Farm Management For Asia System

Farm Management for Asia: a Systems
Approach. (FAO Farm Systems Management

Series - 13)
Table of Contents
Douglas J. McConnell
John L. Dillon
Department of Agricultural and Resource Economics
University of New England
Armidale, New South Wales
Australia
FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS
Rome, 1997
The designations employed and the presentation of material in this publication do not imply the
expression of any opinion whatsoever on the part of the Food and Agriculture Organization of the
United Nations concerning the legal status of any country, territory, city or area or of its authorities,
or concerning the delimitation of its frontiers or boundaries.
M-61
ISBN 92-5-104077-X
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or
transmitted in any form or by any means, electronic, mechanical, photocopying or otherwise, without
the prior permission of the copyright owner. Applications for such permission, with a statement of the
purpose and extent of the reproduction, should be addressed to the Director, Information Division,
Food and Agriculture Organization of the United Nations, Viale delle Terme di Caracalla, 00100
Rome, Italy.
© FAO 1997
This electronic document has been scanned using optical character recognition (OCR) software
and careful manual recorrection. Even if the quality of digitalisation is high, the FAO declines all
responsibility for any discrepancies that may exist between the present document and its original
printed version.
Table of Contents
ACKNOWLEDGEMENTS AND DEDICATION

FOREWORD
ABBREVIATIONS, ACRONYMS AND SYMBOLS USED
1. AGRICULTURAL AND FARM SYSTEMS - CONCEPTS AND DEFINITIONS
1.1 SYSTEM DEFINITION AND HIERARCHY
1.2 GENERAL SYSTEMS CLASSIFICATION
1.2.1 Natural, social and artificial systems
1.2.2 Further sub-classification of systems
1.3 AGRICULTURAL SYSTEMS CLASSIFICATION AND ORDER HIERARCHY
1.3.1 Nature of farm-level systems
1.3.2 Village-level farming systems
1.4 STRUCTURAL ELEMENTS OF THE FARM-HOUSEHOLD SYSTEM
1.5 STRUCTURAL MODEL OF A FARM-HOUSEHOLD SYSTEM
1.6 REFERENCES
2. FARM MANAGEMENT AND FARM TYPES
2.1 FARM MANAGEMENT
2.1.1 Scope
2.1.2 Definition
2.1.3 Optimization
2.1.4 Objectives
2.1.5 Economics as the framework for farm-system analysis
2.1.6 Alternative bases for farm-system analysis
2.1.7 Farm management fields
2.1.8 Farm management modes
2.2 FARM TYPES AND STRUCTURE
2.2.1 Farm types
2.2.2 Structure of small-farm systems
2.3 REFERENCES
3. ELEMENTS OF FARM-HOUSEHOLD SYSTEMS: BOUNDARIES, HOUSEHOLD AND
RESOURCES
3.1 FARM SYSTEM BOUNDARY
3.1.1 Importance of boundary specification
3.2 FARM HOUSEHOLD
3.2.1 Farm household as resource manager
3.2.2 Farm household as system beneficiary
3.3 FARM RESOURCES
3.3.1 Farm resources from an accounting view
3.3.2 Farm resources from an operational view
3.3.3 Operational resource categories
3.3.4 Other relevant resource properties
3.3.5 Resource acquisition and generation
3.3.6 Relationships between resources, capital and costs
3.4 REFERENCES
4. FURTHER FARM-HOUSEHOLD SYSTEM ELEMENTS: ENTERPRISES AND
ACTIVITIES AND THEIR BUDGETING
4.1 ENTERPRISES VERSUS ACTIVITIES
4.2 ENTERPRISES
4.2.1 Enterprise boundaries
4.2.2 Enterprise structural types
4.3 ENTERPRISE AND ACTIVITY BUDGETS
4.3.1 Budget types and purpose
4.3.2 Budget standardization: units of measurement
4.3.3 Level of budget detail
4.3.4 Unit budgets
4.3.5 Extending budget scope: processing and marketing
4.3.6 Economic and financial budgets
4.3.7 Real and imputed input costs and output values
4.3.8 Budget-based measures of performance
4.3.9 Extension of enterprise or activity budgets to whole-farm budgets
4.3.10 Extension of whole-farm budgets to the household
4.3.11 Cost of production
4.4 ACTIVITIES
4.4.1 Types of resource-generating activities
4.4.2 Activity budgets in linear programming format
4.5 FURTHER EXTENSIONS OF ENTERPRISE OR ACTIVITY BUDGETS
4.6 PARTIAL BUDGETING
4.7 CONDITIONAL OR PARAMETRIC BUDGETING
4.7.1 Single-parameter extensions
4.7.2 Two- and three-parameter extensions
4.8 DO'S AND DONT'S OF ENTERPRISE AND ACTIVITY BUDGETING
4.9 REFERENCES
5. FURTHER FARM-HOUSEHOLD SYSTEM ELEMENTS: PROCESSES, STRUCTURAL
COEFFICIENTS AND THE WHOLE-FARM SERVICE MATRIX
5.1 PROCESSES: SYSTEMS OF ORDER LEVELS 1 AND 2
5.1.1 Definition and nature of processes
5.1.2 Number of relevant processes
5.1.3 Analytical problems presented by processes
5.2 INTERRELATEDNESS OF SYSTEM COMPONENTS: STRUCTURAL COEFFICIENTS
5.2.1 Interrelatedness within an activity: internal structural coefficients
5.2.2 Structural coefficients as critical parameters
5.2.3 Interrelatedness among activities: external structural coefficients
5.3 THE FARM SERVICE MATRIX: A SYSTEM OF ORDER LEVEL 9
5.3.1 General charges
5.3.2 Capital fixed costs
5.3.3 Relative importance of total farm fixed costs
5.4 DETERMINATION OF CAPITAL FIXED COSTS
5.4.1 Mandatory versus optional farm fixed costs
5.5 DEPRECIATION METHODS
5.6 REFERENCES
6. HOUSEHOLD GOALS, FARM PLANNING OBJECTIVES, SYSTEM PLANNING AND
PERFORMANCE CRITERIA
6.1 PLANNING AND OPERATING OBJECTIVES
6.2 SYSTEM PROPERTIES AND PERFORMANCE CRITERIA
6.2.1 Productivity
6.2.2 Profitability
6.2.3 Stability
6.2.4 Diversity
6.2.5 Flexibility
6.2.6 Time-dispersion
6.2.7 Sustainability
6.2.8 Complementarity and environmental compatibility
6.2.9 Summary
6.3 REFERENCES
7. ECONOMIC EVALUATION OF FARM SYSTEMS: MEASURES FOR EVALUATION
AND COMPARATIVE ANALYSIS
7.1 SCOPE AND PURPOSE OF EVALUATION
7.2 LIMITED EVALUATION
7.2.1 Using data aggregated on a whole-farm basis
7.2.2 Using data on an activity-specific basis
7.2.3 Productivity of individual resources on a whole-farm basis
7.2.4 Productivity of individual resources on an activity basis
7.2.5 Total productivity measures
7.3 COMPARATIVE ANALYSIS
7.3.1 Level I analysis - the whole farm
7.3.2 Level II analysis - the household
7.3.3 Level III analysis - the fixed-capital structure
7.3.4 Level IV analysis - the individual activities
7.3.5 Level V analysis - the underlying processes
7.3.6 Summary
7.4 DATA SOURCES
7.4.1 Data for the subject farm
7.4.2 Data for the comparative or standard farm
7.5 DIFFICULTIES IN EVALUATION AND COMPARATIVE ANALYSIS
7.6 REFERENCES
8. OPTIMIZATION OF RESOURCE USE LEVELS: RESPONSE ANALYSIS
8.1 INPUT-OUTPUT OR RESPONSE RELATIONSHIPS
8.2 OPTIMIZATION OF A SINGLE-VARIABLE INPUT RESPONSE PROCESS
8.2.1 Optimization by partial budgeting
8.2.2 Optimization by graphical methods
8.2.3 Optimization by using the response equation
8.2.4 Optimization by using the profit function
8.2.5 Maximum output vs optimal economic output
8.2.6 Constrained optimization
8.3 OPTIMIZATION OF MULTI-VARIABLE INPUT RESPONSE PROCESSES
8.3.1 Optimization based on MVP = MC
8.4 DATA SOURCES FOR RESPONSE ANALYSIS
8.4.1 Using existing data
8.4.2 Data from on-farm experiments
8.4.3 Trials with 'permanent' or long-term crops
8.5 DIFFICULTIES IN RESPONSE ANALYSIS
8.6 REFERENCES
9. PLANNING WHOLE-FARM SYSTEMS: ALLOCATION BUDGETING, SIMPLIFIED
PROGRAMMING AND LINEAR PROGRAMMING
9.1 TIME DIMENSION OF WHOLE-FARM PLANNING
9.2 GENERAL PLANNING OBJECTIVES AND PROCEDURES
9.3 PLANNING METHODS AND THEIR APPLICABILITY
9.3.1 Types of farms and activities
9.3.2 Assumed optimality of plans
9.3.3 Specification of non-resource constraints
9.4 ALLOCATION BUDGETING
9.4.1 Operating criteria for allocation budgeting
9.4.2 Allocation budgeting using GM per unit of land
9.4.3 Allocation budgeting using GM per unit of operating capital
9.4.4 Allocation budgeting using GM per family labour day
9.5 SIMPLIFIED PROGRAMMING
9.6 LINEAR PROGRAMMING
9.6.1 Linear programming of systems consisting of only final product-generating activities
9.6.2 Linear programming of Type 2 farms: systems with internally-generated resources
9.6.3 Linear programming of Type 1 farms: subsistence-oriented systems
9.6.4 Relevance of LP
9.7 REFERENCES
10. PLANNING FARM SYSTEMS OVER TIME
10.1 TIME EFFECTS IN AGRICULTURAL PRODUCTION
10.2 OPTIMAL USE OF TIME
10.3 TIME CLASSIFICATION OF ACTIVITIES
10.3.1 Short-term activities
10.3.2 Intermediate- and long-term activities
10.4 BASIC INTEREST-RATE CONCEPTS AND PROCEDURES
10.4.1 Compounding or taking a present value forward through time
10.4.2 Discounting or bringing a future amount back to present value
10.5 SUBJECTIVE (INTERNAL) AND OBJECTIVE (EXTERNAL) INTEREST RATES
10.5.1 Objective interest rates
10.5.2 Subjective interest rates
10.6 EVALUATING FUTURE COSTS
10.7 EVALUATING FUTURE NET RETURNS
10.8 ANNUITIES: EVALUATING REGULAR COST AND RETURN STREAMS OVERTIME
10.8.1 Terminal value of an annuity
10.8.2 Annuity equivalent to a future lump sum
10.8.3 Present value of an annuity
10.9 PERPETUITIES: EVALUATING REGULAR COST AND RETURN STREAMS WITH AN
INDEFINITE LONG LIFE
10.9.1 Perpetuities in the valuation of land and farms
10.10 AMORTIZATION: LIQUIDATING A PRESENT VALUE OVER TIME
10.11 SUMMARY OF PROCEDURES FOR FINANCIAL EVALUATION OF LONG-TERM
INVESTMENTS
10.12 EXAMPLE EVALUATION OF A PROPOSED LONG-TERM ACTIVITY
10.13 EVALUATION CRITERIA FOR LONG-TERM INVESTMENTS
10.14 DIFFICULTIES IN PLANNING FARM SYSTEMS OVER TIME
10.15 REFERENCES
11. PLANNING AND MANAGING FARM SYSTEMS UNDER UNCERTAINTY
11.1 UNCERTAINTY
11.2 RISK
11.3 SOURCES OF FARM-SYSTEM RISK
11.4 MANIFESTATION OF FARM-SYSTEM RISK
11.5 IMPACT OF FARM-SYSTEM RISK
11.6 FARMERS' RISK-MANAGEMENT STRATEGIES
11.7 FORMAL APPROACHES TO RISKY FARM DECISIONS
11.8 SENSITIVITY ANALYSIS
11.9 STOCHASTIC BUDGETING
11.10 SUBJECTIVE EXPECTED UTILITY
11.10.1 Personal preference or utility
11.10.2 Utility function elicitation
11.10.3 Probability elicitation
11.10.4 Example of subjective expected utility analysis
11.11 CERTAINTY EQUIVALENCE
11.12 DECISION TREES
11.13 STOCHASTIC DOMINANCE
11.14 RISK-ORIENTED MATHEMATICAL PROGRAMMING
11.14.1 Risk programming
11.14.2 Stochastic programming
11.15 MONTE CARLO SIMULATION
11.15.1 Steps in simulation modelling
11.15.2 Example of Monte Carlo simulation
11.15.3 Simulation flowcharts and computers
11.15.4 Other uses of Monte Carlo simulation
11.16 DIFFICULTIES IN LONG-TERM FARM DECISION MAKING
11.17 REFERENCES
APPENDIX: MANAGEMENT, FARM MANAGEMENT AND FARM SYSTEMS
1. THE CONCEPT OF MANAGEMENT
1.1 History of Management Thought
1.2 Definition of Management
1.3 Major Features of Management
1.4 Definition of Farm Management
2. THE DISTINCTIVE FEATURES OF AGRICULTURE
2.1 Biological Effects
2.2 Time Effects
2.3 Resource-portfolio Effects
2.4 Small-farm Effects
3. THE THEORY OF FARM MANAGEMENT
3.1 Farm-system Theory
3.2 Theory of Management by Objectives
4. THE PRACTICE OF FARM MANAGEMENT
4.1 Farm-system Approach
4.2 Management by Objectives
5. MAKING SMALL-FARM MANAGEMENT MORE EFFECTIVE
6. REFERENCES
FAO FARM SYSTEMS MANAGEMENT SERIES
ACKNOWLEDGEMENTS AND
DEDICATION
In preparing this book, the authors have received assistance directly or indirectly from many people.
Initial inspiration was provided by Dr Neal Carpenter, then Chief of FAO's Farm Management and
Production Economics Service (AGSP), who saw the need for a more holistic approach to
agricultural development problems at farm, household and village levels. Following Dr Carpenter's
untimely death, the project was carried through with the energetic support of his successors including
Dr Malcolm Hall and Dr Karl Friedrich. The first draft was constructively criticised by Dr John
Dixon, also of AGSP. Nor would the work have been possible without the assistance and cooperation
of the great many specialists, particularly small farmers and professional agriculturalists, with whom
we have talked and worked throughout Asia. Our thanks must also go to Ms Sally Bearman and Ms
Suzanne Blair for assistance in editing, processing and formatting the manuscript as camera-ready
copy.
The book has two source streams. One, the formal stream, consists of analytical frameworks and
techniques pertinent to the planning and management of farm systems. Our consulting of other authors
in these matters is acknowledged in the cited references. Beyond suggested classification schema for
agricultural systems, farm types and the modes and fields of farm management analysis, we add little
that is new. The main task was to apply these formalities to the world of small traditional farm
systems. Thus we exploit also a second stream: that of informal knowledge and village lore which
flows to us from the developing world itself - and we try to gain from it sufficient insight to suggest
what scope (and limitations) there might be for applying modem farm management techniques in the
context of farm systems in old traditional but inevitably changing cultures.
Full acknowledgment of the debt which we owe to the small-farm households of Asia is not possible.
This book is dedicated to them with respect and gratitude.
Douglas J. McConnell
John L. Dillon
FOREWORD
This book was produced by FAO as part of the work program of its Farm Management and
Production Economics Service (AGSP) for the furtherance of a systems approach to farm
development, especially of small traditional farms of the tropical world. Impetus to this has been
given by the ongoing activity of FAO's Regional Commission on Farm Management in Asia and the
Far East.
Because of both the importance of small farms in Asia and the continuing expansion of farm
management analytical concepts and methodology, it is timely to build on previous FAO publications
in this field. Substantially, these began with the publication in 1958 of W.Y. Yang's book Methods of
Farm Management Investigations and, recognizing FAO's special interest in world food security and
the sustainable development of small farms, have continued through to today's FAO Farm Systems
Management Series to which this book belongs.
The present volume is not an updated compendium of methodology. The authors have first developed
an agricultural systems framework and a classification of farm types and the modes and fields of farm
management analysis. They have then selected from the existing body of farm management
methodology those techniques for analysis and planning which, on the experience of several decades,
seem to be of greatest practical use for professionals analysing and planning farm-level agricultural
systems, viz: budgeting, enterprise and whole-farm comparative analysis, response analysis,
mathematical programming, simulation and risk analysis. Although the authors' techno-ideological
framework is economics, they point out that it usually does not matter which discipline - agronomy,
economics, engineering, sociology ... - takes the lead role in the analysis of small-farm systems as
long as these typically diversified farms and their farm-household units are approached holistically as
functioning integrated systems which need to be adequately understood before prescription for
'improvement' is offered.
Most modem textbooks on farm management are set in the context of modem Western commercial
farming. In being oriented to the small farms and rural households of Asia where population pressure
is greatest, this book illustrates the practical application of analytical methods described in J.L.
Dillon and J.B. Hardaker's book Farm Management Research for Small Farmer Development in
this same series.
To contribute positively to sustainable agricultural development, these methods must contribute to the
betterment of food security and income of poor folk supported by small farms using traditional
technology and internally generated resources. To assist in this, the first requirement is that we have a
better understanding of the actual agro-economic structuring, nature and functioning of these small-
farm systems and their households. We must seek such understanding with great care and humility.
For as this book points out, these small farm-household systems which were once dismissed as
'backward' and as prima facie evidence of under-development, will generally on closer inspection
prove to be far more sustainable, efficient and resilient than the bulk of commercial agriculture in the
West.
From decades of experience, we now know better than to try to replace traditional smallholder
agriculture with some allegedly superior form of modem commercial agricultural production
transferred from the West. The degree of interdependence among crop and livestock activities and
between the farm and the household in such small-holder systems is such that one would not attempt
to adjust, improve or modernize any single element without expecting repercussions throughout the
whole system. This characteristic of more-or-less tight structural integration does not necessarily
mean that all these farms are biologically and economically efficient and sustainable. They range
from what are arguably the world's most sustainable, sophisticated and complex systems - the
household forest-gardens of Sri Lanka, Kerala and Java - to systems which, through excessive
demands placed on them, not infrequently generated by external factors, have become degraded to the
point of abandonment.
In this book the authors present a powerful argument for the application of farm management and
farming systems analysis to these complex small-holder agricultural systems in Asia. They leave the
reader with profound respect for what this traditional world of the small farm can offer in return.
Doyle Baker
Chief, Farm Management and Production Economics Service, Agricultural Support Systems
Division
ABBREVIATIONS, ACRONYMS AND
SYMBOLS USED
a: the lower-range or minimum value of a random variable
A: an annual payment amount which, continued over n years, constitutes an annuity with a
terminal value An = A[(1+i)n-1]/i or which amortizes a PV where A =
PV[i(1+i)n]/[(1+i)n-1]
An a future amount to be received or paid in n years' time; it is equivalent to an annuity

