aggregated for the firm.
accounting for farming systems may have to be considerably different.
The evaluation system will have to be simple enough to be used in the
LDC system.

In the R&D sense, evaluation of the farming system will be
expected to identify problems, indicate the most likely points of inter-1 '
vention, and help specify the kinds of data needed.

Linear programming is one of the most powerful tools for this l*jf..
purpose, and will llkye ... vy. Many LDCs have the resources
to handle it, especially in view of the advances in computer technology ,

However, the project will be sensitive to situations in which it may
be necessary to develop more appropriate methodologies.

b. Collection and analysis of data pertaining to small farm
resources become important in light of the fact a growing body of evidence
suggests that an important factor for non-adoption of technology is its
irrelevance to the farmer. This underscores the need to know and under-
stand the farmer in terms of his "relatively abundant and relatively
scarce" resources.

Few, if any, LDC institutions have the knowledge and understanding
of their small farmer clientele that enable them to aim for technology
innovations that fit this "scarcity-abundance" ratio, The required
technology will include an effective sampling technique, so that a small
number of observations reflect a broad area, and a data gathering guidelines
so that the right phenomena-agronomic, ecological, economic, and cultural"-
are observed and analyzed correctly and in relation to each other. The
technique must be relevant to the traditional farmer and also adaptable
to the farm operation which for some reason has become dynamic.

c. A second kind of data needed is that pertaining to technical
coefficients. It is expected that in some cases these coefficients will
be provided from farmer experience. In other cases, the data will come
from experimental plots. In either case methodologies are needed (1) to
reduce the number of observations necessary, (2) to produce data that
has wide adaptability, and (3) to analyze the data efficiently. *Farming
._yrr'" "r in fn !eng grnIQing sea ....0 nf the tropics is currently
generating d n a c=g110 t is becoming unmanageable. Methodologies
must be evaluated on the basis of both etiiclency ana transferability.
Site problems are especially important in LDCs where abrupt changes occut
frequently, and the institutions have to accommodate this variability withJ
fewer resources than the MDC's have.

d. Developing alternatives to current farming systems will, at
the start, be largely on an empirical basis. The growing experience in
the intensification of cropping systems will present many ideas for
testing. Knowledge on the relation of climate to crops species and even
on the climate itself will help. Also helpful will be new technologies
which open up new possibilities, such as short season varieties, drought
tolerant varieties and new crops. The field of ecological botany probably
has some concepts regarding plant association that can be drawn upon for
other ideas. Some guidelines are emerging, such as grain legume or early-
late crop combinations. This experience has not been codified into a
systematic scheme that will help develop alternatives. In fact, some of
the current guidelines may be more old wives' tales than guidelines. The.
conceptual basis for this line of methodology is the weakest in the project.

Adding livestock as a component opens up innumerable other opportunities,
as does forestry, under the new concept of agri-forestry. Some system
is needed to help guide R&D workers to reasonable alternatives rather
than their having to rely on cut-and-try empirical methods. U.S. farm
management methodologies will have some utility.

e. Diffusion methodology needs to be conceptualized in two
major components, one being the center extension (associated with the
extension specialist in the U.S. system) and the other being field
extension (associated with the U.S. county agent.) This project will
address the first, which in effect is the R&D unit's responsibility to
deliver its product to the entities that can take it directly to the
farmer. This project will attempt to define this first step in diffusion
as part of the R&D process, that part which relates to users and is
responsible for feedback from the users.9)%t ais- expected that the RD
ent it 1 ]i have real impact on the various field agencies found in nInnt
all LDCs if it can evelop a significant new tecn and if it can
provide technical support to the field agencies, appropriate to the new

Substantive Output

This project makes much of the methodological output, on the basis
that methodology can be made greatly more transferable than can the
substance of farming systems. However, it acknowledges the difficulty
in practice of separating methodology from substance. In Phase 2, for
example, the three contractors collaborating with LDC institutions in
testing and adapting methodologies will be working on real farming systems.
There is no other way of testing the methodology. Thus, in these countries,
farming system innovations will be produced. To the extent that some
of the methodologies are effective, these methodologies or certain com-
ponents of them will be adapted to rather broad areas within the country.

The conventional wisdom is that farming systems are ultra-site
specific. This project will aim to gain a better understanding of site
specificity. It will give close attention to the relationship between
system performance and site characteristics, in order to broaden the
area of adaptation as much as feasible for a farming system or components
of a system. At this interface it will be difficult to tell which is
methodology and which is substance.

Some indications of wide adaptability are emerging. 4n Central
America, the planting of grain in double rows every two meters rather
than single rows every meter seems to be a technology with little reference
to site. In the same region, cabbage planted in lowland rice or in high

altitude wheat, has produced a crop with no impact on the associated

One of the greatest substantive outputs of this project could
be the guidance that is provided for conventional R&D programs. For
example, this project could provide information on farmer needs on which
'Could be haSpd mIch m-inor r1a r"1 for ro breealng programs --


Inputs for this project can be translated into dollars. It is
planned to implement it in two phases, one to inventory current metho-
dologies either used directly in farming systemss work or taken from
other action areas, to organize and codify these methodologies, and
to present them in a manual for LDC institutions, This phase should
take no more than two years and up to $750,000. The second phase, for
three years and $3 million, will test these methodologies in LDC sites
and adapt them for general LDC use. It is aimed to contract with three
U.S. universities for the entire project. Each phase will end with a
training session for LDC workers. This training -sessibrf, 22.
two functions, one being the obvious transfer of technology& The other
function will be to serve as a discipline on the contractors to produce
a usable product that will be put to a test of sorts in front of an
audience. There will be an A.I.D. input of at least one-fourth personyear
Per year to manage the project witn particular emp as s on coordinafing

Each university will be expected to assume responsibility for
a major geographic area of the world--rca, Asa, and Lain Amerc.
To the extent pos1ible -ah university will be exected to take te--
lead in certain subject matter aspects of the projeef.

Part 3. Project Analysis

Technical Review

Farming systems research is in its infancy. There are almost no
guidelines for conducting the research nor for utilizing its benefits.
Methodologies being used are mainly those of traditional experimental
agriculture or biology. In most programs the work is begun on existing
experimental stations and is usually based on technological or biological
opportunities, intuitively conceived, which appear to be possible.
Usually these perceived opportunities revolve around one or more major
food crops or commodities, usually cereals.

Methods of study in most new programs involve standard plot
techniques which are very good at evaluating single crop responses to
treatment variables, under closely controlled conditions. When multi-
crop situations are encountered, these methods rapidly become inadequate
because of variable harvest or planting dates, intr- or intra-crop
competition (or complementarity) or other related factors.

Design and methodological problems also appear rapidly when one
considers that, in most farming systems research programs, the farmer
is either left out or is considered as an afterthought. Indeed, the
history of many farming systems research programs has been one of
(1) starting conventional plot work on existing experiment stations;
(2) eventually recognizing the need to come to grips with the farmer;
followed by (3) some efforts to study existing farming systems. When
study of existing systems does begin, it is often dominated by a single
discipline and is partial. Economists have taken the lead since farm
surveys have been a function of farm management specialists. Thus,
farming systems research is enmeshed in a major dilemma, (1) how to
improve and conduct farm surveys with a low capability in farm management
economics; and (2) (perhaps even more serious); how to involve social
scientists, and biological and physical scientists in meaningful,
sensitive, integrated farming systems research work. Clearly, to obtain
improved farming systems research, some careful thought and attention
must be paid to methodologies which can be used by multidisciplinary
teams. This is difficult enough for MDCs in sophisticated research
programs, but is even more important when we consider that we are
attempting to find suitable methodologies for use by LDCs where trained
personnel and resources are so limited.

Why conduct farming systems R&D at all? Some benefits of such
research can be appreciated by contemplating what may be considered the
Benefits of Farming Systems Research:

A. To gain an understanding of:

1. The natural resource base

-- its potential

-- its use, patterns of use

-- problems, both existing and potential

-- integration of cultivated and non-cultivated lands.

2. The farmer

-- his utilization of land and water resources

his production systems

of food crops

of cash crops

of livestock


a. use of indigenous crops

b. tillage and land management

c. cultural practices

d. cropping patterns, sequences

e. sowing and seeding methods

f. harvesting methods

-- his use of labor

his management capabilities

his marketing problems and processes

-- his income and food needs

-- his response-to change and technological opportunities.