over n years of A = Ani/[(1+i)n-1] and has a PV of An/(1+i)n
AB: allocation budgeting
ac: acre(s)
ac ft: acre feet (of irrigation water)
AME: adult male equivalent, a standardized unit of labour measurement
APP: average physical product, equal to Y/X for Y = f(X)
AU: animal unit, a standardized unit for aggregation or comparison of different types of
animals
AVP: average value product, equal to (APP)py
b: the upper-range or maximum value of a random variable
B: the resource pool available to the farmer or, in LP, the level of resource constraints
or activities constituting the farm 'plan' at each iteration
B: size of the available cash budget
B/C: ratio of benefits to costs
B-1: in AB and SP, the initial set of resource constraints
B-i: in AB and SP, the balance of resource constraints after sequential introduction of the
i-th activity into the plan, i = 1,2,...
C: depending on context, cost or the coconut-to-copra conversion rate
CDF: cumulative distribution function
CE: a decision maker's certainty equivalent for a risky prospect

cm: centimetre(s)
COP: cost of production, calculated per unit of output as total cost divided by quantity of
output, i.e., TC/Y
CV: coefficient of variation, calculated as 100 to express the standard deviation of a

variable as a percentage of its mean
CY: crop-cycle year of a perennial crop
d: the constant annual rate of depreciation under the declining-balance method
D: the annual depreciation cost as measured by the straight-line method
Dt: the annual depreciation cost in year t as measured by the declining-balance or sum-of-
integers methods
DC: direct cost of an enterprise (or activity), calculated as the sum of its VC plus its share
of farm FC
DI:
Simpson's diversity index, equal to where S is the number of species or
activities present; ni is the number of individuals in the i-th species, or area devoted
to the i-th species or activity, or income or value of the i-th species or activity; and N
(=S ni) is the total population of individuals across all species, or total area across
all species or activities, or total farm income or value across all species or activities
dU/dX: marginal utility of X for U = f(X)
dY/dX: marginal product of X in Y = f(X) or the first derivative of Y with respect to X
E:
mean or expected value, calculated for a variable X as for a sample of
observations X1, X2.... Xn
EMV: the expected money value of a risky prospect
EU: subjective expected utility of a risky prospect
EV: expected value of a risky prospect
f0.i: the i-th fractile of a probability distribution; a proportion 0.i of the distribution of a
random variable X will lie below the value X = f0.i
FC: fixed cost
F(PV): CDF of PV
F(X): the cumulative probability or CDF of a random variable X
G: yield of grain
GAMS: General Algebraic Modelling System
GM: gross margin, calculated as gross return less variable cost
GR: gross return or gross revenue
ha: hectare(s)
HYV: high-yielding variety
i: real annual rate of interest after correcting for inflation or deflation, equal to
[(1+i*)/(1+w)]-1, or, when used as a subscript, denotes the i-th member of a set
i*: nominal annual rate of interest or rate of discount
IRR: internal rate of return defined as the annual interest rate i at which the PV of an
investment's costs is equal to the PV of its gross returns
K: in LP, the maximum level of an activity permitted by a particular constraint
kg: kilogram(s)
km: kilometre(s)
L: the expected total years of useful life of a capital item
LP: linear programming
LU: labour unit, a standardized unit for measuring labour
m: metre(s)
m: the modal value of a random variable
-M: in LP, the notional high negative p value placed on Qi
MC: marginal cost or the change in TC as one more unit of a variable input factor is used
mds: maunds, a volumetric measure used in Pakistan and India
MOTAD: Minimization of Total Absolute Deviations, a form of linear risk programming
MPP: marginal physical product, equal to dY/dX for Y = f(X)
mt: metric tonne(s)

mths: months
MVP: marginal value product, equal to (MPP)py
n: depending on context, the number of observations on a variable, or the size of a

sample, or the number of input factors in a production process, or the number of equal
intervals (usually years) in a period of time
ni: for a given farm system, the number of individuals in the i-th species, or area devoted
to the i-th species or activity, or income from the i-th species or activity; used in
calculating the DI of a farm system
N: depending on context, quantity of nitrogen fertilizer or, for a given farm system, the
total population of individuals across all species, or total area across all species or
activities, or total farm income or value across all species or activities; used in
calculating the DI of a farm system
no.: number
NPK: fertilizer containing nitrogen, phosphorous and potassium
NR: net return or net revenue, calculated as TGR minus TC
Nu: Ngultrum; Bhutanese unit of currency
OC: overhead costs, defined as fixed costs accrued on a whole-farm basis
OGM: operational gross margin of an enterprise (or activity), calculated as TGR minus DC
of the enterprise (or activity)
Oi: the i-th possible outcome of a particular risky decision
p: in LP, the unit price or value of each real or disposal activity; it is equivalent to an
activity's unit GM
pg : unit price of grain
pi: unit price of the i-th good
pn : unit price of N
px : unit price of input factor X
py : unit price of product Y
P: quantity of phosphate fertilizer
pk: patok(s), a Javanese land unit equal to about 0.1 ha

P(Oi): the decision maker's subjective probability for the occurrence of Oi
P(R£ R*): the probability that net return R will be less than or equal to any nominated value R*
PV: the present value of a future lump sum or stream of payments or of an item; if An is to
be received n years from now, its PV is An/(1+i)n assuming an annual interest rate of
i
PVt: PV at time t
P(X): probability of X
q: number of times per year that compounding or discounting is to occur
Qi: in LP, the i-th minimum planning constraint or its corresponding artificial activity
r: interest rate per period of length 1/q years, equal to an annual interest rate i of (1+r)q
-1
R: depending on context, rainfall, or net return, or an annual payment amount that

continues in perpetuity and thus has a PV of R/i, or the income diversity ratio, equal to
(S Ri)²/S Ri² where Ri (i = 1 to n) is the income from the i-th activity and 1 £ R £ n
for Ri³ 0
Ri: income from the i-th activity
R*: any nominated value of net return R
RN: random number
Rp: Rupiah; Indonesian unit of currency
Rs: Rupee; Indian, Pakistani and Sri Lankan unit of currency
RTC: index of relative time-concentration of annual production or income of a product,

calculated as the CV of the product's production or income pattern as a fraction of the
corresponding CV for a perfectly concentrated product
RTD: index of relative time-dispersion of production or income, calculated as 1 - RTC
S: the number of species or activities in a farm system; used in calculating the DI of a

farm system
SD: standard deviation, calculated as the positive square root of variance

SP: simplified programming
SV: the salvage value of a capital item
t: depending on context, ton(s) or tonne(s)
t: time, e.g., year t, or the time-length of an activity
TC: total cost, calculated as the sum of total variable cost and total fixed cost
TDN: total digestible nutrients
TFC: total fixed cost
TGM: depending on context, the total gross margin of an activity, of an enterprise or of the
whole-farm system; equal to TGR - TVC
TGR: total gross return or total gross revenue
TPP: total physical product, equal to Y for Y = f(X)
TVC: total variable cost
TVP: total value product, equal to (TPP)py
U: a decision maker's utility function
U*: the linear transformation of U, equal to aU + b with a > 0
UAE: United Arab Emirates
U(X): a decision maker's utility function specified in terms of the variable X
V:
variance, calculated for a variable for a sample of observations X1,
X2,...Xn
Vt: the depreciated value of an item at the end of year t under the declining-balance
method
VC: variable cost
vs: versus
w: annual rate of inflation (w > 0) or deflation (w < 0)
X: a variable or an input factor
: arithmetic mean of a set of sample observations {Xi}, i = 1 to n, on a variable X,

calculated as
X1 = isoquant equation giving the locus of all combinations of two variable inputs X1 and
h(X2/Y*): X2 to produce a fixed level of output Y*
Xi: depending on context, the i-th observation on a variable X, or the i-th input factor, or
the level of the i-th input factor
Y: depending on context, a product or the yield of product (or output) from a production
process
Y*: some fixed level of output Y
Y = f(X): single-variable production or response function indicating output Y is a function of the

variable input factor X
Y = f(X1, X2): production or response function indicating output Y is a function of two variable
inputs X1 and X2
Y = f(X1, production or response function indicating output Y is a function of n variable inputs

X2,...Xn): X1, X2,...Xn
Y= production or response function indicating output Y as a function of the variable input

f(X1/X2,...Xn): X1 when inputs X are held constant
Z: in LP, the TGM of a farm plan or the opportunity cost of including a unit of an activity
in the plan
Z - p: in LP, for an activity, when multiplied by minus one, indicates the net increase in plan
TGM to be gained by adding one more unit of the activity to the plan
¶ Y/¶ Xi: marginal product of Xi in Y = f(X1, X2,...Xn), i = 1 to n, or the first partial derivative
of Y with respect to Xi
D (.): a small incremental change in (.)
D Xi: a small incremental change in the level of the variable input Xi
p
profit, equal to for the single product activity Y = /(X1,
X2,...Xn); for multi-product situations, equal to TGR - TVC - TFC or, more simply,
GR - VC - FC
for a variable X, the sum of the values X1, X2,...Xn

¥: infinity
>: greater than, i.e., a > b indicates a is greater than b
<: less than, i.e., a < b indicates a is less than b
1. AGRICULTURAL AND FARM SYSTEMS

- CONCEPTS AND DEFINITIONS
1.6 REFERENCES
'For an understanding, not only the elements but their interrelations as well are required.'
Ludwig von Bertalanffy (1973)
The first purpose of this introductory chapter is to develop a conceptual framework for the
examination of the agro-economic structure of farm-level agricultural systems. The second purpose is
to sketch the relationships among these farm-level systems, and between these on the one hand and
higher-level systems on the other. These considerations form the basis for the presentation in later
chapters of an analytical approach to farm management from a systems perspective applied in the
context of Asian agriculture.
While somewhat original in the comprehensiveness of its farm systems' schema, the analytical
framework and approach taken are not in conflict with the approaches to systems theory and
(agricultural) systems analysis as presented by such authors as Ackoff (1973), Ackoff and Emery
(1972), Boulding (1956), Checkland (1981), Dillon (1992), Dillon and Anderson (1990, pp. 164-
174), FAO (1989 and 1990), Fresco and Westphal (1988), Friedrich (1992), Kast and Rosenzweig
(1974), Norman (1980), Ruthenberg (1976 and 1980), Shaner, Philipp and Schmehl (1982), Spedding
(1979) and von Bertalanffy (1973).

An agricultural system is an assemblage of components which are united by some form of interaction
and interdependence and which operate within a prescribed boundary to achieve a specified
agricultural objective on behalf of the beneficiaries of the system.
This definition is analogous to the general definition of any artificial (i.e., man-made) system of
which all managed agricultural systems (including specifically the farm-level systems) form one sub-
division as shown in Figure 1.1.
From a practical production, administration and management point of view, as shown in Figure 1.2,
'all agriculture' can be regarded as consisting of sets of systems at 16 Order Levels or levels of
generality. As discussed in Section 1.3. these 16 Order Levels largely constitute a nested hierarchy.
This book is concerned with the 12 lowest-order systems, those at farm level, i.e., systems of Order
Levels 1 to 12 in Figure 1.2.