B. To improve research and development activities in agriculture.

to make research more relevant to the farmer

to improve problem identification and analysis

to identify plants or plant characteristics needed in
real life production systems

-- to identify animal production problems and potentials

to more effectively mobilize research talent and resources

-- to develop technical innovations which are designed for
the farmer and his resources

to develop and improve the capacity of national R&D
systems to deal with farm problems

to more effectively integrate new technology into common
practice, especially by small farmers.

C. To improve services for farmers and the agricultural sector

through a better understanding of the farmer, his system,
and his problems

--by involvJin the farmer in the process of prrohlm identification,
Analysis, and solution

by mobilization and deployment of knowledge, both from R&D
efforts as well as from traditional or farmer wisdom sources

-- through improved awareness, understanding, and confidence
of agencies and organizations in serving and meeting the
needs of the farmer or herdsmen.

Conventional research approaches are not adequate for study of
farming systems. As has been stated, most of our techniques are based
on plot work centered on or near research stations. Most present research
is also based on single crop systems. Such methods are limited in
situations where (1) the natural resource base varies widely; (2) more
than one crop or enterprise is involved, as in mixed farming, intercropping,
sequential cropping, or relay cropping; (3) seasonal effects affect the
crop and/or animal mix, crop growth, crop utilization, or other factors;
(4) harvest dates are staggered or vary widely; (5) yields or returns
for several crops or enterprises vary in types of measurement or values;
(6) the availability of labor at certain seasons effectively limits the
range of choices among crop-livestock combinations for typical farmers
in the system, and (7) the availability of required inputs or markets
effectively limits farmer's ranges of choice.

The research approach to be used should be able to deal with the
dynamic nature of farming systems while at the same time providing a
means of identifying common elements or factors among systems which are
important to the farmer and his household. In many developing countries,
women share in the decision-making, management and labor required in
the farming system. By considering the farm and household as a unit, it
should be easier to gain a measure of the unmarketed as well as the
marketable inputs and outputs that are necessary to understand the working

of a particular farming system. yamplae of ....marcke-d materials wnu,1'
he the household's need for cooking fuel and the farm's need for feed
for the livestock used as a source of farm powerV'-

Literature Review

A number of international farming systems workshops have been
held and Proceedings have been published. Thus, a considerable core
of literature has been assembled. In addition, CATIE has put together
h.ih!jiography on tropical agricultural systems (UArT, 1974) which
lists it p rk n nrr opics by cropping system categories.

The best reference work on tropical farming systems has been
the excellent book on Tropical Farming Systems by Ruthenberg (1973).

Three general trends can be confirmed in the recent literature
on farming systems: (1) most of the work has been conducted and reported
on a descriptive basis; (2) most work has been started on experiment
stations and only later was it found that it was necessary to work
directly with the farmer and to understand his existing systems; and
(3) most work has been done on multiple cropping systems or on cropping
patterns, and the role of animals has seldom been considered. As a
result, research methodology has been centered more on conventional research
plot techniques and farm management surveys.

Some changes and new trends in methodologies can be detected in
the new research programs. The recognition and realization that small
farmer production systems which employ mixed cropping or intercropping
of two o more plants are more stable biologically has led researcher--
-attempt to find wayd these complex systems. Research programs
at IRRI, CATIE and CIAT have all confirm a -,immhpr of advantages of
multiple cropping, including (1) fewer problems with insect pests and
diseases, (2) higher yields than when either crop was grown singly in
monoculture, (3) fewer and less spring wAP prnhpm and greaterr
profitability for the farmers? Some very large and complex experiments
have been installed with Tne objective of measuring and evaluating the
benefits of a number of cropping patterns and management practices. One
such experiment had over 200 treatments and covered several hectares.
Increasingly, researchers have come to realize that improved techniques
ar--e needed, if w are to he able to handle the scale and research design
e--BTems of complex cropping systems studies. ---

Another clear trend has been in a growing awareness by researchers
that it is necessary to do a better job of understanding and of character-
izing the physical, biological and sociological environment of the farmer.
Of these, the physical environment and its effects on production systems
has received increasing attention. The concepts of zones or zonation
have increasingly been studied as a way to deal with the problem. Hence,

IRRI has been working with "agro-c11marI t zones", and ICRISAT has placed
a great deal of emphasis on characterizing and delineating the "Semi-arid
Tropics", its primary area of concern. CATIE has developed programs
which attempt to understand land resources by studying "soil analogs".
Also, the CATIE program is based heavily on conducting systems research
along or arrnqc "g-ra-fn-ts" in the resource hbase. It can, therefore, be
seen that these programs are mDving toward ecological approaches which
attempt to measure and evaluate the influence of changes in the natural
(or other) resource base on farming patterns and organization. These
trends are clear but no uniformity of methodology exists.

The other new direction being taken is that of working with the
farmer in partnership, on his farm; this relationship is yielding new
information, for it allows the farmer to become a participant and decision-i
maker in research.

Most of the current literature, however, has little to say about
methodology. It is heavily oriented to reports of empirical studies and
experience. Some quite good work has been done in relating climate,
especially rainfall, and specific crops with particular reference to
intensifying land use, but this has been more of substance than of method.
In this area, however, the two tend to become difficult to distinguish.

Conceptualization and Methodological Notes

Two major conceptual frameworks will be used to structure the
project initially. One of them pertains to evaluation of the farming
system and comes from agricultural economics, more particularly farm
management. The other comes from the discipline of ecology and has been
adapted to some degree to agricultural problems in work on pasture and
range management. Both are methodological concepts, and both have enough
flexibility to be adaptable to a wide range of substantive variables.

The farm management methodologies, while aimed specifically at
system evaluation, have proved themselves useful in fairly precise
specification of data that is needed and of providing analytical pro-
cedures for getting the most out of the data. Zransect methodology, on
the other hand. has demonstrated its utility in the economy of data
collection and the eff ry ected. Essentially a
specialized sampling concept, it can provide more data and Detre- da
with a greatly reduced number of observations, compared to otner sampiln

To a large degree, but not entirely these methodologies provide
guidelines for the type of information to be gathered. However, the
guidelines need to be supplemented by inputs from other sources.

These two conceptual frameworks address directly the first
three of the five methodological areas identified under outputs. At
this stage, there is no conceptual framework that directly attends the
fourth, namely the development of alternatives to current, often tra-
ditional, systems. Both of these concepts will provide insights that
when combined with practical judgments of experienced personnel will
be useful in designing alternatives. The intense study which this
project will provide will have as one of its objectives the development
of a framework more directly useful to generation of alternative systems.

The fifth methodological area, diffusion to the field agencies,
is based on the simple concepts that distinguish center extension from
field extension and identify the need for feedback from technology users)
to technology generators.

A more detailed look at the two principal estceptual frameworks
is provided below.

Methodologies for Evaluating Farming Systems

There are two levels of concern in the review, adapting and testing
of economic analysis methodologies in this project. The first level is
selection of appropriate methods to evaluate existing and potential
farming systems at the national research institution level. The second
is selecting analytical methods that can be used to extend information
to the producer level.

Analytical tools such as enterprise budgeting, total farm budgeting,
cash flow, and linear programming have been widely used in farm management
research and extension to analyze farming systems, new technologies, and
investment alternatives. This section discusses the appropriateness
of these tools in small farming systems analysis and the necessary adaptation
to make them useful for small farming systems analysis in LDCSI

Linear Programming

Linear programming is an appropriate tool for economic evaluation
of farming systems by the national research institution. This technique,
which requires a computer, allows for evaluation of a number of production
possibilities and simultaneously considers production, consumption,
resource constraints, and alternative objectives of the farm business and
household. Linear programming is widely used in farm management and pro-
duction economics research and extension in developed countries and the
standardized. c =ut program can be operative'within the computer capacity
of most LDCs. tiara ograMming can also be taught with a minimum of
difficulties to production trained technicians with little economics
training. No knowledge of computer programming is required to set up
models for computer analysis or interpretation of the results. (The simulation

technique is widely used in farming systems analysis but use of this
tool takes a much greater knowledge of computer programming and there are
no standardized programs for total farming systems analysis.),1

Because of the information requirements of linear programming,
a multidisciplinary approach is most effective.

To use linear programming, information must be specified for the

1. The farmer's household and farm business objectives, either
minimize costs or maximize profits.

2. The technically feasible crop and livestock production
possibilities and the associated technologies and management

3. Input-output relationships for each production alternative and
the interrelationship among alternatives.

4. Resource constraints, including seasonal availability of
labor, and other relevant constraints on the total system.

5. Input and product prices.

6. "Household consumption of on-farm commodities and cash income

7. The social constraints relevant to both the household and
farm business.

It is important to specify a range of farming systems that represents
the reality of the farming area and clientele group, both from the standpoint
of identifying a range of production alternatives and for drawing con-
clusions for farmer education and policy decisions.