Discussion and analysis of systems can be of them as actual systems (e.g., of constituent physical
processes in the case of natural physical systems) or as representational systems. Common
representations or models of actual systems take such forms as written descriptions, physical models,
mathematical models, flowcharts, tables of data and computer programs. In the following discussion,
reference is to representational systems.
Systems can be classified into three broad families or divisions as either natural, social or artificial
systems (Figure 1.1).
(a) Natural systems - those that exist in Nature - consist of all the materials (both physical and biological) and interrelated
processes occurring to these materials which constitute the world and, inter alia, provide the physical basis for life. They exist
independent of mankind. Our role in relation to natural systems is to try to understand them and, as need be, make use of them.
We also (increasingly) attempt to duplicate them, in part or whole; but at this point they become, by definition, man-made or
artificial systems. These fundamental natural systems remain unaffected by attempts at imitation. Those natural physical and
biological systems (shown in their totality as the division of natural systems in Figure 1.1) which are relevant to agriculture will be
self-apparent: rock weathering to form soil; plants sustained by such soil; animals sustained by such plants ... are examples of the
outward forms of agriculturally relevant natural systems in operation.
(b) Social systems are more difficult to define. Essentially they consist of the entities forming animate
populations, the institutions or social mechanisms created by such entities, and the interrelationships
among/between individuals, groups, communities, expressed directly or through the medium of
institutions. Social systems involve relationships between animate populations (individuals, groups,
communities), not between things. Concern here is with human social systems as they relate to or
impinge upon farming, and the term social system is used broadly to include institutions and
relationships of an economic, social, religious or political nature. There is a certain degree of
ambiguity in defining social systems. As an example, the law of property is in its essence a social
system. Insofar as it is viewed as consisting of concepts, principles and rules, it is a pure social
system, independent of natural systems. But its existence also presupposes the existence of property,
including natural physical things, some of which exist as systems. To this extent, as a social system
the law of property is dependent on or subordinate to natural systems.
(c) Artificial systems do not exist in Nature. They are of human creation to serve human purposes. All
artificial systems, including agricultural systems, are constructed from either or both of two kinds of
elements: (a) elements taken from either or both of the other two higher-level orders of systems at
division level, i.e., from natural and social systems, and (b) from elements which are constructed or
proposed for specific use by each respective artificial system as the need for this arises.
The upper part of Figure 1.1 depicts the dependence relationship between natural and social systems
on the one hand and between these and artificial systems on the other. The relevant relationships are:
(i) natural systems are independent of systems of the other divisions; (ii) social systems could also be
viewed as being independent, but generally a more legitimate view would be that they depend
immediately or eventually on natural systems for the essentials of their material existence; and (iii)
artificial systems are directly dependent on either or both natural and social systems, or indirectly on
natural systems (through the dependence of social systems themselves on natural systems).
FIGURE 1.1 - Agriculture in relation to Other Systems
In Figure 1.1, agriculture is shown as comprising one of a very large number of actual or potential
artificial systems at the sub-division level. Others are those relating to mining, transport, public
health, education etc. What such systems at this sub-divisional level have in common is that each is
artificial: each is based upon or draws elements from higher-level natural and social systems; and
each also contains elements which are purposefully created by some human agency in order to meet
its needs.
As shown in Figure 1.1, systems within the three broad divisions or their multitudinous subdivisions
can be further classified according to system 'type', a loose term but one which might be used to
differentiate among agricultural systems according to a number of factors of which only two are
shown in the sketch. As outlined below, first, the system might be either an explicit or implicit one;
second, its purpose might be either descriptive or operational. Other 'type' designations could be
added; e.g., operational systems could be further classified according to whether or not they are
amenable to optimization.
· Explicit systems are those in which the constituent elements are more or less closely identified and defined, and the
relationships among these elements are stated formally in quantitative, usually mathematical, terms. Agricultural scientists and
economists who work with farmers are concerned mainly with explicit systems of Order Levels 1 to 10 as specified in Figure 1.2.
But farmers themselves will seldom be concerned with explicit systems - only with systems of a simpler kind, or only with
selected parts of such systems.
· Implicit systems are systems in which only the main or critical elements are acknowledged and only
the major or immediately relevant interrelationships are considered. However, these elements and
relationships are not formally recorded, analysed or evaluated. Farmers themselves deal primarily
with implicit systems. In both traditional and more modem societies particular agricultural systems of
Order Levels 1 to 10 are implied in what farmers do, or deliberately do not do. In more 'advanced'
societies, farmers might formalize and work with a few explicit systems or parts of systems (farm
record books, simple crop budgets, household expenditure accounts) but here also most agro-
management systems will exist by implication.
The purpose in here distinguishing between explicit and implicit systems is to discourage the view
that, because farmers (especially small traditional farmers) do not deal with explicit formal systems,
these farmers are backward, ignorant, unsophisticated and generally inferior as resource managers. If
anything, the facts generally point to a contrary conclusion. While bad farmers can be found
anywhere, any close study of small traditional farmers and farming villages in the developing world
will, with patience, identify implicit systems at agro-technical, enterprise, farm, farm-household and
village levels which are far more complex, sophisticated, sustainable and socially efficient than most
agricultural systems found in developed countries.
· Descriptive systems are usually intended to facilitate an understanding of the organization, structure or operation of a productive
process. This might be their sole purpose; e.g., a farmer might construct a simple input-output budget table in order to learn the
structural configurations of some potential new crop. Depending on the results of this, he or she might then proceed to construct
a more detailed budget (an operational system) to find how best to fit this new crop into his or her farm plan. At higher Order
Levels an organogram describing the administrative structure of a ministry of agriculture or of an extension service might be
constructed or the flowchart of a commodity from farm to consumer might be drawn - these also are descriptive systems.
· Operational systems are constructed (by an analyst or manager or research worker) as a basis for
taking or recommending action aimed at improving the performance of the system. Such systems are
often elaborate (as exemplified in Chapters 9 and 11). However, increased precision is not
infrequently achieved at the cost of decreased practical usefulness. Thus farm managers themselves
work primarily with simple operational systems, although the actual physical systems which these
represent may be very complex.
As outlined by Dillon (1992), it is also sometimes useful to recognize that, like other systems,
agricultural systems may be categorized as:
· Purposeful or non-purposeful depending on whether or not they can select goals and the means by which to achieve them.
· Static or dynamic depending on whether or not they change over time in response to internal or
external influences.
· Open or closed depending on whether or not they interact with their environment.
· Abstract or concrete depending on whether or not they are conceptual or physical in nature.
· Deterministic or stochastic depending on whether or not their behaviour exhibits randomness over
time, i.e., their future behaviour is uncertain.

Agricultural and particularly farming systems exhibit great diversity as shown by, e.g., Duckham and
Masefield (1970), Grigg (1974), Kostrowicki (1974) and Ruthenberg (1980). They have been
classified in various ways as reviewed by Fresco and Westphal (1988) who also present an
ecologically-based classification and typology of farm systems. The hierarchical classification of
farm systems presented here is distinctly different. It is specifically oriented (i) to a farm management
and farm-household perspective and (ii) to use as a framework for analysis of what are proposed as
the six basic types of farms found in Asia (and elsewhere in the developing world).
Figure 1.2 is an elaboration of the lower part of Figure 1.1 and relates specifically to agricultural
systems. These are listed in largely hierarchical order encompassing 16 Order Levels. Alternatively,
with a few minor exceptions, the Order Levels 1 to 16 could have been depicted, reflecting their
nested character, as a set of concentric circles with Order Level 1 as the innermost and Order Level
16 as the outermost circle.
In Figure 1.2, the sectoral system, 'all agriculture', is specified as being of the highest order rank, i.e.,
Order Level 16. Any national or regional agricultural sector, however, consists of such subordinate
sub-sectors or subsystems as agricultural credit, education, research, production, transport etc. Each
of these constitutes and would be analysed, administered and managed as a system of Order Level 15.
Each such (sub)system may then be further disaggregated into commodity-based industry systems of
Order Level 14 such as for coconuts, rubber, wheat, coffee, fish etc. If that flow-path relating to
production is being followed, as depicted in Figure 1.2, this would then lead to villages or other
community units where such production occurs (systems of Order Level 13); these would in turn
consist of and could be disaggregated into the individual farm-household systems of Order Level 12
which comprise such villages. Further lower Order Level systems relate to the agro-economic
structure of individual farms and, in turn, their component crop and livestock enterprises and to the
activities and individual agro-technical processes which underlie such enterprises.
FIGURE 1.2 - The Hierarchy of Agricultural Systems
Systems of Order Levels 1 to 12 comprise the field of farm management (as discussed in Chapter 2).
But systems of Order Level 1 and 2 are also, indeed primarily, the domain of the applied agricultural
sciences. A further proviso is that the 'household' components of farm-household systems of Order
Level 12 remain as yet not very well understood. This component is primarily the province of
workers in such fields as household economics, rural sociology and social anthropology. While these
various farm family-related fields are fairly well established, they have yet to be brought together in a
comprehensive and cohesive way at farm-family level to provide verified models of how rural
families in the developing world think about, plan and operate the 'farm' component of their farm-
household systems (Clayton 1983, Chs 4 and 5).
Figure 1.2 depicts the direction of hierarchical status as proceeding downward from sector to
industry to village to farm to crop etc. But whether this direction of subordination is valid will
depend on circumstances and analytical purpose. Agricultural scientists would probably reverse the
order-ranking shown for the systems on the grounds that, unless the basic agro-technical processes
(Order Level 1 and 2 systems) are well developed, the production of individual crops will be
inefficient, total farm production will be low and the agricultural sector itself will in consequence be
an impoverished one. Similarly extension workers might be inclined to place household systems at
the top of the systems hierarchy on the basis that good farming practices (Order Level 1 and 2
systems) will not be adopted unless the household systems are working well, nor consequently will
the 'higher'-order systems at industry and sector level operate at their full potential.
The nature of each farm-level system (i.e., Order Levels 1 to 12) of the hierarchy presented in Figure
1.2 may be specified from a management point of view as follows:
· Order Level 1: Uni-dimensional process systems. Systems of this lowest order are of an agro-technical nature. They involve an
issue or problem which for purposes of analysis or management is abstracted from the context in which it naturally or normally
occurs. One example is the application of a single fertilizer element, say nitrogen (N), to a crop and consequent plant response to
N in terms of crop yield Y. As noted previously, systems of this order are primarily the domain of physical scientists, but those
systems which have practical relevance for farmers thereby also have an economic dimension and so fall within the scope of farm
economics. Such simple single-dimensional systems are later examined as processes (Chapter 5) and as input-output response
relationships (Chapter 8).
· Order Level 2: Multi-dimensional process systems. Systems of this second order are also
concerned with limited agro-technical relationships and again they are primarily the domain of
physical scientists. They differ from Order Level 1 systems in that they take - or are defined to take -
a wider and more realistic view of a subject or problem. To use the same example of fertilizer
response: at Order Level 2 an agro-technical system might involve the response of plant growth or
yield Y to not one but to several or a large number of input factors such as nitrogen, phosphorous,
irrigation water, crop hygiene, soil tilth etc. These multi-dimensional systems also are later examined
as processes (Chapter 5) and as response relationships (Chapter 8). Order Level 2 systems can be
viewed as aggregations (often interactive) of constituent Order Level 1 systems.
· Order Level 3: Enabling-activity systems. Systems of this order are certain enabling activities
which generate an intermediate product intended for use as an input/resource by enterprises which do
produce a final product. An example is offered by a legume crop turned under to provide fertility for
a following (final product-generating) paddy crop. There will often be alternative ways of obtaining
this resource: e.g., stripping leaves off leguminous trees, keeping cattle for their manure, or buying a
bag of fertilizer. These are all enabling, resource-generating activities but only some of them, the
complex ones, warrant designation as systems. They are intended to supply resources to systems of
Order Levels 4 and 6.
· Order Level 4: Crop systems. Systems of this order relate to the production of individual crops; but
if these are primarily intended to produce inputs for other crops or livestock, they are regarded as
systems of Order Level 3. On many small farms, crop and livestock enterprises produce both final
products and resources (as discussed in the context of activities in Chapters 3, 4 and 9).
· Order Level 5: All crop systems. Systems of this order, known also as cropping systems, refer to
the combined system of all the individual crops on a farm. On a farm with a single mono-crop, this
Order Level 5 system will obviously be equivalent to an Order Level 3 system; but on small mixed
farms there will usually be four, five, six or more different crops (of Order Levels 3 and 4) grown in
some degree of combination and as many as 20 or more on the highly diversified forest-garden farms
of South Asia.
· Order Level 6: Animal systems. These systems relate to single-species animal enterprises or
activities - e.g., dairy cows, camels, fish, ducks. They are the animal equivalent of Order Level 4
(i.e., individual crop) systems.
· Order Level 7: All animal systems. These systems are the aggregation of all Order Level 6
(sub)systems on a farm. Known as livestock systems, they are the animal equivalent of Order Level 5
(i.e., all crop) systems.
· Order Level 8: Resource pool. This subsystem is a conceptual device for farm-system planning in
which resources and fixed-capital services required by other subsystems are 'stored' in a 'resource
pool' from which they are allocated to the other subsystems (of Order Levels 1, 2, 3, 4 and 6). The
resource pool is central to operation of the whole farm-household system. It is discussed in Chapter
3.
· Order Level 9: Farm service matrix. A system of this Order Level consists of all the fixed capital
resources of a farm which are pertinent to the operation of the farm as a whole but are not assigned to
the exclusive use of any particular enterprise or activity: land, fences, barns, irrigation channels and
work oxen are common examples. Some of these capital items are true (sub)systems, having
interdependence among their component parts (as in an irrigation storage/delivery/distribution
network, a grain drying facility, an integrated network of soil conservation structures etc.). Some are
only things (e.g., fences, a plough, a barn). But, in its totality, such capital is managed and
manipulated as a system for the purpose of providing general services which, while not specific to
them, enable the functioning of lower Order Level systems of the farm. This service matrix is
discussed in Chapter 5.
· Order Level 10: Whole-farm systems. Systems of this Order Level consist of all the lower Order
Level (sub)systems which go to make up a farm. They consolidate in a single entity all the farm fixed
capital, all the operating capital, all the final-product enterprises, all the activities and all the agro-
technical processes which underlie such enterprises and activities. Structuring and managing systems
of this Order Level are the main tasks or focus of farm management as carried out, on the one hand, by
farmers and as investigated, on the other hand, by farm management economists in their professional
capacity of providing advice to farm managers, development agencies and governments.
The terms farm system and farming system are often used interchangeably. Here the practice is to use
farm system to refer to the structure of an individual farm, and farming system to refer to broadly
similar farm types in specific geographical areas or recommendation domains, e.g., the wet paddy
farming system of West Java or the grain-livestock fanning systems of Sind.
· Order Level 11: Household systems. On small farms the household itself is the most dynamic and
complex of all farm-level systems, although it is a social system not an agricultural one. It dominates
the agricultural systems which comprise the farm component. It has two functions: as household it
provides purpose and management to the farm component, and as major system beneficiary it
receives and allocates system outputs to itself and other beneficiaries.
· Order Level 12: Farm-household systems. These consist of two components or (sub)systems of
Order Levels 10 and 11, i.e., the whole-farm system and its associated household system,
respectively. The term is a very useful if not mandatory one when used to refer to the small farms of
Asia. It carries an insistence that the technical analysis discussed in following chapters will amount
to nothing at all unless it is applied to achieving the real needs and aspirations of the household -
which, as discussed in Chapter 6, might be quite a different thing from evaluating the performance of
a farm system according to the subjective or preconceived ideas of agricultural technicians and
economists (Chambers and Ghildyal 1985; Rhoades and Booth 1982). As the peak farm-level system,
the farm-household system may be described in system terms as a goal-setting (i.e., purposeful) open
stochastic dynamic system with a major aim of production from agricultural resources. These
attributes are sufficient to make it also a complex system. The purposefulness of a farm-household
system is ensured by its human and social involvement which enables the system to vary its goals and
their means of achievement under a given environment. The openness of the farm-household system is
obvious from its physical, economic and social interaction with its environment. The non-
deterministic or stochastic nature of the farm-household system is guaranteed both by the free-choice
capacity of its human (and, if present, animal) elements and by the stochastic nature of the
environment with which it (and all its subsystems) interacts. Necessarily, a farm-household system is
also dynamic by virtue of its purposefulness, openness and stochasticity which ensure that the system
changes over time. Too, any farm-household system is a mixture of abstract and concrete elements or
subsystems. The concrete elements are associated with the physical activities and processes that
occur in the system. The abstract elements relate to the managerial and social aspects of the system.
Not infrequently in parts of Asia, as also elsewhere in the developing world, the village may replace
the farm-household in whole or part as the focal entity for agricultural production. Systems of Order
Level 13, i.e., village or community systems, are thus often relevant to the performance of farming
systems (Cederroth 1995; Walker and Ryan 1990).
· Order Level 13: Village-community systems. Village-level systems or community systems in some situations replace all or part of
individual farm-household systems. Three situations are common. First, some production activity in its entirety, including the
operation of whole farms as production units, may be on a formal cooperative or group basis. Second, only part of an activity
might be carried on by individual farmers while critical parts of it (such as land preparation, the supply of inputs, harvesting and/or
marketing) are the responsibility of a formal farmers' club or cooperative. Third, and most difficult to analyse, is the situation
found in many Indonesian villages where informal and temporary groups form to perform certain production tasks in common
(such as land preparation, irrigation and/or harvesting) then disband and re-form to do different tasks on different crops, with
membership continuously changing as individuals drop in and out of groups according to their interests, needs and mutual
obligations. In a village there might be 10, 20 or 30 such 'cooperatives', though none might exist officially. Other examples are
offered by the semi-nomadic livestock farmers of West Asia who sometimes operate as individual households and sometimes as
members of a collective. In all these situations the boundaries of individual units are often so fluid and obscure that the focus for
productive analysis has to be the group or village community. (Nevertheless, much externally sponsored farm-development
planning remains locked into the mythology of agricultural individualism; perhaps that is why on the small farms of Asia it has
borne so little and often poisonous fruit.)
Farm-level systems of Order Levels 1 to 12 are discussed more fully in the following chapters.
Before proceeding, however, it will be useful to examine those constituent structural elements of a
farm-household system which are relevant to its organization and management.

The definition of an agricultural system given in Section 1.1 is a general one and applies broadly to
systems of all the Order Levels. When applied specifically to a farm-household system of Order
Level 12 it implies the system involves ten structural elements or components:
1. Boundaries
*2. Household 3. Operating plan
*4. Production-enabling resources: the resource pool
*5. Final product-generating enterprises
*6. Resource-generating activities
*7. Agro-technical processes
* 8. Whole-farm service matrix
9. Structural (interdependence) coefficients
10. Time dimension.
Those elements marked by an asterisk have been considered above as subsystems of the farm-
household system (see Figure 1.2). The ten elements are briefly discussed below and, except for
structural coefficients and the time dimension, their interrelationships as components of a farm-
household system of Order Level 12 are sketched in the example of Figure 1.3 where they are denoted
E1, E2 ... E8.
1. Boundaries: This first element, the boundaries of the farm-household system, set it apart from other systems and from the
world at large. These boundaries are provided partly by the structural characteristics of the particular type of farm (Chapter 2), and
partly by the purpose of analysis, i.e., to some extent they are subjective and relate to more than the simple physical boundary of
the farm. Boundaries are discussed in Chapter 3.
2. Household: As previously noted, the household plays two roles: first, it provides purpose and
management to its associated farm system and, second, it is the major beneficiary of its associated
farm system. Its role as beneficiary is discussed in Chapter 3. In its first role it provides purpose,
operating objectives and management to the farm component of the farm-household system according
to its broad domestic and social goals. Obviously these goals vary widely with culture, tradition and
the degree of commercialisation and external influences to which the household is exposed. However,
one would probably be not too far wrong in offering a generalization that the primary economic goal
on most small farms (Types 1, 2, 3 of Chapter 2) is security and the primary non-economic goal is
social acceptance (Clayton 1983, Ch. 4). If this is correct, the primary objectives for the farm are,
first, production of a low-risk sustainable subsistence for primary system beneficiaries; second,
generation of a cash income to meet needs not directly met in the form of food and other farm-
produced materials; and third, pursuit of both of these in ways which are not in conflict with local
culture and tradition. Goals, objectives and planning criteria are discussed in Chapter 6.
3. Operating plan: The above objectives are pursued through preparation and execution of a farm
operating plan. The core of this may be taken as selection of the best possible mix of agro-technical
processes, activities, enterprises and fixed capital (systems of Order Levels 1, 2, 3, 4, 6 and 8).
Formulation of operating plans is discussed in Chapter 9.
4. Resource pool: This element was noted above as a system of Order Level 8 central to the
management of other subsystems within the farm system. It is discussed in Chapter 3.
5. Final product-generating enterprises: These were noted as systems of Order Levels 5 and 7 in
the previous section and are discussed in Chapter 4 and Section 9.3.
6. Resource-generating activities: These also were previously discussed as systems of Order Level
3. They are intended to supplement or entirely supply the resource pool as discussed in Sections 4 4.1
and 9.3.1.
7. Agro-technical processes: These were defined above as systems of Order Levels 1 and 2.
Processes may be of a biological or mechanical kind. They are a shorthand designation of all the
potentially complex and interrelated physical and biological factors underlying production from crop
or livestock species, only some of which may be economically relevant. They are discussed in
Chapter 5.
8. Whole-farm service matrix: This was discussed previously as a system of Order Level 9. It is
further examined in Chapter 5.
9. System structural coefficients: These coefficients identify and quantify linkage relationships (a)
among the various parts or elements within each subsystem and (b) between subsystems. From the
general system definition, an essential property of any system is that there be interrelatedness between
its parts. In farm-household systems (and in subordinate subsystems of lesser Order Level,
particularly Order Levels 4 and 6) such interrelatedness is specified by these coefficients. They are
discussed in Chapter 5.
10. Time dimension: Unlike mechanical systems which stamp out buttons or TV sets, agricultural
systems rest on biological processes which occur over considerable periods of time - from, e.g., a
few days in the case of quick-response agricides to 70 or more years in the case of growth and
decline of a coconut palm. Agricultural systems are thus inherently stochastic: being dependent on the
passage of time, ex ante, their outcomes are uncertain. Moreover, because agriculture is also a set of
economic activities, the old adage applies: time is money. Other things being equal, a system which
yields its product or ties up resources over a short time is better than one which yields its output or
occupies resources over a long time. Strictly speaking, time is not a system component; rather it is a
dimension in which the system operates. The time dimension in relation to resource use is discussed
in Section 3.3.4 and in relation to farm planning in Section 9.1. The evaluation of activities which
occur over long time periods is examined in Chapters 10 and 11. The latter chapter also considers
uncertainty as it occurs in farm planning and decision making. Also important from a time perspective
are the sustainability and environmental compatibility of the farm system being used. If, over time, the
farm system is not biologically and economically sustainable or causes resource degradation, as
discussed in Sections 6.2.7 and 8, this is to the disadvantage of both the farm household and society at
large.

Before examining the elements of a farm-household system in more detail in later chapters it is useful
to consider where they lie in relation to each other in the structure of a small mixed farm as
exemplified in Figure 1.3.
Element 1, system boundaries: Depending on the purpose of analysis, the farm-household system
may be specified with different boundaries. In Figure 1.3, these are suggested by the shaded circle
around the system which sets it apart from other neighbouring systems and from the larger community
or environment in which it is imbedded.
Element 2, household: As noted at the top of Figure 1.3, the household provides objectives and
management of the farm-household system and, at the bottom, it exists as the primary internal
beneficiary of the system, while distributing some of the system output to external beneficiaries.
FIGURE 1.3 - Interrelationships of Elements in a Simple Farm-household System
Element 3, operating plan: As shown in Figure 1.3, this is determined largely by the household but it
might also be influenced by the requirements of external individuals, agencies or other influences,
some of whom might be (external) beneficiaries of the system as outlined below.
Element 4, resource pool: This element consists of resources which are initially present at the time
of planning or commencing operation of the system - some pool or stock of land, water, seed, cash
etc. which the other elements of the farm system may draw upon. Once the system begins operating,
certain components of it (the resource-generating activities and by-products of the enterprises) will
replenish the pool. In the schematic sketch of Figure 1.3, the arrows from the farm resource pool
indicate that items from the resource pool flow to processes as well as to activities and to enterprises
(as well as possibly to maintenance of the whole-farm service matrix). Strictly speaking, resources
should be shown as flowing directly only to processes, since this is the level (subsystems of Order
Levels 1 and 2) at which they are actually used. But from a practical viewpoint and because most of
the potential processes are actually ignored in planning the operation of a farm system, resources may
also be viewed as flowing directly to activities and enterprises as indicated.
Element 5, final-product enterprises: Only four enterprises are shown in the system of Figure 1.3:
wheat, paddy (rice), cotton and fish. Only the flow lines of inputs to and output from the paddy crop
are shown.
Element 6, resource-generating activities: In the example of Figure 1.3, there are two of these
supplying some resources to paddy (e.g., a prior fertility-generating legume crop and possibly a cattle
activity providing oxpower).
Element 7, agro-technical processes: These underlie the activities and enterprises. Only three are
indicated as relevant to paddy in Figure 1.3 but (as discussed in Chapter 5) there are a very large
number of biological and mechanical processes actually present in any form of agricultural
production.
Element 8, whole-farm service matrix: This was defined previously in relation to Figure 1.2 as a
system of Order Level 9. It consists of fixed farm capital which provides a flow of services (not
shown in Figure 1.3) to all other elements of the system, particularly to Elements 5, 6 and 7 but it is
not specific to any one of them.
Element 9, structural coefficients: These are not depicted in Figure 1.3. They are discussed in
Chapter 5.
Element 10, time: In the schematic example of Figure 1.3, the time dimension is not specified
explicitly but the model would probably refer to a single operating phase with a duration equal to the
life of the longest-term enterprise subsystem, here cotton with a term of six to seven months, or, if
climatic seasons are constraining, to a full seasonal cycle of one year. If the system proves to be a
'good' system in terms of household objectives, it might be reactivated in successive phases and
continue indefinitely; if a 'very good' system it might permit further farm intensification and
development; if a 'bad' system it might be restructured (as discussed in Chapters 6 and 9); if a 'very
bad' system it will, in the absence of restructuring, prove to be non-sustainable and eventually
collapse.
Figure 1.3 is intended only as a schematic example to indicate the broad relationships which the
elements of a farm-household and its associated whole-farm system bear to each other. Real small-
farm systems are much more complex than Figure 1.3 suggests, particularly in their internal cycling of
resources. The following chapters are essentially an elaboration of Figure 1.3 in terms of managerial
considerations.
Outputs of the farm component of the farm-household system depicted in Figure 1.3 are, as an
example, shown as flowing from the paddy enterprise. They do not comprise a separate element but
are part of Element 5 (final-product enterprises). Outputs are allocated to the farm component and to
system beneficiaries. Thus, to enable continuation of the system in subsequent phases (seasons or
years), some paddy grain output is cycled back to the resource pool to replenish the seed stock.
Similarly, to replenish the cash resource which was used up in this phase, some rice grain is sold.
Whatever grain is left might also be sold to generate cash income or be used as food. These latter
outputs will be distributed among system beneficiaries who might be internal to the system, i.e., the
household, or external to it. External primary beneficiaries may consist of needy relatives and friends,
landless neighbours, a landlord receiving cash or part of the crop as rent etc. External secondary
beneficiaries, as discussed in Section 3.2, might also be present - usually the tax collector. External
beneficiaries often also play another role: they can exercise a direct influence on how the system is
planned, constructed and operated. This influence is indicated in Figure 1.3 by the broken line
indicating feedback from external beneficiaries to Element 3, formulation of the farm plan. Returning
to the paddy flow column of Figure 1.3, the system will also generate a second type of output apart
from the material one of grain, namely non-material output. This consists of the knowledge and
experience gained from operating the system over any phase. Farm-household systems are dynamic:
the results of one operating phase flow back not only as products giving sustenance and cash income
to the household but also as knowledge and experience to influence the formulation of plans for future
phases.
1.6 REFERENCES
Ackoff, R.L. (1973). 'Science in the Systems Age: Beyond IE, OR, and MS', Operations Research
21(3): 661-671.
Ackoff, R.L. and F.E. Emery (1972). On Purposeful Systems, Tavistock, London.
Boulding, K.E. (1956). 'General Systems Theory - The Skeleton of Science', Management Science 2:
197-208.
Cederroth, S. (1995). Survival and Profit in Rural Java: The Case of an East Javanese Village,
Curzon Press, Richmond.
Chambers, R. and B.P. Ghildyal (1985). 'Agricultural Research for Resource-poor Farmers: The
Farmer-first-and-last Model', Agricultural Administration 20(1): 1-30.
Checkland, P.B. (1981). Systems Thinking: Systems Practice, Wiley, Chichester.
Clayton, E. (1983). Agriculture, Poverty and Freedom in Developing Countries, Macmillan Press,
London.
Dillon, J.L. (1992). The Farm as a Purposeful System, Miscellaneous Publication No. 10,
Department of Agricultural Economics and Business Management, University of New England,
Armidale.
Dillon, J.L. and J.R. Anderson (1990). The Analysis of Response in Crop and Livestock
Production, 3rd edn, Pergamon Press, Oxford.
Duckham, A.N. and G.B. Masefield (1970). Farming Systems of the World, Chatto and Windus,
London.
FAO (1989). Farming Systems Development: Concept, Methods, Applications, Food and
Agriculture Organization of the United Nations, Rome.
FAO (1990). Guidelines for the Conduct of a Training Course in Farming Systems Development,
FAO Farm Systems Management Series No. 1, Food and Agriculture Organization of the United
Nations, Rome.
Fresco, L.O. and E. Westphal (1988). 'A Hierarchical Classification of Farm Systems',
Experimental Agriculture 24: 399-419.
Friedrich, K.-H. (ed.) (1992). Readings in Farming Systems Development, Food and Agriculture
Organization of the United Nations, Rome.
Grigg, D.B. (1974). The Agricultural Systems of the World: An Evolutionary Approach, Cambridge
University Press.
Kast, F.E. and J.E. Rosenzweig (1974). Organization and Management: A Systems Approach, 2nd
edn, McGraw-Hill Kogakusha, Tokyo.
Kostrowicki, J. (1974). The Typology of World Agriculture. Principles, Methods and Model Types,
International Geographic Union, Warsaw.
Norman, D.W. (1980). The Farming Systems Approach: Relevancy for the Small Farmer, Rural
Development Paper No. 5, Michigan State University, East Lansing.
Rhoades, R.E. and R.H. Booth (1982). 'Farmer-back-to-farmer: A Model for Generating
Acceptable Agricultural Technology', Agricultural Administration 11(1): 127-137.
Ruthenberg, H. (1976). 'Farm Systems and Farming Systems', Zietschrift für Ausländische
Landwirtschaft 15(1): 42-55.
Ruthenberg, H. (1980). Farming Systems in the Tropics, 3rd edn, Oxford University Press.
Shaner, W.W., P.F. Philipp and W.R. Schmehl (1982). Farming Systems Research and
Development: Guidelines for Developing Countries, Westview, Boulder.
Spedding, C.R.W. (1979). An Introduction to Agricultural Systems, Applied Science Publishers,
London.
von Bertalanffy, L. (1973). General System Theory: Foundations, Development, Applications,
Penguin, Harmondsworth.
Walker, T.S. and J.G. Ryan (1990). Village and Household Economies in India's Semi-arid
Tropics, Johns Hopkins University Press, Baltimore.
2. FARM MANAGEMENT AND FARM
TYPES
2.1 FARM MANAGEMENT
2.3 REFERENCES
'He that by the plow would thrive