The direct products from a representative farming systems analysis
using linear programming would include estimates of:

1.. Income and standard of living potential or the level of
.. achievement for other household and farm business goals.

2. Input requirements for resources used in the farming system.

3. The productivity at the margin of added resources; i.e., land,
labor, and capital.

4. Optimum combination of production activities given the specified
resource constraints, input costs, and product prices.

5. Income sacrificed by the non-optimum combination of-F
production activities.

In addition, there are a number of by-products that can result from
linear programming analysis including

1. A fuller understanding of the farming system production
possibilities and constraints*

2. A large amount of farm level evaluated micro-information,
including enterprise budgets, that can be used in agricultural
sector analysis and agricultural extension activities.

3. An important interdisciplinary educational activity for
training technicians.

4. Generation of a great deal of information useful for policy
and for research and development project efforts.

Linear programming is particularly useful in analysis of the effect
different technically and managerially feasible production alternatives
have on the utilization of abundant and scarce resources throughout the
annual production cycle. In defining the input required for each production
alternative, resource requirements are defined. If seasonal availability
of a resource is relevant input is specified on a seasonal or monthly
basis; i.e., seasonal labor requirements, monthly cash flow, etc. The
availability of each resource for the farm unit is then specified as a
resource constraint. As noted before, one of the direct products from
the linear programming analysis is the estimated productivity at the margin
of added resources. Not only are the scarce resources identified, but
the estimated value of added resources indicates what could be paid for
an added amount of the scarce resource given the production alternatives
considered, prices, and other resource constraints.

Linear programming is an appropriate tool for analysis of the role
of livestock in farming systems, Widely used in the livestock feed
industry in developed countries, the tool could be adapted to evaluate
livestock production possibilities for small farms in LDCs. Livestock
production often can complement crop production on small farms through
utilization of crop residue, non-crop land, and forages grown in crop
rotations as well as a more pfb-i t Year-round ui1 i-1on of lahnr to
produce meat, mik, traction power, and cash income. In the linear pro-
ranming framework, livestock production possibilities are evaluated by
comparing nutritional requirements of the livestock to the nutritional
value of available feedstuff to determine the optimal production activity
while simultaneously considering other resource constraints and production
possibilities. Due to the complexity of the interrelationship between
small farm crop, forage, and livestock enterprises and year-round resource

use, linear programming is a valuable tool for analysis of the role
of livestocK in alternative farming systems.

Although linear programming has been widely used in developed
countries in agricultural research and extension, its use in LDCs in
farming systems analysis has been limited. Efforts are needed to adapt
the tool to farming systems in LDCs and refine procedures to assemble
the appropriate information necessary for implementation of the technique.
Educational materials to train technicians in how to fully utilize linear
programming technique for farming system analysis also need to be

It is important to note that the significant advances in computer
technology of the last few years has made the computer much more widely
available than it was only a short time ago, and taking advantage of this
technology will make much more information potentially available to LDC
institutions than before. Still, the project will be sensitive to the
resource constraints of th1 TiTr-T-tut nn, o nf which still may not
have access to computer far'il'tiej mP S p.chke.

Budgetin and Cash Flow Analysis

Enterprise and whole farm budgeting have been used in farm management
economics from the beginning of economic evaluation at the farm level.
Budgeting is a systematic way to make estimates and is a flexible tool
that can be used for most farm resource allocation problems.

Budgeting can be used to evaluate production and investment alternatives
and is especially useful in evaluating the right direction to modify present
resource use. It will be important in this project to adapt the tools
to handle on-farm consumption requirements and tie together the household
and farm business aspects of the farming system. Efforts will have to be
made to insure a close link between the linear programming and budgeting
at the farm level for both the generation of information for farm models
and dissimulating results to farmers.

Cash flow analysis is a tool used to quantify cash expenditures
and income for a specified period of time. Cash flow projections are
based on a farm business and household plans and expected prices and costs.
Due to the limited cash surplus of small farms, cash flow requirements
of alternative farming systems will be an important measure of the
financial feasibility of alternative farming systems and the potential
cash flow risk associated with different technologies requiring capital
investment that pay back over time. Cash flow' analysis procedures thar
ar_ widely used in farm management etinwrk aj B ay agricultural
cqgdftnstitutions wilL require t hesall farm
environment in SICs.

Transect Methodology

Ecological research techniques have been employed by range and
pasture ecologists in order to understand the dynamics of the growth
and behavior of a number of pasture plants, both desirable and undesirable,
across a widely variable land and climatic resource base as modified over
time by activities of man and grazing animals. The research is based
on the philosophy that it is necessary first- en dae4nte hrnoad nrolo-ical
zones which possess certain characteristics and th-n tn progressively
refine the measurements an characterizations of sub-units which are
identified until a satisfactory precision is atta4lai.4

In order that the limits and parameters of the system can be under-
stood, the ecologist lays out transects, lines that cut across the main
ecological variations as a guide to sampling. He then uses spot sampling
along the transect in order to obtain more detailed information on
representative areas. Applied to farming systems, variables entering
sample design would include cultural, social and economic as well as
biological, physical and climatic factors.

The line transects areas characterized by important differences in'
one or several significant variables. A line that cuts across areas of
variable precipitation, or altitude, or soil types, or social organization,
for example, would be a useful transect. Once the transect is located,
observations are made along it. These observations can be made at large
intervals for some purposes and at small intervals for others.

The transect approach has a number of strengths in farming
systems research. Its greatest indicated value is to detect with a
minimum number of observations the variations in the components of current
systems that need attention; i.e., it should result in the detection of
important problems (or limiting factors) and in the providing of a context
within which several problems can be compared to each other in their
severity. It can also Iabe i f to- lcate -gperimental plots so that a
small number of plots can generate data that have relevance over a broad

No rules exist as to number of transects or their direction, but
guidelines can be developed that will enable such decisions to be made.
Judgments are needed on how to orient them to adequately sample the
important macro and micro variations. The scope of a transect is
determined by the macro-variation in climate and resources of the region,
the differing population densities, and the variation in customs and
traditions of the people of the region. The transect must also cut across
the relevant microenvironments such as field slope, drainage, etc.,
encountered along the transect. Hence, in farming systems research,
the width of the transect will probably not be uniform throughout its
length. The width must be sufficient to take in a whole farm. Where
farms are small, several farms may have to be included at a given point

along the transect. Transects may be laid out in parallel to cross the
region, or they may cross at angles to sample the macro-variation. Xarms
selected for sampling must be representative of the region and reflect
the resulting interactions etwee physical biological, social a
economic forces. Valid criteria for selecting represent tatve farms for
the purpose of investigation are extremely important, because to a
significant extent the benefits of the methodology outlined draw heavily
on farmer-researcher interaction.

The transect methodology is especially relevant in an R&D process
that aims to include the farm and the farmer. Since it involves a straight
line that cuts across variations, it can mark the boundaries where the
variable changes. With a small number of transects to provide the second
dimension, it is relatively easy to locate or identify an area homogeneous
for an important variable. It is also relatively easy to identify and
measure non-contiguous areas having the same characteristics, a factor
quite important in transferability of technology. Further, since the
transect operates from straight lines, a great deal of information on the
area can be produced from a relatively very small number of observations
compared to other types of sampling. When the number of observations
needed is significantly reduced, resources are freed to observe more
variables or the same variables over a range of time. All of the above
pertains to the complete range of data in which the farming system is
interested, and it pertains equally to data generated autonomously on the
site or to data generated by purposive experimentation.

This transect approach was first proposed by Study Team 4a (Farming
Systems) of the World Food and Nutrition Study (NAS-WFNS, 1976). A
synopsis of the approach is presented in Annex 2.

This methodology relies heavily on ecological research techniques,
and would appear to be especially useful for evaluating and understanding
the natural resource base of the farmer as well as to provide a basis for
measuring and evaluating dynamic relationships between edaphic, climatic
and biological factors under the management of man. Analytical techniques
borrowed from vegetation ecology will be analyzed to determine their
usefulness in farming systems studies. Such ecological measurements and
techniques also appear to be a potential bridge to tie physical/biological
information with socio-economic studies.