Himself must either hold or drive.'
Benjamin Franklin, 1706-1790
This chapter introduces the discipline of farm management as it applies to the main structural types of
Asian farms, particularly small family farms (Types 1 and 2 of Section 2.2.1).
2.1 FARM MANAGEMENT
2.1.1 Scope
2.1.2 Definition
2.1.3 Optimization
2.1.4 Objectives
2.1.1 Scope
To a layman 'farm management' probably means just that - a body of activities and procedures carried
out by a farmer in the ongoing management of his or her farm and for which advice may be available
from professional specialists in farm management.1 To an extent this is correct (as per farm
management in Field A of Section 2.1.7) but, more broadly, farm management as considered here is a
professional discipline which relates to the description, construction, analysis and evaluation of farm
systems of Order Level 10 (Figure 1.2). In this wider sense, farm management is the discipline within
whose ambit farm-level systems analysis most clearly falls. (This does not exclude from farm systems
analysis other disciplines of a technical or special-purpose nature.) Farm management system
analysis can have several operating objectives (Section 2.1.4); there are several alternative bases
(Section 2.1.6) on which analysis can proceed; and it can operate in four fields (Section 2.1.7) and in
four modes (Section 2.1.8) within these fields. These several aspects of farm management as a
systems-related discipline are now briefly discussed in turn.
1 Farm management as carried out by farmers has been defined (Dillon 1980, p. 258) as 'the process by which resources and
situations are manipulated by the farm manager in trying, with less than full information, to achieve his [or her] goals'. See also
Makeham and Malcolm (1986, Ch. 2) and Upton (1973, Ch. 1). The appendix to this text gives the authors' perspective on farm
management as it relates to management per se and to farm systems theory.
2.1.2 Definition
Except when it serves a descriptive purpose, farm management is the science (and art) of optimizing
the use of resources in the farm component of farm-households, i.e., in systems of Order Level 10
(see Figure 1.2), and of achieving the optimal functioning of these systems in relation to household-
specified objectives; and since Order Level 10 systems consist structurally of subsystems, farm
management is also concerned with the operation of subservient subsystems of Order Levels 1 to 9 in
such fashion as to optimize the whole-farm system. However, for reasons discussed in Chapter 1, it is
often essential, especially when dealing with small farms, that farm management extends also to the
family or household component, thus its true scope extends to Order Level 12 systems. A second
consideration is that the village is sometimes a more relevant unit for analysis than the farm, and
where this is so the scope of 'farm' management extends to systems of Order Levels 1 to 13 as
discussed in Section 1.3.2.
2.1.3 Optimization
Optimization of the planning objective is defined as achieving the farm household's goals as
efficiently as possible in the face of whatever constraints of a physical, environmental, legal or socio-
cultural nature may be relevant. This implies obtaining maximum possible net benefit over time from
the operation of the farm system. Net benefit is measured, as appropriate, in terms of output or profit
or, more broadly, as satisfaction or utility. Maximization of net benefit implies efficient use of
available resources and opportunities. For the achievement of a given level of net benefit, it implies
the minimization of costs. This reflects a theoretical view. In the real world, as discussed in Chapter
6, the general objective is often constrained by household and social factors other than availability of
physical inputs and their costs. Thus many small-farm households place a high value on the long-term
sustainability of their farm system. Also, in the real world, uncertainty will generally prevail about
yields, prices and other relevant influences so that the farmer's choice will lie not between sure
alternatives but between alternative (subjective) probability distributions of net benefits. This aspect
is considered in Chapter 11.
Optimization can occur at two levels: local or global. When operating in Field A (on-farm problem
solving - see Section 2.1.7 below), farm management will seek optimization at the global level of the
Order Level 12 farm-household system. This sets it apart from other farm-related agricultural
sciences which are usually (though not always) concerned primarily with optimization of lower order
subsystems, i.e., local optimization. Two examples will clarify this point. First, a farm might involve
only two irrigated crops, cotton and sugarcane. If only the cotton is considered, the local optimum
might be to use all of the water supply on cotton, but if the farm as a whole is considered, i.e., a
global optimum sought, this might well require that the water be shared between both crops. Second,
a farm-household system itself might be only a subsystem within some larger system. For example, the
optimal irrigation water supply to a farm might, from the viewpoint of the farm, be 1 000 m3, but if
such water is to be provided only as a minor by-product by a large multipurpose dam project (the
chief purposes of which are power generation and flood control), these purposes will determine what
discharge rate/farm supply is optimal from an overall project or global viewpoint and this would
override whatever supply rate might be optimal from a farm perspective.
2.1.4 Objectives
Optimization of farm-household systems takes the form of conditional maximization over time of the
socioeconomic welfare of farm families. The term 'welfare' is used broadly to include money income,
sustenance food, farm-produced consumption goods and factors of production, non-material benefits
such as those enabling the attainment of education and health standards, and satisfactions derived
from work well done as well as from cultural and religious sources. Whichever of these system
outputs/family benefits are relevant in a particular farm situation will depend on the farm type and on
the values held by the particular family - values which will normally reflect the society and cultural
context in which the farm-household exists. Welfare maximization is conditional because it is
constrained by resource availability and, as relevant, legal constraints and socio-cultural mores.
Typically the farm plays only an enabling role towards achieving broad family goals. Thus farm
management is concerned with conditional optimization of only part of a farm-household system - but
usually the most important material part. The specific objective might be to maximize money profit or,
recognizing the presence of uncertainty, to maximize the expected utility (Chapter 11) of risky profit
(farms of Type 4, 5 and 6 and possibly Type 3 as defined in Section 2.2.1), or such money profit
might be only incidental to other objectives (Type 2 farms), or hardly relevant at all (Type 1 farms).
Goals and objectives are discussed in Chapter 6.
Economics or economic analysis is the science of making choices so as to best achieve desired
objectives given that only limited (physical and other) resources and opportunities are available and
that the future is uncertain. There are no choices to which the science of economics cannot be applied.
It is just as pertinent, e.g., to the choice of a spouse as to the choice of which crops to grow or to the
choice between using an insecticide or using environmentally friendly integrated pest management. In
contrast to this wide applicability of economic analysis, financial analysis is restricted to matters
that are naturally of a financial or monetary nature. Financial analysis is thus a subset of economic
analysis and, in circumstances where everything is valued in money terms, may be the natural way in
which to conduct economic analysis. In other cases, it may be feasible to facilitate economic analysis
of possible choices by imputing money values to possible gains and losses. And in yet other cases,
such as assessing the resource sustainability and environmental compatibility of alternative farm
systems, it may often be infeasible to impute money values to the gains and losses of alternative
choices. Decisions must then be made using economic analysis based on non-money values, intuition
and judgement.
Farm management economics (i.e., economic analysis applied to the choices confronting farmers)
provides the general disciplinary basis for farm-level systems analysis (Dillon and Hardaker 1993,
pp. 22-30; Upton 1973, Ch. 1). Obviously other farm and family-related disciplines will be involved
in systems' construction: agronomy, animal husbandry, soil and water conservation/management,
human nutrition etc. However, except in the case of special-purpose technical systems (e.g., when the
farm-household unit is analysed in terms of nutritional or energy flows among components as
discussed below), these other disciplines should play subordinate contributing roles coordinated by
farm management economics as the lead discipline. That in fact this often does not happen and the
lead is taken instead by workers in other disciplines is really not important. It might just reflect the
fact that many agriculturists are aware of the necessity for a systems approach if application of their
expertise is to be effective; or that many agricultural economists are content in the more modest role
of economics apparachnik.
Nevertheless, the disciplinary basis of farm management remains economics - but economics of a
special wide-ranging kind, the core of which is production economics supported by other branches of
economics of which marketing, resource economics, agricultural credit and data analysis (including
operations research, econometrics and risk analysis) are probably the most important. When working
with the household component, especially of small traditional farms, the most important supporting
disciplines are sociology and social anthropology.
There are several reasons why farm economics provides a good conceptual framework for most
farm-household systems analysis. The most important of these is the necessity to bring the many
relationships of a system and between systems to some common unit or basis of comparison. Unless
this is done, systems analysis and the comparison of alternatives will not be possible. The base
usually most convenient - and in the case of commercial farm systems most relevant and which has the
highest degree of universality - is money or financial value. But several other bases for systems
analysis are possible and in certain circumstances they might well be more relevant than money
value. The four most important bases of comparison are as follows:
(a) Money value: The convenience of using money or financial values as the basis of commercial farm systems analysis will be
obvious: it permits the various system inputs (e.g., seed, fertilizer, power, labour etc.) to be standardized as money costs and the
various system outputs to be standardized as money returns so that net revenue, i.e., money returns minus money costs, can be
used as the basis of comparison between alternatives. In commercial farming, all or most of these inputs/costs and
outputs/revenues can be stated in explicit quantitative terms. On the other hand, when dealing with less commercialized systems
where there are no actual price-setting markets for farm inputs and/or outputs, one is often obliged to base the analysis on
imputed values. However, this is possible only up to a point; beyond this point, as one moves further from a commercial
environment towards a traditional one (as discussed in Chapter 6), the attempted use of money value as the basis or numéraire
for analysis becomes too abstract to be useful and one has to search for some other base (Dillon and Hardaker 1993, pp. 28-30).
(b) Family labour effort: Probably the best alternative to money value on small family farms of a
subsistence or semi-subsistence nature is labour input, both as a measure of inputs and as a yardstick
to judge the worth of outputs. At least this is so in the eyes of the majority of Asian (and African)
small-farm families for two reasons. First, on these farms most production activities involve few if
any commercial inputs and most outputs are also not disposed of through commercial channels.
Money hardly enters into the matter at all. What such activities do have as their common factor is
family labour - often very hard labour - from hand-preparation of fields, to carrying all inputs/outputs
perhaps long distances, to hand-pounding the harvested grain. Not unnaturally then, these families
plan, compare and evaluate their several different farming activities and alternatives (i.e., analyse
their systems) in terms of labour content. To conduct such analysis on any other basis such as money
value would be an incomprehensible abstraction. However, 'labour' is not a simple quantity. It can
have several dimensions: quantity when labour is measured in terms of standardized units (e.g.,
labour-days or task-days on estates); quality where the relevant factor is the actual effort required or
the degree of skill or unpleasantness associated with separate tasks; and agency where the labour
measurement reflects the social position or status of the person performing the task. Thus, in different
societies, patriarchal or matriarchal, women's labour will be valued less or more highly than the
labour of men regardless of the actual effort expended, while the labour performed by children might
also be valued according to their (usually inferior) social status rather than to the actual work they
perform. These dimensions of labour and the implied difficulties of measurement often limit the use of
this factor as an alternative to money value. Nevertheless labour often provides a more relevant basis
for systems analysis of a very large number of small traditional farms than does money.
(c) Bio-mechanical energy: A factor which all farm-household systems and their subsystems have in
common is their explicit or implicit energy content (including labour, above). Farm-system models
have sometimes been structured on the basis of such energy content and inter-component energy flows
- see, e.g., Axinn and Axinn (1983). Use of energy-based farm systems analysis rests on the view that,
in a world of declining energy resources and materials that can be represented by their energy
content, the energy generation and consumption of farm-household systems is a more valid basis for
systems analysis than is money profit, and usually also that energy flows which are directly or
indirectly involved in all economic activities (including agriculture) are not properly represented -
indeed they are often severely distorted - by commercial pricing mechanisms. However, these views
involve issues and require solutions at much higher than farm level. Farm systems analysis based on
energy flow is more appropriate for some aspects of macro/industry/sector strategic planning than for
farm-level operational planning where the immediate interest of farm families is in income (in
whatever form it takes) and the effort required to achieve it.
(d) Water consumption: A fourth possible basis for analysis of farm systems is offered by water with
systems analysis conducted in terms of the relative water consumption of different crops and animal
populations and the implicit water content of products and by-products. Water is obviously the
critical common factor in all the farming systems of that great belt of lands stretching from North
Africa to India, so much so that even the very wealthy Gulf States, while they have been able to
import or create all other agricultural resources, including soils, micro-environments and farmers,
remain constrained by water. Moreover, in the 'wet' tropics, the critical nature of this input common
to all parts of all farming systems is not yet widely recognized; e.g., even 'well-watered' Java will
probably exhaust its water supplies before its soils. However, as important as water is, like bio-
mechanical energy it is more appropriate as a basis for some aspects of macro-level systems analysis
than for operational-oriented systems analysis at farm level.
In summary, except when used in connection with special-purpose systems, such bases of analysis as
energy, water, ecological balance etc. lack the universality and the value orientation required of a
general systems base. Money value and labour will probably continue to be used as such a base,
either separately in the case of commercial and near-subsistence farms respectively, or jointly in the
case of the bulk of small traditional partly commercialized farms.
Farm management analysis and advisory activities can be categorized in terms of four fields defined
in terms of the purpose of the analysis as follows:
Field A consists of those problems and analyses which are only or primarily of direct interest to the
farmer subjects of the analysis and where solutions to problems are offered on the basis of their
beneficial effect on the welfare of these farmers and their families. The great bulk of farm
management systems analysis occurs within this field. Field A is the conventional area in which farm
management operates, directed to solving the on-farm problems of individuals and groups. Except
where otherwise noted, this book is concerned with farm management within Field A; it needs no
further discussion at this point except to note that such analysis should, whenever possible, involve
farmer participation so as to ensure that the farmer's felt needs are considered (Ashby and Sperling
1995; Chambers 1983; Chambers and Ghildyal 1985; Matlon et al. 1984; Mikkelsen 1995; Rhoades
and Booth 1982; Tripp 1991; Werner 1993).
Field B consists of those problems and analyses which should not really fall within Field A (i.e., they
do not properly constitute farm-level problems) but which for convenience or purposes of analysis
can be defined, regarded and treated as if they do. Examples of the scope of Field B farm management
analysis are offered by the agricultural industries and sectors of some of the mini-states. For example,
the agricultural sector of the island nation Kiribati is equivalent to not much more than a single good-
sized coconut estate with a few supplementary enterprises added. This sector (a system of Order
Level 16) could easily and probably most effectively be analysed, of course with the necessary
modifications, as if it were a system of Order Level 10 or 12. An example at industry level is offered
by the banana industry of Western Samoa which, although it consists of a large number of individual
farms, has been centralized (through the government) in fruit collection/inspection/transport/export
and other important aspects. The industry could justifiably be analysed as a single large 'farm system'
even though in fact (and in respect of banana-growing activities) it really consists of many farm-
household systems.
Obviously, since this type of higher-than-farm-level analysis will be concerned with a range of
subject matter in addition to farm economics - processing, marketing, transport, research, extension
etc. - farm management can operate successfully in Field B only if the analyst can ignore artificial
divisions which are conventionally imposed between the various disciplines. Another condition is
that the analysis could not be better performed by a systems analyst working within the conceptual
framework of some other discipline. (If this is the case then farm management analysis would operate
in a subservient role in Field C.)
Field C consists of problems or issues arising within or in relation to higher systems of Order Levels
13 to 16, towards the resolution of which farm management plays only a secondary contributing or
partial role, e.g., the lead discipline might be water resource engineering if the problem is planning of
a public irrigation project, or agronomy if it consists of planning for the introduction of new crops, or
agricultural credit if it is to plan the establishment of a farmers' bank. In this type of supporting role,
farm management can operate in any or successively all of Modes 2, 3 and 4, i.e., description,
diagnosis and prescription, respectively, as defined in Section 2.1.8 below. However, any
'prescription' that is offered will be of a limited kind and fall short of being a plan for the overall
project or program. Analysis will be directed towards the achievement of some global optimum
which is not defined in terms of farm management itself. A few of the very large number of situations
in which farm management operates in Field C are:
(i) Descriptive studies of farm-household systems to provide background for local or multi/bilateral investment programs in
agriculture and/or agricultural infrastructure (e.g., the Country Background Reports of the World Bank). This type of analysis is in
Mode 2 (Section 2.1.8 below).
(ii) Diagnostic studies of farm-households to determine just what developmental or investment

assistance is needed (e.g., roads, health, transport, extension, credit etc.), and the priority ordering of
specific projects to provide such assistance (Mode 3).
(iii) Prescriptive analysis (i.e., Mode 4) aimed at providing the farm-related part of some uni- or
multi-purpose project or plan: e.g., farm-level demand schedules for irrigation water in a multi-
purpose water storage project (where irrigation is only one planned activity in addition to power
generation, flood control, fish production etc.); scheduling of produce supply as part of a feasibility
study for the establishment of a cannery.
Field D consists of farm management in the role of generating data for the guidance or support of
agricultural policy making. Provision of such data might not require special studies or systems
analysis for this particular purpose; often such data will be an incidental output of analysis undertaken
for some other purpose, e.g., in Field A. Field D analysis is also of a supportive kind and operates in
Modes 2 and 3 (quantitative description and diagnosis as outlined in Section 2.1.8 below). The aim is
usually to generate knowledge about farm-households or their component subsystems which is to be
used by governments, public agencies etc. as a basis for structuring agricultural or broader economic
policies - setting farm-input prices or consumer food prices, establishing transport services or credit
programs etc. (Dixon et al. 1994; Upton and Dixon 1994). These policies might imply either
enhancing farmer welfare or reducing it (e.g., if their thrust is to minimize urban living costs). This is
a very important and wide-ranging field: it is difficult to think of any policy which is to affect farmers
which should not be based at least in part on farm-level analysis, despite the fact that such farm-level
analysis is in fact frequently not carried out, much to the detriment of sound policy making.
Farm management operates in four modes within the above fields.
Mode 1 encompasses routine operational and control activities. It is concerned with the day-to-day
operation and management of an actual farm, estate, cooperative or other farm-based
producing/marketing entity. This may be thought of as practical or 'muddy-boots' farm management.
Management in this mode is largely outside the scope of the present discussion, except that the
systems concepts discussed here will, it is hoped, provide principles to guide practical (i.e., Mode 1)
management.
Mode 2 refers to descriptive activities whereby farm management provides a conceptual framework
for the study, understanding and description of farm systems or farm-related problems. This might be
an end in itself; or more likely it will be a necessary stage in the logical-event sequence towards
action, as suggested in Figure 2.1. The chief function of descriptive farm management studies is to
provide a basis of understanding before problem diagnosis is attempted. There are still many
societies in the world with farm-household systems of which we are in nearly complete ignorance;
understanding and description of these systems must precede problem diagnosis and, if need be,
prescription of solutions.
Mode 3 refers to diagnostic activities concerned with the identification of problems and weaknesses
in farm-level systems of all Order Levels 1 to 10 and those parts of Order Level 11 household
systems relating to the farm. Such problem diagnosis includes the identification of potential
opportunities. Problem diagnosis is usually carried out as a separate mode, but on some commercial
farms it might be built into their routine monitoring and management mechanisms (as also on more
sophisticated estates).
Mode 4 refers to prescriptive activities in which farm management is aimed at the prescription of
action plans for both (a) the overcoming of problems or weaknesses and (b) the seizing of
opportunities uncovered in Mode 3 (diagnostic) analysis.
Thus farm planning, as discussed in Chapters 8, 9, 10 and 11, is a prescriptive (Mode 4) activity
based on the descriptive (Mode 2) and diagnostic (Mode 3) activity of farm evaluation as discussed
in Chapter 7.
Analytical situations within modes
In Modes 3 and 4, three analytical situations will arise, viz.:
(i) Diagnose and prescribe: First, the problem might require that a diagnostic analysis be made of a system of Order Level from 1 to
10 (or 1 to 12) leading to the identification of some specific weakness or opportunity to be investigated by research on the farm or
experiment station - see Norman et al. (1995). If the problem falls within the competence of the investigator, the analysis would
at this point go into prescriptive Mode 4 to develop and offer solutions.
(ii) Diagnose and refer: This second situation arises when the diagnosed problem lies beyond the
competence of the analyst: e.g., low milk production on a dairy farm might be due to animal disease
or a genetic factor or to the household itself (such as a low educational level leading to product
adulteration). In such situations the role of the farm analyst is - or should be - to refer the problem to
some relevant agency or specialist, i.e., to diagnose but not to prescribe.
(iii) Prescribe only: Often only a prescriptive analysis is called for: e.g., if the problem is to
generate land-use plans intended to serve as the agro-economic basis of new settlement projects or
transmigration schemes.
FIGURE 2.1 - Relationship between the Four Modes of Farm Management Activity
2.2.1 Farm types

To date, farms have most often been classified on the basis of agro-ecological factors (such as
climate, soil, slope, altitude and, not unrelated to these factors, the crop and livestock systems used)
overlaid, to a lesser extent, with socioeconomic criteria (Fresco and Westphal 1988). Inevitably such
an approach leads to a plethora of farm types. A different approach is taken here. Emphasis is on
farm-system structure from a farm management and farm-household perspective with classification
based on: (1) the main purpose of the farm, (2) its degree of independence and (3) its 'size'. From
such a structural viewpoint there are basically six major types of farm system to be found in Asia and
elsewhere around the developing world with dozens of subtypes constituting a continuum of farm
types between the extremes of a totally subsistence to a totally commercial orientation.
2.2.1 Farm types
The six basic farm types are:
Type 1. Small subsistence-oriented family farms.
Type 2. Small semi-subsistence or part-commercial family farms, usually of one half to two
hectares, but area is not a good criterion: the same basic structure can be found on much larger 20- to
30-hectare farms as in the Punjab, Sind, and North West Frontier Provinces of Pakistan.
Type 3. Small independent specialized family farms.
Type 4. Small dependent specialized family farms, often with the family as tenants.
Type 5. Large commercial family farms, usually specialized and operated along modified estate
lines.
Type 6. Commercial estates, usually mono-crop and with hired management and absentee ownership.
Each of the six farm types is now discussed in turn.
Type 1: Small subsistence-oriented family farms
There are two main subtypes. First, and of lesser numerical importance, are those based on only one
or two crops or livestock types (e.g., on maize or cassava or coconuts; or on yaks or camels). Some
farms of this subtype are based more on exploitation or management of a local natural resource - in
the extreme case, by use of shifting cultivation or by nomadism - than on deliberate choice of their
main farm enterprise (e.g., on indigenous sago palm, palmyrah, coconut or nipah). However, the main
group of Asian subsistence-oriented farms is based on a wide range of crops and animal types. This
second subtype is of necessity more highly mixed than are Type 2 part-commercial farms. Farms
which are completely self-sufficient are rare, but self-sufficiency remains the operating objective and,
if forced by circumstances, farms of this type could exist in isolation from the outside world. The
structure of a Type 1 farm is exemplified in Figure 2.5 below. The focus for evaluation and analysis
of Type 1 farms is the household rather than the farm component of the system. However, Type 1
farms have most of the characteristics of Type 2 farms and these are discussed below in relation to
this latter type.
Type 2: Small semi-subsistence or part-commercial family farms
This type is predominant throughout South and South East Asia in terms of the number of such units,
the large number of people supported by them and the total volume of their production - especially of
basic foodstuffs.
Operating objective: The general operating objective of this farm type is family sustenance, pursued
first by production of foodstuffs for consumption and of produce/materials for use on the farm, and
second by generation of some cash income for the purchase of (a) non-farm produced food essentials
(salt, tea etc.); (b) other essentials such as clothing, medicines, transistor radio, batteries etc.; and (c)
some farm inputs (such as agricides and fertilizer). Such cash is obtained primarily by sale of
commodities which are surplus to family requirements, and secondarily - where this is possible - by
production and sale of some cash crop raised specifically for this purpose. The comparative
operating objectives of this and other farm types are discussed in Chapter 6.
Production activities: Type 2 farms can be further classified according to geographical occurrence
(e.g., wet tropics; sub-tropics of India/Pakistan; temperate zone of North India, Nepal, Bhutan), and
by whether these farms are dryland (as most are) or part irrigated. However, they are all basically
similar in their crop activities which consist essentially of one or more staple food crops plus a
leguminous protein source plus an oil crop (see Section 9.6.3). Some examples of geographically
typical crop mixes are:
Himalayan hills: barley or buckwheat; beans or peas; mustard.
sub-tropics: wheat; soybeans; sesame.
wet tropics: cassava or rice; groundnuts; coconut or sesame.
In general, land is cropped to its maximum intensity, but the number of crop species grown in each of
the energy/protein/oil crop subgroups, as well as the area of each, is limited by length of growing
season (dependent upon temperature), rainfall occurrence and/or irrigation water supply.
Livestock, whether fish, poultry or larger animals, are typically important on Type 2 farms. They are
closely integrated with the crop activities, and here - unlike the situation on farms in developed
countries - they are kept for a range of purposes: direct production, draught power (except on the
smallest farms), transport, manure production to sustain field and pond fertility levels, and as a store
of wealth. The combination of livestock with crops results in a large number of activities, and an
even larger number of different farm products.
A special subtype of this highly-mixed farm type consists of the forest-garden farms of the wet tropics
as found in Kerala, Sri Lanka, Malaysia and Indonesia. These consist of both whole farms, e.g., the
forest-garden farms of Kandy in Sri Lanka - see McConnell (1992) - or the house-yard parts of farms
forested with a more or less dense mix of economic species as in the Pekarangan lands of Java.
Except for poultry, livestock are relatively unimportant on this subtype.
Farm system boundaries: Discussion of system boundaries in Chapter 3 mainly relates to farms of
this type. Briefly, boundaries of Type 2 farm systems (and of Types 1, 3 and 4) segregate them
distinctly from the external world, but the boundaries between individual farms are relatively weak.
In contrast with farms in developed countries, they often have much stronger links to and
interdependence with other farms in the local community than they do with the outside world, i.e.,
with commercial input suppliers and markets for their outputs. Through such practices as exchange of
labour and animal services among farms, group farming, community work ('gotong royong' in
Indonesia) to develop irrigation channels, village roads etc., barter of food grains for animal
products, and village-level savings' mobilization ('arisan' in Indonesia), each farm-household unit is a
link in an often highly developed agro-socioeconomic network within the local community. Whatever
the basis for such informal integration - culture, religion, isolation - its effect is to provide strong
structural boundaries around groups of farms, hamlets and villages rather than around individual
farms. Each Type 2 farm is very much a part of the community and often could not function effectively
if divorced from it. It might well be said that 'No such farm is an Island, entire unto itself.'
Activity and product diversity: Diversity, or the degree to which farm income (however measured) is
derived from a range of activities and products rather than from a single source, is discussed in
Section 6.2.4. Type 2 farms are typically the most diverse of all farms. Diversity has three elements:
the number of crop/livestock activities present; the number of products obtained from those activities;
and the number of ways in which each product can be used or disposed of.
The mixed farms of the Punjab commonly consist of four to six crop activities and three to six
livestock activities; those of Bhutan somewhat fewer. In Sri Lanka the forest-garden farms commonly
grow up to 16 or so tree and vine species producing 20 to 30 different products; and since each can
be disposed of in up to four different ways (sold/bartered, consumed, processed, stored), such a
system will possibly generate some 60 or so end products. This contrasts sharply with the situation on
farms of Type 6, the estates producing a single product (tea, rubber etc.) with a single end use (sale).
Even a common field crop such as maize may be managed so as to yield four or five primary products
(green pick, dry grain, fodder leaves and stalks, fuel, live stripped stalks as supports for a companion
bean crop), and two or three subsequent processed products (maize cakes - an important kitchen
industry in parts of Bhutan, alcohol etc.).
Diversification of Type 1 and 2 farms has several bases. Broadly it follows from their sustenance
orientation. In remote tracts of Nepal and Bhutan it is a necessity. In the Punjab it results largely from
the possibility of growing a wide range of summer and winter crops and combining these with
livestock. On the closely integrated vegetable-poultry-pig-fish farms of West Malaysia and Sarawak
it results from a business-like approach to profit maximization. On the forest-garden farms of Kandy
in Sri Lanka it results partly from historical circumstance (the wide botanical base provided by
Indian, Arab, Portuguese, Dutch and British immigrants and colonizers) and partly from a tendency of
the Kandyans to plant a tree/vine/shrub in any vacant space. (If they did not, the space would soon be
filled anyway through natural seed fall and germination.)
Sources and uses of farm resources: An important characteristic of Type 2 farms (and of farms of
Type 1) is the high proportion of farm and household resources generated on the farm and,
correspondingly, the low level of dependence on purchased inputs. Further, where purchased
resources are used, it is common practice to restrict their use to cash crops (cotton, sugarcane,
tobacco etc.) with non-cash crops receiving no purchased inputs or being grown on the residual
fertilizer (and often soil tilth and soil moisture) from some previous commercial crop. Those farms
which do use purchased inputs often operate at dual levels of technology - 'advanced' for some main
crop, 'traditional' for the rest. Farm-generated resources including food supply (as distinct from
purchased resources) are obtained in a wide variety of ways as follows:
· By operating separate specific-purpose resource-generating activities. These, as distinct from cash-generating activities, are
relatively important on farms of Types 1, 2 and 3. The most common example is livestock kept primarily for manure production (as
well as for other purposes). Growing a green manure crop serves a similar purpose. Growing and lopping the leaves from
leguminous trees for paddy fertilizer is still common in Java. Such activities need not be elaborate: in Bali and Sri Lanka the most
common 'resource-generating' activity is simply growing a clump of bamboo in the house-yard (for construction material, produce
containers, fences, water pipes).
· Through simple apportionment of part of a commercial crop for household use as food or livestock
feed.
· By carrying on parallel crop activities by growing one variety/type for the market and another for
the family's own use. The first, typically a high-yielding improved variety, might be deficient in taste
and storability but will generate cash. The second might be capable of long storage and possess other
qualities valuable in rural but not in sophisticated urban markets.
· Through extensive use of all by-products: stalks of maize, tobacco, pigeon pea, cotton etc. as
household fuel; wheat straw for mud brick making; etc. With some crops - especially tree crops - the
'by-products' may take on such relative importance in the range of their uses as to make unclear just
what the 'main' crop is.
· By growing/keeping generally low-yielding but multi-purpose types of crops and livestock rather
than high-yielding specific-purpose types. Maize, wheat etc. might be grown nearly as much for
green/dry animal fodder as for grain, or - in the case of maize - as a standing trellis for some
following climbing crop, e.g., beans or cucumbers. (Thus the not uncommon observation that farmers
are 'backward' because they do not adopt 'improved' varieties is often based on ignorance as to the
real reasons why the 'old' varieties are grown.)
· Finally, resources are also generated by the multiple use of farm capital. A common example is
provided by farm boundary and roadside fences. In the Matale district of Sri Lanka most of the fences
consist of kapok trees planted at very close spacing. They also support pepper vines and thus yield
four 'products': kapok floss and seed (for oil) and black pepper, as well as field security. Farm fences
in the Yogyakarta-Boyolali area of Central Java are used to generate a wider range of resources (or
to directly produce a marketable commodity). There are four main types: (a) bamboo-lattice fences,
invariably used also as supports for long-beans; (b) napier grass fences consisting of a single row of
grass, laced together into an upright position by a strip of bamboo about a metre above ground level
which permits the growing top of the fence to be regularly cut for cattle feed; (c) live leguminous
trees whose leaves are harvested periodically for cattle feed or seasonally as field green manure; and
(d) cassava fences, formed by planting cassava very close (15 to 20 cm) and weaving one or two
bamboo strips through the line of cassava stems at about one metre from ground level. The cassava
stems are then used also as a bean trellis.
Following is a partial list of the farm and household resources/inputs/capital equipment commonly
generated on farms of Types 1 and 2. The list suggests the high level of self-sufficiency that
characterizes these farm-system types, especially in isolated areas.
- Labour and power: all labour (except at peak periods, then by labour exchange or mutual help); all ox/buffalo/camel draught
power, grain grinding power and transport; ox/buffalo treading power for grain threshing (mainly of paddy).
- Crop inputs: most seed, animal manure (with a chemical fertilizer supplement on some cash crops);
lopped high-nitrogen tree leaves for paddy fields (wet tropics); packing materials for market (woven
bamboo baskets, teak and plantain leaf wrappers).
- Capital equipment: all ploughs, harrows, rakes and levellers from farm or village timber; all
animal harness and repairs; fences (live kapok, areca palm, napier grass, cassava, woven bamboo);
sugarcane roller crushers; oil extractors; irrigation water delivery systems (bamboo pipes in parts of
Bali and the Madiun Valley of Indonesia).
- Buildings and household: building panels/roofing/flooring (woven bamboo and sago/nipah/coconut
leaves); buckets, containers (areca palm spathe and woven bamboo); cordage, twine, ropes; granary
containers (woven, wood, packed earth); stoves and jaggery sugar boilers (packed earth); household
fuel (crop residues, tree prunings); domestic light (coconut oil, butter lamps).
Type 3: Small independent specialized family farms
The key characteristics of Type 3 farms are (a) their specialization in some particular crop or
livestock activity which distinguishes them from the mixed farms of Types 1 and 2; and (b) their
management independence which distinguishes them from Type 4 farms.
Type 3 farms fall into three subgroups according to their management orientation/purpose and type of
income: (A) commercially-oriented farms, and family sustenance-oriented farms which achieve this
objective through either (B) sale of part of their production (which makes them of necessity part-
commercial farms) or (C) multiple-use of produce from their single specialized activity and/or barter
of some of this produce for necessary commodities/goods which cannot be produced or purchased. In
this latter situation such farms are also a subtype of subsistence farms (Type 1), but differ from the
main body of near-subsistence farms in that only one main production activity is pursued. A sub-
classification of Type 3 farms is shown in Figure 2.2. Some examples of these Type 3 farm subtypes
are noted below. Probably the most important are the Subtype B near-continuous paddy farms of the
wet tropics.
Subtype A (commercial):
· small farms specializing in poultry, pig, dairy or vegetable production around metropolitan areas.
· orchid and horticulture farms.

· vegetable farms in upland areas throughout Malaysia, North Sumatra and Java.
· smallholder rubber, oil palm or pepper holdings in Malaysia and Indonesia.
· citronella and cinnamon farms in southern Sri Lanka.
Subtype B (part-commercial):
· continuous and near-continuous paddy farms of the monsoon lands.
· upland/dryland maize and cassava farms.
· smallholder coffee or cacao farms.
Subtype C (near-subsistence):
· near-subsistence maize farms of East Bhutan, Nepalese hills and Sarawak.
· cassava-based farms on poor soil in South Java.
· sago farms of South East Asia and New Guinea.
· yak/sheep migratory farms of the high Himalayan valleys.
FIGURE 2.2 - Sub-classification of Type 3 Fanning Systems