The ecological approach will compare the relationships between farms
across varying natural resource, economic, or cultural conditions. As a
technique it can produce these outputs: (1) Analysis of the soil/water/
climate environment for various farming systems and the integration of
these resource data to the benchmark soil concept to specify and characterize
the manner in which farming systems are determined by soil/climatic
factors; (2) understanding of plant types need for specific cropping
patterns--multiple cropping in all its variations--thereby improving

identification of breeding objectives, special management needs, and
livestock feeding opportunities; (3) characterization and understanding
of agricultural zones, "agro-climatic zones", "ecological analogs" (the
concept of Nuttonson) -- in short, how, where and in what fashion can
agricultural systems be characterized and understood, what factors cause
zonation and are of greatest importance; (4) a systematic approach to
study of existing farms, without having to resort to conventional field
plot techniques, and an understanding of both strengths and weaknesses
of existing systems, as well as a basis for monitoring benefits or weak-
nesses of technological innovations and interventions; and (5) a basis
for multidisciplinary research.

The ecological methodology will be thoroughly studied in at least
one LDC (in Phase 2) at two or more levels of sophistication and
intensification. A key part of the project will be the development of
techniques for integrating vegetation ecological methodology with linear
programming techniques.

S The activity will be conducted in the lowland humid tropics, and
will emphasize tropical crops of special importance to small farmers, such
as tropical root crops, tree crops, and cereals. Systems which include
intercropping of perennial and annual crops such as tree crops and food
crops or intercropping of annual crops will be studied, for it is in
intercropping or mixed crop/livestock farming that the ecological approach
may prove of most benefit.'PMAt,,

Adaptation of the Methodology

Several levels of sophistication in farming systems methodology will
be required, in order to fit the various needs and capabilities of LCs.
Some countries have an extremely limited core of agricultural scientists;
for them a "minimum-level" team consisting of an agronomist and agricultural
economist, for example, neither with advanced degrees, may be all that
it is possible to mobilize. Operating on the assumption that some study
of existing farms and of possible innovations for them is better than no
program at all then it will be necessary to design methodologies
and approaches which will be manageable by such a team. The questions
here must then be what minimum data are needed to understand existing
farming systems, including farmer needs, how to gain an understanding of
potential innovations for that system, how can such data be gathered and
handled and, what supporting services may be needed to assist such a team?

For countries having a larger and better trained agricultural science
staff, more sophisticated and sensitive farming systems methodologies
will be needed. However, at present it is not known how the size and
composition of the teams will constrain project efforts; nor is it known
which methodological adaptation may be most useful or appropriate.

One area of special concern will be to find analytical instruments
suitable for LDC use, at differing levels of sophistication. Many
inexpensive, easily portable machines, especially small electronic
calculators and computers, are currently available which appear to offer
data handling opportunities to less sophisticated LDCs not available
just a few years ago. Farming system innovation requires rapid analysis
of the farming system and its components, even though sophisticated or
complex computer technology is not available. To use some of the less
expensive instruments in this research will require specific, careful
methodological planning proven by field evaluation of the techniques.
There will also be onppnrtinity for innovation in the training of R&D
workers outside the M.S.-Ph.D. smir

It may be fruitful to conduct paired studies in the same region,
in which team composition and methodology are the main variables, Also,
such comparisons should be made between geographical areas or varying
natural resource situations, For example, one contractor might work
in the South Pacific or Asia, another might work in the Andean region,
while still a third works in Africa. Comparing levels of sophistication,
lowland humid tropical conditions and highland Andean conditions, and
large geographic cultural areas should provide a good basis for comparing
the usefulness and efficiency of methodologies over a wide range of
natural resource and cultural conditions.

The search for wide adaptability of the methodology will continue
throughout the project. It will be handled in a sequence which will
develop and test "approximations" of methodology relevant for use. The
first approximation will be made in the state of the art study. It
will be largely conceptual or inductive in natures tested in the seminar-
workshop at the end of Phase 1. Succeeding approximations' however)
will be developed on the basis of empirical tests, under the range of
conditions noted above, and should be correspondingly more ussP"1

Establishment of the farming systems network is expected to be
useful in increasing the range of adaptability of the methodology. If
the project is successful in capturing the imagination of other MDC
farming system R&D projects and in arousing an interest among them in
methodology, not only will the improved methodologies resulting from
this project have a wider test but also the other projects will develop
methodological innovations. To the extent that happens,this project
will attempt to communicate the innovations.


1. Blaut, J. M. 1959. Microgeographic Sampling: A Quantitative
Approach to Regional Agricultural Geography. Econ.
Geography 35:79-88.

2. Dalrymple, Dana .G. 1971. Survey of Multiple Cropping in Less
Developed Nations. Foreign Economic Development Service,
USAID/AID, Washington, D. C. 20250. 108 pp.

3. Ensminger, Douglas. constraintss to Millions of Small Farmr in
Developing Countries Risking Changgs in Farming PractiLes
San .Family Living Patterns." The World Food Conference of
1976 Proceedingse owa stafe University Press, Ames, 1977.

4. Evenson, J. P., D. L.. Plucknett, and I. Horton. 1973. A Proposed
Classification for Agricultural Systems. Proc. 2nd Int. Symp.
on Tropical Root and Tuber Crops, College of Tropical
Agriculture, University of Hawaii, Honolulu, Vol. 11:63-69.

5. Francis, C. A. "Impact of New Technology on Small Farm Agriculture,"
Paper presented at ICRISAT Farming Systems Workshop, Hyderabad,
India, 1974.

6. Falcon, Walter P. "The Green Revolution: Generations of Problems."
American Journal of Agricultural Economics, December, 1970.

7. 1974. Resource Utilization Approach to Cropping
Systems Improvement. International Workshop on Farming
Systems. ICRISAT, Hyderabad, India. p. 249-260.

8. Hildetbrand, Peter E. "A Multi-Disciplinary Methodology for Developing
New Cropping Systems Technology for Traditional Agriculture."
ICIA report, Guatemala, no date.

9. International Crops Research Institute for the Semi-Arid Tropics
(ICRISAT). 1975. International Workshop on Farming Systems.
Nov. 18-21, 1974. Hyderabad, India. 547 pp.

10. International Rice Research Institute, 1975. Proceedings of the
Cropping Systems Workshop. Los Banos, Philippines. 396 pp.

11. 1975. Changes in Rice Farming in Selected
Areas of Asia. Los Banos, Philippines. 377 pp.

12. 1976. Annual Report for 1975.

Los Banos, Philippines. 479 pp.

13. Institute Interamericano de Ciencios Agricolas, Centro Inter-
americano de Documentacion e Informacion Agricola, 1974.
Bibliografia sobre sistemas de agriculture tropical.
Bibliografia No. 27. Turrialba, Costa Rica. 145 pp.

14. National Academy of Sciences National Research Council. 1962.
Problems and Techniques in Range Research. NAS, Washington, D.C.
341 pp.

15. 1976. World Food and Nutrition Study --
"Farming Systems", a panel report (D. L. Plucknett; E. Oyer,
F. Viets, J. Vicente-Chandler and L. Martin). Unpublished

16. Ruthenberg, H. 1971. Farming Systems in the Tropics. Clarendon
Press, Oxford.

17. Spedding, C. R. W., and N. R. Brockington, 1976. Experimentation
in Agricultural Systems. Agricultural Systems 1(1):47-56.

Part 3, C. Project Design and Work Plan

The project will be implemented in two phases and will involve
multidisciplinary efforts in both. The first can be called state-of-
the art analysis or methodology inventory and evaluation. The other
can be called methodology testing and adaptation. Each will result
in (1) handbook or manual type publications designed specifically
as guidelines for R&D personnel of the LDC's, plus (2) training sessions
for those personnel. Other publications that may be indicated at the end
of each phase will be considered optional.

Specific Objectives

Phase 1 will seek to achieve the following detailed objectives.

1. A description and analysis of the relevant methodologies
in use by current multiple cropping and farming systems
programs for tDC4

2. Identification and evaluation of methodologies of conventional
disciplines for their probable utility in R&D efforts in
farming systems in the LDCs.

3. Integration of the findings from objectives 1 and 2.

4. Publication of the product of objective 3 in a handbook
or manual format.

5. Training of R&D personnel teams from three LDCs in one or
more LDC sites on findings packages in the manuals.

6. A preliminary finding regarding the desirability of each
of several LDCs for field test activities.

7. A design of and plans for Phase 2.

8. Establishment of an international network of R&D personnel
working in farming systems, including mainly MDC insti-
tutions but also the more advanced LDC programs.

Plan of Work

In order to accomplish these detailed objectives, the following
activities will likely be necessary. They are subject to change, however,
as new knowledge becomes available and as interaction among contractors

Activity 1. Contractors will seminar for the purpose of
conceptualizing the inventory and analysis task and for planning the
field visits; i.e., deciding on the countries to visit and the makeup
of the teams to visit them and designing a standardized observational
schedule so that the data collected can be treated in a similar manner.
The seminar will be preceded by preparatory staff work. This seminar
will also plan other details of the plan of work of the first phase,
including ways to establish linkages with others interested in the
field, ways to keep contact and coordination among the contractors,
the relevant literature for detailed review, and draft outline of

At least three persons from each contracting institution, the A.I.D.
project manager, and consultants as needed will attend this seminar.

This activity will be relevant to all objectives, especially
1, 2, 3, and 6.

Activity 2. Field visits will be made. Each contractor will
have responsibility for one large geographic area (Asia, Latin America,
or Africa) and probably for a certain area of subject matter, such as
system evaluation, data improvement, and relating with the farmer.
In Phase 1, however, the work load will be evened out by an exchange
of personnel among contractors as a means of improving coordination
of contractors and integration of results. This activity is mainly
relevant to objectives 1, 6, and 8. These visits will include ongoing
MDC projects as well as LDC programs, some of which will be candidates
for test sites.

Much of the effort during the first year will be devoted to sur-
veying methodology being used in farming systems R&D efforts. This work
must be done in concert with other institutions conducting research
in farming systems. To complete this task will require written and
direct contact with research institutions around the world. Determined
efforts to establish the R&D network will be made along with the survey
to take advantage of the contacts.

Visits will be made to IRRI, ICRISAT, IITA, CIAT and ILCA to
gain an understanding of research methods being used by international
centers. Regional or national organizations will also be surveyed
to obtain an understanding of the research approaches and methods being
used in current and past programs. Key programs and organizations in
LDCs and MDCs will include: CATIE; PCARR (Phil.); WARDA; the all-
coordinated Dryland Farming program in India;-the CSU-Pakistan Water
Management Program; ICARDA programs in the Middle East; the Coconut
Research Board of Sri Lanka; national programs in Guatemala (ICTA),
Nicaragua (INVIERNO), Indonesia (CRIA), Korea, El Salvador and the

Dominican Republic; the BARTAD project in the Bahamas; South Pacific
Commission; Alafua Agricultural College, Western Samoa; the University
of the South Pacific, Fiji; the Yurinaguas research program (NCSU) in
Peru; and EMBRAPA programs in Brazil.

More developed countries which must be contacted concerning their
experiences include Canada, France, the United Kingdom, Belgium, The
Netherlands, Australia, Mexico, and Taiwan. The SOTA work will be
accomplished by literature reviews and visits to existing programs,
followed by a period of analysis and synthesis to compare the effective-
ness and usefulness of the methodology; (1) to identify key problems
facing LDC farmers, (2) to serve as a basis for technology innovation,
(3) to serve as a basis for technology transfer, (4) to serve as an
appropriate tool for LDCs where resources and trained personnel are -
limited, (5) to provide a basis for involvement of the farmer in
technology innovation, (6) to provide a basis for studies of intensi-
fication on small farms, and (7) to provide a basis for multidisciplinary
research involving biological, physical and social scientists. In
other words, methodologies used will be reviewed and analyzed for
suitability to serve complex farming systems research needs as well
as for appropriateness for LDC institutions and e farm-ig omuniuie
which they serve.

Activity 3. Concurrently with Activity 2, an inventory and evaluation
of methodology from related fields will be made, largely from literature,
but by using consultants and possibly visits in connection with the field
visits listed above. (Objective 2).

Activity 4. Coordination meetings will be held at least every
six months involving the principal investigators and the A.I.D. project
manager. To the extent feasible, project meetings will be held on
one of the contractor campuses, and advantage will be taken of the
visiting personnel, through seminars and conferences, to share experience,
knowledge, and insights for the purpose of deepening the interest of
the contractor faculties in the project. (All objectives).

Activity 5. Upon termination of the field visits, draft manuals
will be prepared by the staffs of the contractors. The organization of
this exercise will be decided by the contractors and A.I.D. project
manager before the end of the field visit activity, when much more
information is available than currently. Preliminary plans for this
activity will be made in the initial seminar, but they too will likely
need modification in light of information and experience gained.
(Objectives 3 and 4).

Activity 6. The drafts of the manuals will be reviewed, revised,
and completed with the help of a seminar-workshop to be attended by
contractor personnel, A.I.D. project manager, project consultants,
and experts from other MDC (or donor) entities who have an interest
and capacity in farming systems and who have been in contact with
the project previously. Some attention in this seminar-workshop will
also be given to the preliminary design of the next phase. (Objectives
4 and 7).

The main output of the workshop will be to recommend a limited
number of major methodologies for testing and use in LDCs. These
recommended methodologies will be considered as "approximations"
for trial and may not be considered as final products, but as candidates
for further study. Handbooks and workbooks will be prepared for -
each methodology-including methods of analysis and steps to follow
in laying out and conducting the research. These handbooks will be
geared for use in LDCs.

The workshop will also help in developing a formal network for
and in improving linkages in farming systems. Also, it will
serve as a focus for negotiations on locations and targets for collaborative
research by the AID-funded farming systems methodology consortium
partners. In a way, the workshop will serve as a design workshop for
the succeeding research program. By doing this, the resulting research
program should be greatly improved.

The SOTA studies will also result in a series of technical
analyses of the various research methodologies being used. Such
comparisons and analyses have not been conducted; however, they should
be of major importance and benefit to inform existing and contemplated
programs as to their adequacy and suitability for use in future.

Activity 7. Publication of the manual will follow immediately
after the. seminar-workshop. For the most part this will be an editing
and printing exercise. (Objective 4.)

Activity 8. Concurrently with the editing and printing of the
manual, contractors will complete the design of the second phase. This
will be an iterative process that involves contractors, all relevant
entities of A.I.D., and LDC institutions. (Objective 7).

Activity 9. Concurrently with Activities 7 and 8, the project
will be evaluated and steps will be taken for approval of second phase
and allocation of funds. (Project management).

Activity 10. Training will be implemented, taking one of several
possible forms. It may be a single session with teams of farming
system R&D personnel from several LDCs, or it may be held in three
sessions, with only one country involved per session. Sessions may
be held in countries which are candidates for the siting of Phase 2
activities. (Objective 5).

Part 3, D. Environmental Impact

The intent of this project is to stimulate the adaptation and
testing of methodologies for technology innovation for the small farm
operator who is under severe resource constraints of all kinds, including
purchased inputs. In order to realize this intent, it will be necessary
to aim for technologies that conserve chemical fertilizers, pesticides,
and herbicides, resulting in better use of those purchased and even
a reduction in purchases. This project will be operating in small,
controlled plots. If the project is successful and the technology
resulting from it is used on a broad scale, any impact on the environ-
ment will be positive. Not only will waste of chemicals be reduced,
but other environmental hazards such as deforestation and soil erosion
will be reduced. More food will be produced on less land, and pressure
will be reduced on lands that are marginal for agricultural produ-.tion.

Part 3, E. Impact of Project on Role of Women

It is difficult to predict the impact of this project on the role
of women, because we do not know some important relationships. Logic
and a few data regarding relationships indicate that the project will
be neutral to positive. One of the positive relationships is that
between women in the labor force and family income. Some studies have
indicated that if family income rises to a certain point women feel less
need to work and voluntarily withdraw from the labor force. That
must be considered a quality of life improvement, and to the extent that
this project helps raise incomes, it may have that effect, at least in
certain cultures and certain situations. Even if it does not relieve
the women of heavy farm work, the increase in family income would also
be considered plus, but that is an advantage gained by the male as well
and counts only as neutral in this analysis. The project will also
contribute to the development of a management capability among small
farm operators and their families. The extent to which women will
benefit from thigh rr-p'4"h"'t c4q a function or the culture and not of
this project.