The specialization of Type 3 farming systems is based on a wide range of factors as follows:
- Subtype A farms: on commercial/profit opportunity and proximity to urban markets (poultry, pig, dairy and vegetable farms). The
development of such profit-oriented peri-urban farm activities is a reflection of economic growth with its demands for intensively
produced products with a high income elasticity of demand.
- Subtype B farms: on the presence of specific physical/geographical factors (water and good soils
for continuous paddy farms; favourable temperature and rainfall regimes for coffee or cacao farms).
- Subtype C farms: on necessity (cassava on the poor soils of South Java; yaks and sheep in the high
valleys of Nepal and Bhutan).
Yet other bases for specialization are historical accident (e.g., smallholder tea in Sri Lanka on lands
acquired from previous tea estates); or prior presence of some natural resource (e.g., the sago and
nipah farms in the coastal swamps of the Philippines and New Guinea).
Type 4: Small dependent specialized family farms
Structurally, except for their lack of independence, Type 4 farms are quite akin to Type 3 farms and
contain the same three (A) commercial (B) part-commercial and (C) near-subsistence subtypes;
however, they are sufficiently important to be examined as a separate type. The characteristics which
set them apart from farms of Types 1, 2 and 3 are their high degree of activity specialization and the
lack of real decision-making power possessed by the farm family. The specialization characteristic
may be based on the same factors noted above for Type 3 (independent specialized) farms. The
dependence characteristic arises from the fact that on Type 4 farms the family is not free to decide
what to produce, nor frequently the conditions under which some obligatory activity is to be carried
on. This lack of independence can be due to several factors, viz.:
(a) Terms of tenancy: Tenant farmers are often obliged to produce one or more specific crop or livestock products, as dictated in a
landlord-tenant agreement. The tenant-operated vegetable farms of Qatar are an example.
(b) Structural integration: In this situation small family farms are integrated more or less closely as
the production arm of some larger farming cum processing system. Small tenant-operated farms
supplying sugarcane to a mill or leaf to a tobacco-processing factory are common examples of such
vertically integrated farms. Not only is the crop which is to be grown specified, but the conditions of
production - timing of planting and harvesting, amounts of fertilizers to be used, spraying programs
etc. - are also dictated by the controlling authority.
(c) Debt: Some agro-industrial units (such as milk processing plants) often provide farmers with
input factors (such as cattle, feed and technical assistance) in order to achieve a regular or higher
quality supply of their needed raw material (e.g., milk). These advances are usually made in the form
of a loan at attractive terms, but often the only way farmers can liquidate this loan (and perhaps
eventually regain their independence) is to continue to produce the particular commodity - usually
under conditions set by and to the relative advantage of the lender.
A second kind of debt, that entered into for consumption rather than production purposes, can also
provide the basis for farmer dependence. Thus for generations the small cardamom farmers of the
southern Bhutan hills have been indebted to the cardamom traders/money lenders of the towns along
the West Bengal border. The only hope these farmers have of liquidating such debt - usually used for
food, clothing and household items - is to continue to grow this specialist crop and sell it at whatever
terms may be offered by the traders.
(d) Government policy directives: In some countries, farmers' lack of independence in production
decision making is the result of government power to issue production directives. In Indonesia, e.g.,
the Government has the broad power to direct that some percentage of those village lands which lie
within the command area of each sugar factory be planted to sugarcane. Typically each hectare of
land, owned by individuals of the village, might be under sugar for one year during which time it is
farmed by the company as part of a larger estate. It then reverts to its owner for three years during
which period he or she will operate it as a complete and independent farm, until it is again taken for
sugar. During this three-year period the farmer has all normal decision-making powers (crop
selection and how each is grown). During the one-year cane phase, for which the farmer receives
payment as a 'landlord', he or she has no decision-making power whatsoever. This system thus
involves the alternating of two distinctly different farming systems (as shown diagrammatically in
Figure 3.1).
(e) Lack of alternative market outlets: Absence of any real independence in management can also be
due to lack of alternative market outlets, especially when the product is too bulky or fragile to be
transported far from the farm. For example, most of the small cassava farms of Perak in West
Malaysia are located on poor soils which would grow little else except cassava (many are located on
tailings or spoil from tin mines). This accounts for their specialization. The second factor, their lack
of management independence, is due to the high bulk/low value of their product which must be
disposed of to chipping factories in the immediate vicinity. There is little practical possibility of
seeking higher prices by transporting the raw cassava further afield. Similar situations face the
citronella grass and cinnamon leaf farmers of the Galle-Matara district in southern Sri Lanka. Here
again the high bulk/low value of these farm outputs deny the growers any real choice in disposing of
their crops to other than the local oil mills under price/quality conditions set by the mills.
Source of farm resources: Farms of Type 4 (and Type 3) are usually not self-sufficient in resource-
generation; e.g., the continuous cropping of specialized paddy farms (Subtype B in Figure 2.2) is
usually possible only because it is based on some purchased package of 'high technology' inputs -
HYV seed, artificial fertilizer and agricides. Dependence on commercial inputs is even greater on
farms of Subtype A, e.g., purchased feed and veterinary supplies for specialist poultry and pig farms.
However, Subtype C farms might exist with only minimal purchased inputs; e.g., the upland near-
subsistence maize farms of Nepal, Bhutan and Sarawak are based on use of retained 'local' seed, no
artificial fertilizer and no agricides.
System boundaries: The boundaries of the specialized farms also vary with subtype. Those of the
commercially-oriented farms of Subtype A will interface more to the outside world (i.e., to suppliers
of inputs and product markets) than to other farms; however, the boundaries of farm-system Subtypes
B and C will tend to be stronger around groups of contiguous farms or all the farms of a hamlet or
village in much the same way as the communal boundaries of Type 1 and 2 farming systems set these
apart from the outside world.
Type 5: Large commercial family farms
Type 5 farms are similar in most respects to estates except that usually the primary beneficiaries are
members of an (often extended) family rather than absentee owners or shareholders. They fall into
two subtypes. The first consists of mono-crop farms which are at the fringe of the estate sector proper
and which are usually dependent on this estate sector for research, availability of new crop varieties
and often for processing and marketing facilities. The 10- to 20-hectare coconut farms of Sri Lanka
which exist side-by-side with the large (now nationalized) coconut estates are examples.
The second subtype consists of either mono-product or mixed farms which are not part of any estate
sector but are organized along commercial lines, e.g., using hired labour, being dependent on
purchased rather than farm-produced inputs and, except in the case of tree-crop farms, adjusting the
activity or activity mix according to commercial opportunity. The larger cinnamon farms of Galle-
Matara and the mixed coconut-dairy farms of Sri Lanka are examples of this subtype. So also are the
large 30- to 50-hectare mixed grain-livestock farms of Sind and Punjab.
The operating objective of Type 5 farms is profit or utility maximization through market sales. As a
group and in pursuit of that objective, they are the most dynamic of the six farm types discussed here.
Type 6: Commercial estates
Commercial estates are generally mono-crop in nature. They are largely a colonial legacy, first
established to provide cheap raw materials (and later some food and beverage products) to the
industries of Europe and North America. This role continues except that they now also serve national
industrialization. The chief characteristics of this farm type are as follows:
Crops: The main crops on which Type 6 farms were initially based are rubber, sugar, cinchona,
cacao, tea, coffee, cinnamon, cloves, nutmeg, coconut and the coarse fibres. Some of the old
traditional crops have become uneconomic (sisal and to some extent cinchona); some have become
primarily smallholder crops (the spices and coffee); and new estate crops (such as flowers, oil palm
and citronella) or improved varieties of old crops have emerged. Recent years have also seen the
emergence (usually close to metropolitan areas) of livestock-based estates, particularly for pork and
broiler production.
On-estate processing: Primary processing is an integral part of the operation of most estates (e.g., tea
manufacture, sheet and crepe rubber production, copra curing). This requires a high level of capital
investment which, to be fully utilized, requires a flow-type of operation rather than a batch-type. This
has two effects. On the one hand it tends to restrict estate production to those crops which yield a
fairly uniform year-round flow of produce (tea, rubber, coconut, cocoa etc.). On the other hand it
gives estates certain advantages, e.g., quality control, relative to smallholders producing these same
products. Some crops which naturally give an intermittent or irregular product flow are also made
amenable to continuous estate-type production by relay-planting or chemical control of growth time-
patterns (e.g., sugar, sisal and pineapple).
Size: Estate size is commonly from 200 to 2 000 hectares but area itself is not an important criterion:
a 40-hectare orchid estate will generate about as much income and employment as will a 200-hectare
tea estate or a 400-hectare coconut estate.
Marketing: Marketing plays a very important role in estate operations. Most estates are jealous of
their product reputation or 'mark' and make deliberate attempts at product differentiation. They also
maintain close contact with buyers and monitor demand trends. Thus the larger cacao estates of
Malaysia might be in daily telex or e-mail contact with buyers in Hamburg and Amsterdam. This
contrasts sharply with the situation on smallholder farms growing the same crops: most smallholders
have little interest in their product once it leaves the farm gate and, not infrequently, have no
knowledge of its use after export.
System beneficiaries and operating objectives: Previously, the primary beneficiaries of estates were
usually absentee shareholders who employed professional expatriate management and often also a
docile expatriate labour force (as was the case in Malaysia, Sri Lanka, Fiji and Mauritius). Thus,
before their nationalization, the estates of Sri Lanka were referred to accurately, if somewhat
emotively, as 'islands of privilege and prosperity in a sea of poverty'. With exceptions, this situation
has changed markedly; consideration of the interests of secondary beneficiaries now receives far
greater attention than formerly and these are more widely defined to include the host government, the
estate labour force and their dependants, local communities and councils. Profit remains the main
operating objective but this is increasingly tempered by the condition that worker retirement schemes,
schools, clinics, roads, village welfare centres etc. be provided at estate cost. In short, increasing
proportions of operating profit are being diverted from the primary to the secondary beneficiaries of
this system type.
Management: A mono-product estate system is at once more simple and more complex than the
systems found on mixed family farms. Since only one product is usually involved, only one
production activity exists, and there is no need to allocate resources among five, six, seven or more
competing production enterprises, as on a typical mixed family farm. Also, very little if any of an
estate's resources have to be generated within the system's boundaries. (On a typical tea estate usually
only fuelwood and hydro power might be produced as inputs to tea production, and even the use of
these is declining.) Thus, again in contrast with family farms, there are no resource-generating
activities to divert attention from the main production task. However, although only one production
enterprise subsystem rather than a multiplicity exists, it is carried on at a sophisticated level. Volume
production usually means that per unit profit margins are thin; the wrong decision at some critical
time, especially with long-term tree crops, can have serious and long-lasting consequences.
Specialization means that the advantages of crop diversification are not available; there is no
possibility of making up on the swings what might be lost on the roundabouts. If a Bhutanese farm
with one pig suffers an outbreak of swine fever the farmer would probably shrug his shoulders and go
off to the paddy field or do something else. If it happened on a 500-sow estate it could spell disaster.
Moreover, insofar as most estates are based on one or other of the tree crops, the effects of some sub-
optimal decisions (e.g., regarding variety/strain of crop to be planted, spacing, initial fertilizer etc.)
might well have long-lasting if not permanent adverse effects, possibly over the 30-year life of a
rubber stand, the 65-year life of a coconut stand, or even longer in the case of tea.
While not requiring the allocation of resources among enterprises, planning and management of a
mono-product estate system requires the explicit recognition, organization and optimization of a large
number of agro-technical processes (as discussed in Chapter 5). An example is given in Figure 2.3
which shows the sequential steps in establishing and operating a tea crop. For each step, several
alternatives are possible. If this were only one of several crops to be grown (e.g., as on a mixed
smallholding), most of the questions listed in Figure 2.3 would not be asked; each operation would
proceed on the basis of village tradition, farmer experience or local lore. But on an estate they have
to be explicitly asked and answered if the optimal level of long-term sustainable production and
profit is to be achieved.
In the management planning of estates, three kinds of management analysis can be particularly
important: (i) evaluation of some single production enterprise (crop or livestock) over a long time
period (as presented in Chapter 10); (ii) the optimization of processes because even small marginal
reductions in inputs/costs will become important when spread over many hundreds of hectares or
thousands of tonnes of produce (Chapters 5 and 8); and (iii) simulation of estate operations as a
whole system under uncertainty (Chapter 11).
Having defined the six main farm types and outlined their chief structural characteristics, it is now
possible to turn in following chapters to a consideration of the field of farm management analysis as
this would be applied to these farm types, especially the small farms, i.e., Types 1 to 4. First,
however, two contrasting examples of small-farm systems are presented in the following section.
FIGURE 2.3 - Alternative Processes in producing Tea on an Estate
Sequential operations Alternative processes for each operation
(1) Clear land Method: labour? elephants? tractor?
Level: what season? depth? tilth?
(2) Establish cover crop Method: which crop?
Level: for how long? 15,18 or 24 months?
(3) Plant tea Method: what kind?

seedling or vegetatively propagated?
Level: what population density? 3-4-5 000 plants/acre?

(4) Weed Method: how? hand? chemical? men? women?
Level: every 4, 6, 8... weeks?
(5) Fertilize Method: what type(s)?
Level: what levels of application? frequency? timing
(6) Pick Method: how? machine? hand? male or female?
Level: frequency? every 25, 30, 35... days?
(7) Cure/pack Method: how? for which of many markets?
Level: mix grades? in what proportions?
(8) Etc.

A useful way of introducing the discussion of following chapters is to look briefly via examples at the
structure of two of the small-farm types, the partly commercialized farms (Type 2) and the near-
subsistence farms (Type 1).
Model of a Type 2 farm
Based on McConnell (1972), Figure 2.4 presents a model of the annual operation of a 'representative'
Pathan farm in the Peshawar district of North West Frontier Province, Pakistan (the data are actually
means of a group of 11 similar farms). The central core of the farm system consists of two livestock,
seven crop and two on-farm processing activities: dairy and draught cattle, berseem (clover), a mixed
orchard, sugar beet, sugarcane, maize, millet and wheat. The processing activities consist of
converting some of the milk to ghee for sale or consumption (in this particular year it was all
consumed), and crushing/boiling some of the sugarcane to make gur (unrefined 'country' sugar), some
of which is consumed, some sold. These 11 activities are represented by bar columns in the middle
part of Figure 2.4. Input and output values are in rupee (Rs) terms. The levels of the various
activities, livestock population and other relevant structural data are summarized below.
Crops Level
berseem (cattle feed) 0.6 acres
orchard (use + sale) 1.1 acres
sugar beet (sale) 0.7 acres
sugarcane (sale + gur) 1.3 acres
maize (use) 1.3 acres
millet (use) 0.2 acres
wheat (use) 2.0 acres
sugarcane:
produced 542.0 maunds2 plus 15.5 maunds of gur
sold as cane 542.0 maunds
Livestock
bullocks 2.0 head
milk animals: 1.7 head
producing 1.1 head
dry 0.6 head
young stock (dairy + draught) 1.2 head
ox traction available (pair) 365 days
actually used 74 days
milk produced: 22.0 maunds
used/sold as milk 18.4 maunds
converted to ghee 3.6 maunds