Project Initiation,
Planning Seminar 1,2,3,6

Field Survey 1, 6

Related Literature 2

Prepare Draft
Manuals '

Coordination All

Review, Revise
Manual Preliminary '
Design Phase 2

Publish Manual' 4

Design Field Test
Evaluate Phase I

Prepare, Hold
Training Sessions


Month from Beginning of Project

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24


What How Who When

Design of Field Testing, In Phase I Contractors 3 6 months
Selection of Country Seminar and Consultants before Phase II
Candidates Collaborative Work A.I.D. starts

(About July 1979)

Final Selection of Visits to Candidate Contractors, Immediately upon
Test Sites and Countries; Consultations LDC's personnel project approval
Negotiation with Countries and USAIDs, A.I.D.
Missions Regional Bureaus Month 1
Program Manager (Aug. 1979)

Implementation of On-farm work in collabora- Contractors,
Country Tests tion with LDC institution, Consultants, Months 3 36
plus seminars and con- primary and
ferences among contractors secondary
to induce integration collaborators

Revise Manuals, Seminar Workshop of Contractor, Months 24 33
Prepare Training' experts, consultants Consultants,
Materials working with draft Collaborating (Aug. '81 to May '82)
materials prepared by experts

Hold Training Session

For 3 LDCs in each
site country

Contract personnel
and collaborators

Months 30 36

(Feb July '82)

Part 3, F. Estimated Budget and Life-of-Project Costs
(in 000's)
1 2 3 4 5 TOTAL

1. Salaries
(108 person months)- 200 200 -- -

2. Fringe Benefits 33 33 -

3. Travel 26 26 --

4. Other Direct Costs 16 16 -

a. Supplies (5) (5)

b. Computer (3) (3)

c. Communications (3) (3)

d. Printing (2) (2)

e. Miscellaneous (3) (3)

5. Overhead 100 100

TOTAL 375 375 1,000 1,000 1,000 3,750

1/Including clerical and varying levels of professional talent.
- Including clerical and varying levels of professional talent.

Part 4. Implementation


There are some general considerations regarding design of the

1. The design of the first phase accommodates a certain amount
of flexibility. It provides plenty of time for each activity, perhaps
more than is needed. If the phase can be implemented at a rate more
rapid than programmed here, or if test site cropping season necessitates,
activities can be accelerated and Bhase 2 will begin earlier. Funds
for Phase 2 are requested for FY 1979, with plans being for expenditure
to begin in FY 1980. No reason is foreseen that those funds could not
be expended starting in FY 1979. It is planned to bring the project
back for a full review before further commitment of funds.

2. The design presented is provisional, to be authenticated with
needed modification with the project manager.

Contractor Coordination

A major consideration in the design and implementation of this
project is the division of functions or responsibilities among contractors
and in turn the putting of the pieces back together to accomplish an
acceptable level of project integrity. Efforts are being made to employ
a technical specialist within TA/AGR on an IPA contract for this task.
Until that is realized, heavy reliance will be placed on one or more
project consultants. Efforts will also be made to have the consultants
in contact with a range of LDC projects, ideally as consultants to them,
so that a well-grounded knowledge of LDC conditions is part of his
expertise in consultation to the project. He will also consult on other
TA/AGR projects involving system technology.

The project manager will also be responsible for seeing to it
that the principal investigators have regular interaction and as appropriate
will attend their meetings. Attempts will be made to develop the dynamics
of the group of personnel on the project. Such elements of project
design as the joint authorship of the manuals and handbooks will contribute
to that end.

Test Site Selection

The contractors will be gathering information, according to a
structured criteria list, on possible sites for the testing and adaptation
phase. However, selection and negotiation will involve A.I.D., especially
the missions, and LDCs, as well as the contractors.

The-possibility is held open for two site categories. One
site category will involve a team in residence working along with
the LDC institution. This will be a collaborative operation to the
fullest extent that it can be. Ideally, this project will join forces
with LDC institutions already engaged in farming systems work but
not so far advanced that other LDCs could not identify with them or
with those who are seriously interested in initiating a program
independent of the influence of this project.

The second category of site would be one in which the resident
team made periodic visits for monitoring and consulting. This would
likely be a program that is fairly well along, such as the CATIE
program will be by Phase 2, which may be interested in keeping abreast
of the work in this project. The field visits of phase 1 can be expected
to stir up some interest among LDC institutions, and some of those not
selected as a primary site might well want to collaborate as secondary


Strong linkages are expected to develop between the contractor
group for this project and other non-LDC entities engaged in R&D work
and even among the other entities. Given the empirical nature of
-farming systems, the substantive orientation of the programs, and the
site specificity of most systems, communication among farming systems
program has not been extensive, even among a group as closely knit as
the international agricultural research center (as noted below). However,
a focus on methodology will make it easier to develop linkages.
Methodology can be made transferable. Even though the solutions may
be site specific, the processes for achieving them are not so site-bound.

Through a fortuitous set of circumstances, the project may be
greatly facilitated in its efforts to effect linkages with other entities.
The Technical Advisory Committee (TAC) of the Consultative Group of
International Agricultural Research (CGIAR) is staging a review of
farming systems work of at least four centers. A professor of the
University of Hawaii, from whom a proposal has been received, is on the
TAC review team. In this exercise he will be in contact with the
Centers and with other entities who have much to contribute to this
project and much to gain from an improved farming systems R&D methodology.

Plans for Utilization

In one sense this is a utilization project in that it aims to
mobilize relevant knowledge and technology, process it, and put it in
a form for utilization. The actual training that is done in this

project is a form of utilization, of course, but a follow-on project
will likely be needed for adequate utilization. However, at the end of
the project handbooks and manuals will be available, as well as a
cadre of 12 or so U.S. professional personnel from three institutions
who will be available for technical assistance to LDC institutions.
It can also be anticipated that other MDC institutions engaged in
farming systems R&D work will have had a part in the synthesis of the
methodologies this project aspires to present and that they will also
make use of the methodology, even as they continue adapting and refining

Evaluation Plan

The evaluation of this project is expected to be simple. There
are clearly marked target dates for the completion of specific activities
which in turn are related to specific objectives or outputs. In a
sense, the handbooks and the training sessions are in themselves an
evaluation of Phase 1. Observers, constituting an A.I.D. evaluation
team, will monitor the training sessions and analyze the training
materials, including handbooks,as the final evaluation.

However, it may be necessary to move quickly into Phase 2 in order
to fit into the growing season that prevails at the test site. In order
to make this possible, evaluation by the project manager will be
continuous, and approval documents for the second phase will be prepared
as far in advance of Phase 1 termination as appears feasible.

Evaluation of Phase 2 will be a little more difficult, since much
more is expected of it. Ideally, it should be evaluated by what happens
on farms. To a certain extent this will be possible by a survey of
the area which the testing site typifies. However, that sample will
be defective, and the survey will be more useful for the insights it
provides than for its representativeness. Still a skillful interpretation
of these results will be one input into the evaluation. Reliance will
also be placed on an analysis of the documents produced, an evaluation
of the training sessions, and periodic reviews of the project in progress,
with particular attention to the performance of the LDC personnel.
In Phase 2, the possibility of a secondary site in an LDC different
from the primary site will be considered. If there is a secondary site,
the evolution of the farming systems program at that site will also be
taken into account for evaluation. One of the important elements of
evaluation is the continuous surveillance which will be maintained by
the A.I.D. project manager and by a technical consultant who enjoys
the confidence of the project manager.

Selecting Contractors .

Unsolicited proposals have been received from two contractors.
Their interests in farming systems are similar enough that integration
of efforts appears to be achievable. At the same time. each has a
special subject matter intern whi-h it proposes to emphasize while
attending to the other needs of the project. The two proposals are
compatible in that one proposes work in Asia and the other in Latin

Currently, no proposal has been received from a potential contractor
willing to work in the African area. We intend to seek proposals,
through conventional A.I.D. procedures, and hope to have a contractor
before the beginning of Phase 1. If such is not possible, the two
who submitted unsolicited proposals will be requested to carry a bigger
load in the worldwide methodology inventory and evaluation phase, and
efforts will continue to find a contractor to take geographic responsi-
bility for the African region for Phase 2p la5us-es.aSdi d t Wt.

Relationship with Other Projects S.L. 'iA.

This project will draw heavily from the outputs of other projects,
for two reasons. The first reason is that essence of farming systems
technology is the integration of other pieces of technology with cultural
and economic forces into a purposeful human endeavor or enterprise.
For this reason, it can be expected that attention to farming systems
will identify a need for other R&D projects of A.I.D., either in the
central program or in mission programs. Several of the current projects
are specifically relevant to small farm agriculture. One is the small
farm machinery project of IRRI. Another is the nitrogen fixation project
which aims at developing means by which the farmer can take nitrogen
from the air, supply unlimited, for use in crop production. No matter
how much the cost of manufacturing nitrogen fertilizer can be reduced,
it will be an expensive commodity by the time it reaches the fields of
many of the world's small farmers. The neglect of nitrogen fixation
by the world's technological community with the advent of cheap cnmmorial
nitrogen one e most dramatic examples of how R&D has been biased
away from the small farmer and how R&D institutions of the LDCs have._
been influenced by the R&D interests of theMCs who reacted correctly
t5 the "scarcity-abundance" rel--onships of their clientele..

The second reason for the close relationship to other projects
is that it holds some purposes in common with them. For example, the
on-farm water management project of Colorado State in Pakistan has
developed methodologies similar to ones this project is concerned
with, especially those of knowing the farmer and identifying his problems.
A project to be initiated in 1978, Technology Innovation Management, aims
to strengthen national R&D systems.



Project Goal

To increase income and
welfare of the poor
majority in the LDCs



1. Farm income,
average and

2. Nutritional Level

Means of Verification

1. Census data over the
long run.

2. Special surveys,
particularly to supply data
on the poor majority.

3. Educational Level
and literacy rate

4. Health


1. Goal will be
achieved in long run
and by contributions
of many projects and

2. Measures listed
will not be able to
identify contributions
of individual projects.

3. Means used to
achieve goal will
include activities
designed to (1) increase
production so that
total supply of food
will be increased and
(2) increase efficiency
and lower input costs
so that farm profit-
ability does not have
to come from higher
farm prices except in
cases to correct obvious

Project Purpose

To improve the capability
of agricultural R&D sys-
tems of LDCs to generate
and diffuse technology
innovations in intensive
production systems for
small farm agriculture
that increase food pro-
duction and farm income.

End of Project Status

1. Farming System R&D Metho-
dology used on any appreciable
scale anywhere in the world will
be organized, codified, and
packaged in a form useful to any
LDC to which it is relevant.

2. Every LDC that has an
interest in or an active R&D
project in farming systems will
have a collection of manuals
and handbooks on farming systems
R&D methodology.

3. A dozen LDCs each will have
had at least three of its farming
systems R&D personnel trained in
R&D methodology.

4. Three LDCs will have farming
systems R&D projects that meet
criteria for adequate performance
for this type of project.

5. A cadre of U.S. technicians
will be available for field sup-
port to missions.

Means of Verification

1. Experience of AID
contractors as reported
in annual reports and
other means.

2. Observations of AID

3. Formal evaluation of
project, if decided to be


1. This is a five-year

2. Other projects will
contribute to purpose.

3. The current interest in
farming systems is not a
passing fad.

4. "Systems technology" is
biased toward the small
farmer, since intensive
systems do not perform
well with mechanization and
hired labor.

5. R&D systems of the LDCs
are under varying degrees of
resource constraints,
necessitating a range of
R&D methodologies.

6. Current projects in
farming systems are not in
condition to provide
technical assistance in
methodology to R&D systems
in other countries.

7. There will be a sus-
tained interest in AID
and other donors to assist
LDCs in the development of
their agricultural R&D



Two primary outputs are expected
from this project.

1. A set of integrated, multi-
disciplinary farming system R&D
methodologies adapted to man-
power and financial constraints
of the LDCs, packaged for easy
delivery to LDC institutions.

2. Training in the methodology
for LDC personnel engaged in
farming system work.

These methodologies will pertain
a. Evaluation of farming systems
(incl. problem identification and
data needed).

b. Collection and analysis of
small-farm resource data, (incl.
social-cultural factors, insti-
tutional services, ecological
conditions, and economic

c. Collection, generation, and
analysis of data on technical
coefficients, emphasizing
efficiency and transferability.

d. Generation and identification
of alternative systems.

Objective Verifiable Indicators

1. Existence of a set of
manuals and other documents
that organizes current know-
ledge of farming systems R&D

2. Existence of a set of
manuals and other documents
that complement the original
material (or revise it) based
on experience in testing and

3. Numbers of LDC personnel
trained by all types of

Means of Verification

Indicators can be
verified by inspec-
tion of contractor
reports, published
material from the
project and Agency
R&D portfolio.


1. Three systems are
relevant to this project:
the household-farm
system, the production
system, and the multi-
farm system.

2. The production system
is the chief unit of
analysis. The household-
farm system will provide
data to the analysis. The
multi-farm system will be
important in diffusion and

3. This project must come
to terms with the LDC
entity for agricultural
R&D, which in turn must
come to terms with the

4. Effective methodologies
for farming systems work
exists in specific pro-
grams around the world.
Further methodologies
from certain other subject
matters can be adapted to
farming systems work.

e. Wholesale delivery of technology
and feedback from users.

Objective Verifiable Indicators

Three secondary outputs or
by-products are expected.

1. One or a few improved
systems for specific areas
of countries in which
testing and adaptive work
is carried out.

2. Some urgent problems
identified that need attention
is conventional R&D programs,
both for countries and for
international R&D programs.

4. A series of simple publications
explaining technologies and their
applications which have demonstrated
adaptability over a wide range of

5. R&D project statements or proposals
aimed at addressing problems identified
in the project.

5. Most LDCs have
various organizations
and media which are in
position to inform the
farmer of improved tech-
nology and instruct him
in its use, but in few
cases is there an ef-
fective activity for
packaging information for
these entities and
instructing them in its

6. Although farming
systems technology is
highly site specific,
there is a potential for
considerable trans-
ferability if proper
variables can be
identified and understood.

Means of Verification




Implementation Targets

1. AID funding and

2. Documents, time and
collaboration of MDC
entities working in
farming systems.

3. 66 months of
time of professional
personnel from three
U.S. universities.

Phase 2

1. About 30 years of
professional time of
U.S. personnel and
the same or more of
LDC personnel.

2. AID funding and

3. Collaboration of
MDC personnel from
other entities.

1. Initial seminar.

2. Completion of field visits
and literature search.

3. Synthesis of first approxi-
mation methodology.

4. Verification seminar-workshop.

5. Publication of manuals.

6. Training of LDC personnel
Phase 2

1. Test site selection and

2. Field test implementation.

3. Revision of manuals.

4. Training of LDC personnel.

Means of Verification

1. Project records,
annual reports, and


1. This effort on the
part of AID will tend
to focus attention of
other IDC entities on
the methodological
issues of farming
system R&D6

2. *rhe current interest
P farming systems will
no out to be a
*pfJn _ffal .




Proposed phases of the research

The subgroup proposes that the research can be done in four phases which will
usually be sequential but not necessarily so. The phases are:

Phase I. Identification of Target Systems

The first phase involves the collection, collation and synthesis of as much
pertinent information as feasible on: 1) the natural resource base for
agriculture; 2) the current status of agricultural exploitation of these
resources; 3) identification of apparent constraints and opportunities for
betterment; 4) selection of candidate farming systems that seem worthy of
study; 5) "windshield" surveys of potential target areas; and 6) selection of
transects for Phase II. Most of this data will come from secondary sources.

Phase II. Conducting Survey Transects

Survey transects are established to collect data on existing farming systems,
the management of farms, the customs and traditions of the people, and a
more in-depth examination of the resource base. The survey will be conducted
by a multidisciplinary team. The farm (the farmer and his household) is the
sampling unit. Survey transect studies will identify priorities and representative
farms for a more detailed examination in Phase III. Data on household behavior
and human resources on the farm should be collected in this phase as well as
data on the physical and biological character of the farm.

Phase III. Conducting Detailed Transects

Detailed transect studies will evaluate important components of the farming
system which were identified in Phase II. Here the farm field is the sampling
unit. Examples of a detailed transect might be a study of disease incidence
or spread in a field, or competition between two crops in an intercrop system.
Such detailed studies should be done by specialists in a number of fields.
Phase III will identify high priority research needs for Phase IV. Note the
increase in number and specificity of disciplines likely to be involved as one
moves from Phase I to Phase IV.

*Adapted from: World Food and Nutrition Study Team 4a report, National
Academy of Sciences

A- 2

Phase IV. Derived Research Needs

Most Phase IV studies could be termed derived research%, in that problems
identified have been derived from a systematic study of the farming system and
its needs. Some Phase IV studies might originate from new biological and
technological opportunities, but results from these should be suitable for
ready transfer to the farm, for Phases II and III have characterized and analyzed
the farmer and his needs, as well as the biophysical and socioeconomic environ-
ment of the farm itself.

An indication of the study team's assessment of the activities, outputs and
potential benefits of each phase of this ecological approach to farming systems
research is presented in outline form as follows:


Purpose: Use what is known to make better selection of target systems and to
improve research planning.


1) Identification of priority systems and initial transects.

2) Identification of additional data needed.

3) Identification of needs for information systems and capacity to gather

4) Identification of multidisciplinary team.

Examples of activities:

1) Select priority production systems. (See 8 questions on page 6.)

2) Using existing information on soils, climate, agricultural and socio-
economic status to establish baselines.

3) Windshield surveys "sensory" -data collection.

Follow-up: Interpretation, synthesis and evaluation of information: The
collected information will be interpreted to determine how it can be used to
plan farming systems research. Such information should identify:

1) major topographical units and changes,


2) rainfall boundaries,

3) soil boundaries$

4) probable cropping system cellss'

5) major economic factors which affect the farmer, and

6) tribal boundaries and customs#

Follow-up: Intervention, utilization and transfer capability:

1) Intervention possibilities will be small.

2) Utilization and transfer likely to be restricted to information gathered
which will then be available for wider use.

3) A major benefit should be an increased awareness of available information
on a country or a region.


Purpose: Understanding the farm (the farmer and the household) as a unit
(basically qualitatively), its characteristics and elements common with
other systems.

Outputs: ______ ....

1) Description and characterization of representative farms and repre-
sentative systems.

2) Identification of targets of opportunity (both problems and opportunities)
for detailed study in Phase III.

3) State-of-the-art studies on existing practices.

Follow-up: Intervention, utilization and transfer capability:

1) Intervention possibilities would usually be small, except in cases
where policy issues are a major problem.

2) Opportunities to transfer simple, innovative farming practices may be
good. State-of-the-art publications should assist in the transfer and adoption
of such information in other areas more rapidly.


Purpose: Understanding the factors and components affecting the farm
where the field is the sampling unit for biological, physical, economic
and social measurements. Put another way, measuring the fit of the
system to the environment and assessing the potential of improving the


1) Recommendations for high priority research to solve real problems.

2) Recommendations for change in the system.

3) The appropriate tool to monitor and measure changes on the farm with

Examples of activities: Detailed measurement of:

1) water management and irrigation use and potential,

2) crop competition and varietal interactions,

3) role and use of livestock, sources of feed,

4) pest ecology, and

5) socioeconomic analysis-tribal customs, populations, labor supply,
risk factors, infrastructure and institutions available.

Follow-up: Interpretation, synthesis and evaluation of information.

1) Multi-variate analysis can be used in analyzing and interpreting
socioeconomic as well as biological and physical data.

2) Detailed transects in farmers' fields will provide a basis for
measuring and evaluating the stability of existing farming systems.

3) Components of systems in the field, including measurements of impact
of new technology and analysis of the different sources of risk can be
most effectively evaluated here.

Follow-up: Intervention, utilization and transfer capability.

1) Intervention possibilities are great. Detailed transects can
evaluate the dynamics and impact of the components of a farming system.
Some intervention opportunities will be quite clear after these studies are

2) Utilization and transfer opportunities will also be great; research
identified will be of high priority and specific for the farming system



Purpose: Conduct problem-oriented, action-oriented studies which will
support improvements in existing farming systems, or work out the components
for improved systems.


1) Systems interventions.

2) Better adapted varieties suitable for specific systems,

3) Methodology for the testing and adaptation of new technology on the farm.

4) Fertilizer, pest control, water management, etc., recommendations.

5) Innovations on new systems.

Examples of activities:

1) Research on components of systems.

2) Developing technology specific for existing or improved farming systems.

3) Systematic study of physical, biological, and socioeconomic needs (MAJOR

4) Identifying and implementing roles of institutions at the country, region
and international levels.

5) Training of researchers, technicians and farmers.

6) Synthesis and evaluation of improved farming systems.

7) Economic and social research on such problems as: a) functioning of
irrigation systems; b) input and output markets; c) market demand; d) credit;
e) land tenure; f) transportation; g) educational institutions; h) health

Folow-up: Research and its results in this phase will be:

1) focused on the farmer and his system,

2) multidisciplinary,

3) action-oriented, and

4) strongly focused on implementation of ideas and new technology, the
need for which is identified in the preceding phases.

A- 6

Follow-up: Intervention, utilization and transfer capability.

1) Intervention opportunity great because research results have been
focused on the farmer and the action-oriented research approach constantly
involves the farmer in the systems Intervention planned and implemented can
be monitored by on-going Phase III detailed transects,.

2) Utilization and transfer will be strengthened, especially if the
systems research is tied into existing soil and land classification
systems and related or adapts to varying socioeconomic situations.
A longer time frame (Z to 5 years) for completion.


If one agrees that the four phases of the ecological approach to farming
systems research provide a valid framework for improvement of farming systems,
it is necessary to insure that the results of the research accomplished will be
applied through some form of "extension". If one finds a great variation in
research capabilities in developing countries, the "extension" or delivery"
systems are even more variable in quality.

It is assumed that some mechanism fo transmitting new research information
to farmers is needed and that such a system should be reversible to transmit
needs and problems that exist at the farm level back to the researchers.
Without suggesting an organizational framework for such a system, several
important components will be identified.

After new information is developed, it must be interpreted in light of
existing situations and practices so that it can be an integral part of the
synthesis of an improved farming system. After synthesis, the new technology
must be evaluated in the real-life farming system. These three tasks of
interpretation, synthesis and evaluation can be conducted by the researcher
or by the researcher and farmer working as a team. However, for the new
technology to receive broad application, it will need to be transmitted by a
large group of "technicians" or "change agents" who can implement its utilization
by a much larger number of farmers. Thus, the final steps in the adaptation
of the new technology can be termed intervention to stimulate change in present
practices to utilize the new technology.

These important steps can be achieved in many ways, but certainly dedicated and
committed people and institutions will be required. It is essential that the
informational flow be in two directions: 1) from the researcher to the farmer;
and 2) from the farmer to the researcher. The latter process should communicate
perceived or realized problems with the new technology or needs for other

A diagram of such an information flow system through the four phases of the
research process is illustrated in Figure 1. There is connection and inter-
action between the last three phases of research and the processes of inter-
vention and utilizati by the farmer* These processes are interconnected
and information canhould flow in each direction.

An example of the cost of the application of this methodology to a simple
system would be one in which there was little variation in soil and
climate over the transect, In this case, assuming secondary data was
available, Phases I, II and III might be accomplished with as little as
2 to 6 scientist years (SY). In a more complex system, where a transect
might cross,wide variations in topography, soils, climate and social customs,
from 20 to 30 SY could easily be consumed in accomplishing Phases I through
III. Phase IV, derived research, itson-going and its extent will depend not
only on the complexity of the problem but also the need and resources available
for its solution. Examples of the cost, time and complexity of Phase IV can
be taken from agricultural universities, national organizations and international
research centers whose missions are to conduct agricultural research.

The successful implementation of such research and delivery systems will
depend on the priorities of governments and concerns and commitment of the
numerous participants in the process. It is expected that the public research
and educational institutions of a country would be involved in all phases of
the approach. Training of more researchers in an increasing number of disciplines
will probably be necessary. Whether this is done in a university or government
agency will depend on the research organization of the country, but it should
be emphasized that this approach is a participatory one. In a developing
country, it can involve expatriate and local scientists; some participants could
come from universities in developed countries, some from international research
centers and some from local institutions. The expected result would be the
self-sufficiency of the developing country to conduct its own farming systems
research program.

The time required to implement this research methodology will vary with the
stage of development of the country where it is applied. For example, if
institution-building to provide basic education has to occur before implementation,
at least a generation (20 years) will be required before there would be any
hope of self-sufficiency of research effort in a developing country. Shorter
periods of time will be required if part of the educational and research
infrastructure are in place.

When one critically evaluates the operational mode of much "modern" agricultural
research, it is quite obvious that Phases II and III are often ommited and the
researcher moves from Phase I (literature review) to Phase IV (laboratory
or field plot experiments). The farmer is really not utilized in the planning,
execution or evaluation of research which is being done on his behalf. Perhaps
this has resulted from our"modern" energy-intensive monoculture agriculture.
The derived researchprocess is expensive. Experienced, highly trained
specialists are needed at all levels of the research and intervention process,

Figure 1. Basic Operation of the Ecological Research Approach

Survey Transects


Problem Identification, Ideas for
Research Priorities
Assessment of Impact
| &,Need for Adjustment

Conventional Researc

Trial on








Heavy involve-
ment in


Work o
I '-1
I j



Detailed Transects
a. Studies on lb. Monitoring of
Phase II I Improvements &
Targets | Adjustments in
the system
---- -- ---.



Compare this situation with the one described in the preceding model where
modest-sized multidisciplinary research teams function in each of the four
phases, but two of the four phases employ the farmer as a partner and his
resources as a laboratory. Even in Phase IV where derived research is
conducted, junior partners are being trained to make the entire cycle
self-perpetuating and adapted to the facilities of the local institutions.