2 Maund (abbreviated to mds in Figure 2.4) is a volumetric measure corresponding to about 90 pounds (40 kg) of sugarcane; ghee is
measured by the seer (about 14 pounds or 6 kg).
Figure 2.4 consists of two sections: the household component, circumscribed by a boundary line in
the top right section of the diagram, and the farm component of the farm-household system which
constitutes the remainder of the diagram.
Referring to the household component of Figure 2.4, the farm family consists of seven members of all
ages. Together these members are capable of supplying 1 053 labour days annually. In fact there is
only enough work on the farm to occupy 443 labour days, and only 21 days of off-farm work can be
found, thus 589 days are either occupied in farm maintenance or development work not connected
with any particular crop, or are occupied in social/religious activities, or are idle. (Here they are
shown as 'idle'.)
The second general input by the family into the farm component is cash for meeting the direct costs of
each of the production activities. This cash is obtained by the household as Rs 2 848 from sale of
farm produce plus Rs 65 from 21 days of off-farm work. (Of course the family component also
supplies other vital inputs to the system - management, direction and purpose - but these cannot be
measured.) These family-provided inputs into each of the activities, cash and labour days, are shown
in Figure 2.4 in the two top rows of the farm component.
Consider now the farm component as depicted in Figure 2.4.
Activities: Referring to the activity columns in the body of Figure 2.4, each of these is an abbreviated
activity budget (Chapter 4) showing first the inputs to and then the outputs from the activity. For
example, the inputs to 0.7 acres of sugar beet are cash Rs 79, labour 48 days, bullocks 8.5 days,
manure 66.5 maunds; and the outputs are 32 maunds of leaves (by-product) and 175 maunds of beets.
Similarly, the inputs to sugarcane are cash Rs 250, labour 82 days, bullocks 14.4 days, manure 80.6
maunds; and the outputs are 185 maunds of cane tops (for bullock feed) and 542 maunds of sale cane,
plus a small but unknown amount of cane diverted to home processing for gur which as a separate
activity uses Rs 77 cash, 25 labour days and 2.2 bullock days, and yields 15.5 maunds of such
'country' sugar of which 6.8 maunds is consumed and 8.7 maunds is sold.
FIGURE 2.4 - Structural Model of a Pathan Farm exemplifying a Type 2 Farm
The two livestock activities - dairy and draught oxen - are shown at the left side of the model. Inputs
are cash costs, labour and livestock feed. This latter is somewhat complex.
Livestock feed: Feed is shown in two units of measurement: amount (i.e., maunds of around 40 kg) of
actual material, and the equivalent in maunds of total digestible nutrients (TDN). This conversion is
desirable in order to standardize each of the several different feedstuffs produced on the farm, each
having a different nutritional value, into a common basis of units of TDN. The total amount of feed fed
to the livestock is shown at the top of the livestock columns, 101 maunds of TDN, coming from the
'total feed pool'. All feed entering this common pool comes from one of three sources: (a) grass (cut
from the orchard) equivalent to 5.9 maunds of TDN; (b) 613 maunds of green feed, equivalent to 63.8
maunds of TDN, which comes from some of the crop activities (berseem 246 maunds, beet leaves 32
maunds, cane tops 185 maunds, maize green chop 150 maunds) and (c) 79.3 maunds of dry feed,
equivalent to 31.3 maunds of TDN, from some of the crops (maize stover 14.8 maunds, millet 33.2
maunds, bhoosa/wheat straw 31.3 maunds).
The green feed is shown as cycling to the left and being accumulated in a 'green feed pool' totalling
613 maunds of material, equivalent to 63.8 maunds of TDN; the dry feed items are accumulated to the
right into a 'dry feed pool' totalling 79.3 maunds of material or 31.3 maunds of TDN. Then, as the
arrows indicate, both these green and dry feed pools are accumulated above the livestock activities
into a 'total feed pool' which, when supplemented by cut grass, totals 101 maunds of TDN flowing to
all the dairy and draught animals (which include young and dry stock as previously listed). These
feed flows refer only to feed produced on the farm. In addition, stock are grazed on village common
lands when grass is available there (the amounts of such grazing could not be recorded).
In summary, the following widely diversified feedstuffs are obtained on this type of farm: grass from
the orchard, some rough common-lands' and roadside grazing, berseem clover, beet tops, sugarcane
tops, green maize chop, dry maize stalks, millet and wheat bhoosa (straw). Only two of these items
are specially grown for the cattle, namely berseem and millet.
Livestock outputs: The cattle activities generate three outputs: bullock power, manure for the crops,
and milk. The number of bullock days flowing to each of the crops are shown, being 2.0 for the
clover, 5.5 for the orchard, 8.5 for the sugar beet etc. Crushing of sugarcane for gur also uses 2.2 days
of bullock power. Bullock power not used is shown as 291 'idle' bullock days. Below the livestock
activities, a total of 242 maunds of manure are accumulated from all the livestock. This also is shown
flowing to the crops: 25.8 maunds to berseem. 0.4 for the orchard, 66.5 for the sugar beet etc.
Final activity outputs: The lower section of the model shows the final products flowing from each
activity and the amount of each product consumed by the household or sold, each in quantity and value
terms.
Household income: Farm income consists first of the value of produce consumed. This is
accumulated to the right in Figure 2.4 and enters the household component of the system as a total
value of Rs 1 638. Second, cash from farm sales is similarly accumulated and has a total value of Rs
2 848. Total income, real plus imputed, is Rs 4 551 which includes the small income from non-farm
work. Thus, for this year, income in kind from home-consumed production constitutes 1 638/4 551 or
36 per cent of total gross family income. On this basis, it might be said that the farm is about one third
subsistence oriented and two-thirds commercially oriented.
As shown, total cash inputs into all the activities amount to Rs 675. Thus, in this particular year, the
household would have a cash 'surplus' of Rs 2 848 + 65 - 675 or Rs 2 238. This amount would be
available to meet any cash costs in farm maintenance (which were not considered as a cost in the
model), and to meet cash living expenses for purchased clothing, food, medical expenses etc. Any
final surplus after meeting these latter expenses would be available as savings.
In summary, this Pathan fanning system from the North West Frontier Province of Pakistan is a highly
diversified one. It has 11 major production and processing activities and it produces 15 separate
products and by-products (excluding bullock power, young livestock and manure for the fields). Cash
inputs are low, mainly for fertilizer for some of the crops and a minimal amount of agricides; most
seed is retained. Although a significant degree of self-sufficiency is present, it is not a true
subsistence farm. This latter type is examined in the following section.
Model of a Type 1 farm
Most farms of Type 1 (i.e., small subsistence-oriented family farms) in Asia now have at least some
element of commercialisation and generate at least some small amount of cash for the purchase of
essential items. At the top end of the structural scale this type merges into the small mixed Type 2
farms; at the bottom end it includes the locally-shifting cultivators of Sarawak-Kalimantan (and these
merge into the hunter-gatherers and forest dwellers of New Guinea, Kalimantan, Sarawak and
Sumatra).
Subsistence or near-subsistence is a condition more often imposed than voluntary: one sub-
classification could be on the basis of external causal factors; e.g., the near-subsistence farms of
India, the hills of Nepal and parts of Java exist because of shortage of land; those of Bhutan (where
land is seldom limiting) because of isolation and the lack of roads and markets; and those of the new
settlement areas of Sri Lanka because of the lack of family labour and oxpower to till more than a
minimal subsistence area.
When examined from the viewpoint of their range of activity, variation of this farm type ranges all the
way from being highly mixed to almost mono-crop. In the first of these conditions, this type merges
into Type 2 farms. In the second extreme condition these farms are structured around production of a
single bulk staple - usually maize, dry paddy, cassava, palm sago or coconut. Secondary foodstuffs
and non-food subsistence items are obtained by supplementary activities: fishing; hunting or
collecting in nearby forest areas for food or items for sale (birds, monkeys, butterflies, orchids,
beeswax, rattan etc.); or by sending a family member off to work somewhere outside the subsistence
environment.
Although they can generally be described as resource-poor, poverty is not necessarily a characteristic
of Type 1 farm families. At one end of the scale the economic condition might be poverty verging on
destitution, even starvation. At the other extreme it might be prosperity when that condition is judged
by an availability of resources in excess of those needed to maintain a reasonable physical existence.
Again the best way of illustrating systems of this subsistence type is with the aid of a model. Such a
model is shown in Figure 2.5 which describes the structure and operation of a near-subsistence farm
located at about 1 700 m (5 500 feet) above the town of Wangdiprdan in Central Bhutan. The chief
features of this farm are its highly diversified activities and its very high level of self-sufficiency. In
this example the farm is a prosperous one (as that term was defined above) and its subsistence nature
is based on isolation, poor roads, lack of markets and therefore of incentive - indeed opportunity - to
enter the commercial world. The model refers to an operating period of one year.3
3 Data from a 1985 farm survey by Nim Dorji and the senior author. The monetary unit is Ngultrum (Nu) = US 8.7 cents in 1985.
The farm's land resources consist of 1.75 langdo4 (0.25 hectares) of dry paddy fields, 1.25 langdo
(0.18 hectares) of dry fields, and 4.00 langdo (0.4 hectares) of wetland rented from a local
monastery. However, the effective size of this farm is only 5.75 langdo: the 1.25 langdo of dryland is
too far away to be easily worked and, since more accessible land can be rented from the local
monastery, the dryland portion is not used. (If it were cultivated there would be no market for the
extra produce, which would also be surplus to family requirements.)
4 A langdo is the area a pair of oxen can plough in a day: a dry langdo is about one seventh of a hectare and a wet paddy-field
langdo is about one tenth of a hectare.
The farm family consists of five adults, all able to work full-time if necessary, i.e., a population
density and potential workforce of six persons per hectare. It is assumed that this family could
provide a potential supply of 1 200 labour-days annually. Livestock resources are also high relative
to farm size: two oxen, two milkcows, three young cattle, three pigs, three hens. Cattle are grazed off
the farm on common lands for more than half the year.
The farm structural model of Figure 2.5 consists of six parts as indicated by the circled numbers 1 to
6 in the diagram. Part 1 shows human, livestock and land resources, as noted above. Part 3 shows the
farm resource pool. All resources except rented land are owned or are generated by the system itself
(all seed, manure, bran, labour etc.). There are no purchased inputs; even the rented land is paid for
by barter with the monks.
In Part 2 of the structural model, all productive activities (subsystems) are listed in separate activity
columns. There are nine of these (plus one external resource-generating activity): tending cattle, pigs
and poultry; growing paddy, wheat, buckwheat, mustard, vegetables/chillies and fruit. The external
resource-generating activity consists of one family member working part-time off the farm (for local
government, clearing paths and roads after landslides).
The coefficients within each upper column of Part 2 are the resources used by the respective
subsystems: e.g., the paddy crop uses 5.25 langdo of land, 115 labour days, 29 kg of seed, 14 tonnes
of manure and 23 draught ox days. Of the used total area of 5.75 langdo, two langdo are
doublecropped.
Part 4 shows all intermediate products (bran, grain, straw, draught days, manure and seed) which are
produced by the nine farm activities. These are generated in the activity columns, accumulated to the
left in the 'intermediate outputs' section (Part 4), then cycled back to the resources pool of Part 3 and
from there to the activities which use each such resource/input. As shown by the left-side arrows
from Part 4 to Part 3 and thence to Part 2 of Figure 2.5, resources used up by the activities (Part 2) in
any year are replenished by the activities producing intermediate products, but these need not exactly
balance because some resources may be held over in storage: e.g., in this year 21.5 tonnes of manure
are used on the crops but only 15 tonnes were produced; 6.5 tonnes came from storage not recorded
in Part 3 of Figure 2.5.
FIGURE 2.5 Structural Model of a Bhutanese Farm exemplifying a Type 1 Farm
Part 5 of Figure 2.5 lists and aggregates final outputs from the nine activities: e.g., for the paddy
activity the intermediate products from paddy (bran, straw, seed) cycle to the left in Part 4 and then
up to the resource pool. But the final paddy products consist of 934 kg of grain used for family food,
340 kg converted to alcohol (consumed not sold), 410 kg bartered to other families for other types of
food not produced on this farm and 112 kg of paddy paid as rent to the monks for the four langdo of
wetland.
The various food items produced and consumed on the farm are shown in the 'food' line of Part 5: 13
kg of butter and 50 kg cheese from the cattle, a 50 kg pig, 10 kg eggs, 934 kg of paddy, 20 kg of wheat
etc. Alcohol was made from a total of 366 kg of grain (paddy and wheat).
The bottom of Part 5 shows those outputs which were sold. In this particular year the only marketed
items were some citrus fruit (76 kg) and vegetables (70 kg) which were sold for Nu 16 and Nu 50
respectively at the Wangdiprdan weekly 'hat' (street market).
Clearly this farm system is one very close to complete subsistence. There are no purchased inputs
into the crop/livestock activities and all farm capital items such as ploughs, harrows and ox gear are
home-made, as are the woven storage bins for grain storage. The mustard seed oil for home use is
extracted using a kitchen press and used for both cooking and in the lamps before the family's
Buddhist altar.
Parts 1 to 5 of the model refer to the farm components of the farm-household system. Part 6 refers to
the household component: this consists of data relating to the use of family labour, and to family
income (cash plus food) and non-farm family expenditure. Of the 1 200 days of family labour
available, 444 days are used in farm activities and off-farm work, leaving 756 days designated as
'idle'. (In fact some of these would be used in general maintenance around the farm in jobs not
directly related to any of the nine production activities.)
Total cash family income as shown in Part 6 amounts to Nu 786: Nu 720 is from off-farm work; only
Nu 66 is from sale of farm produce. From this cash income, family cash (non-farm) living costs are
deducted, leaving an apparent net deficit this particular operating year of Nu 189. The items
comprising family cash expenditure are shown: Nu 600 for purchased food, Nu 70 for purchased fuel
(kerosene), Nu 80 for household items, Nu 75 for insurance and Nu 150 for clothing. The apparent
negative cash balance of Nu 189 would be made up from savings or by obtaining credit for purchases.
Due to off-farm work and cash expenditures for the items shown, the farm-household system at this
point has ceased to be a purely subsistence one. However, if the family's economic conditions
changed for the worse, the expenditure pattern could be easily adjusted to reduce or eliminate some
of the cash expenditure items (especially food which consists mainly of 'luxury' items), and the
remaining needs for cash could be met by sale of a pig or a little mustard oil.
It is now possible to turn in following chapters to an examination of the individual structural elements
of this and other types of farm systems.
2.3 REFERENCES
Ashby, J.A. and L. Sperling (1995). 'Institutionalizing Participatory, Client-driven Research and
Technology Development in Agriculture', Development and Change 26(4): 753-770.
Axinn, N.H. and G.H. Axinn (1983). Small Farms in Nepal: A Farming Systems Approach to
Description, Rural Life Associates, Kathmandu.
Chambers, R. (1983). Rural Development: Putting the Last First, Longman, London.
Chambers, R. and B.P. Ghildyal (1985). 'Agricultural Research for Resource-poor Farmers: The
Farmer-first-and-last Model', Agricultural Administration 20(1): 1-30.
Dillon, J.L. (1980). The Definition of Farm Management', Journal of Agricultural Economics 31(2):
257-258.
Dillon, J.L. and J.B. Hardaker (1993). Farm Management Research for Small Farmer
Development, FAO Farm Systems Management Series No. 6, Food and Agriculture Organization of
the United Nations, Rome.
Dixon, J.M., M. Hall, J.B. Hardaker and V.S. Vyas (1994). Farm and Community Information
Use for Agricultural Programmes and Policies, FAO Farm Systems Management Series No. 8, Food
and Agriculture Organization of the United Nations, Rome.
Fresco, L.O. and E. Wesphal (1988). 'A Hierarchical Classification of Farm Systems',
Experimental Agriculture 24: 399-419.
McConnell, D.J. (1972). The Structure of Small Farms in Peshawar, NWFP, Pakistan, UNDP-FAO
Consultant's Report TA3070, Food and Agriculture Organization of the United Nations, Rome.
McConnell, D.J. (1992). The Forest-garden Farms of Kandy, Sri Lanka, FAO Farm Systems
Management Series No. 3, Food and Agriculture Organization of the United Nations, Rome.
Makeham, J.P. and L.R. Malcolm (1986). The Economics of Tropical Farm Management,
Cambridge University Press.
Matlon, P., R. Cantrell, D. King and M. Benoit-Cattin (eds) (1984). Coming Full Circle:
Farmers' Participation in the Development of Technology, IDRC, Ottawa.
Mikkelsen, B. (1995). Methods for Development Work and Research: A Guide for Practitioners,
Sage Publications, London.
Norman, D.W., F.D. Worman, J.D. Siebert and E. Modiakgotia (1995). The Farming Systems
Approach to Development and Appropriate Technology Generation, FAO Farm Systems
Rhoades, R.E. and R.M. Booth (1982). 'Farmer-back-to-farmer: A Model for Generating
Acceptable Agricultural Technology', Agricultural Administration 11(2): 127-137.
Tripp, R. (ed.) (1991). Planned Change in Farming Systems, Wiley, Chichester.
Upton, M. (1973). Farm Management in Africa, Oxford University Press, London.
Upton, M. and J.M. Dixon (eds) (1994). Methods of Micro-level Analysis for Agricultural
Programmes and Policies, FAO Farm Systems Management Series No. 9, Food and Agriculture
Werner, J. (1993). Participatory Development of Agricultural Innovations: Procedures and
Methods of On-farm Research, Schriftenreihe der GTZ No. 234, GTZ, Eschborn and SDC, Bern.
3. ELEMENTS OF FARM-HOUSEHOLD
SYSTEMS: BOUNDARIES, HOUSEHOLD
AND RESOURCES
3.2 FARM HOUSEHOLD
3.3 FARM RESOURCES
3.4 REFERENCES
'Before I built a wall I'd ask to know what I was walling in or walling out ...'
Robert Frost. 1874-1963
In this and in Chapters 4 and 5 the several structural elements of a farm-household system noted in
Section 1.4 are more closely examined.
In analysing any farm-household system or any other agricultural system, an obvious first step is to
define the scope of such a system, i.e., its boundary as relevant to the purpose of analysis (FAO
1990). Sometimes this will present no problem, particularly if, as here, the focus of analysis is on
farm production and its management. In this case, unless they relate significantly to production
management, interfaces of a purely social, religious or political nature between the farm-household
and its environment can be ignored (Dillon 1992). However, as the following examples illustrate,
even if the focus of analysis is production and its management, the specification of the farm-system
boundary (and likewise of the farm-household system boundary) may be quite complicated. Boundary
definition may also vary with the purpose of analysis. Thus the relevant boundary for annual
enterprise planning may simply correspond to the boundary of the physical farm entity. In contrast, for
long-term development planning, the boundary may need to include off-farm income-generating
activities and the interface with suppliers of long-term credit.
Modern Western and Asian farms in a highly commercial environment (e.g., West Texas cotton farms
and Malaysian rubber smallholdings and estates) have one or a limited number of farm production
enterprises and clear sets of trading relationships between the farm and its input suppliers on the one
hand and its output markets on the other. Relationships between a given farm and other farms will
often be minimal, even non-existent. From a production management perspective, the boundary of this
type of farm system will encompass the farm in its physical extent and its input suppliers and market
outlets, but will typically exclude other farms.
In a second situation, as exemplified by many isolated near-subsistence farm-household systems of
the Himalayan valleys, the boundary of the farm system also corresponds closely with the farm's
physical boundary but in this case the system excludes the outside world.
More common is a third situation. It comprises the great bulk of small traditional farms of Types 1, 2
and 3 (i.e., subsistence, semi-subsistence and specialized independent, respectively) which are
moderately or heavily dependent on each other for supply of inputs - exchange or hired labour,
oxpower, village transport etc. - and often also for the disposal of produce, e.g., by barter or trading
among friends and neighbouring families. In this situation the real boundary of a farm-household
system, from an operational management viewpoint, might encompass all the farms of a hamlet or
even of several mutually dependent villages.
This third situation in some of its variations can be quite complex. For example, the farms of the
villages around Karanganyar in the Solo Valley of Java commonly consist of some area of household
garden devoted to fruit trees and bamboo plus a larger area of fields (Prabowo and McConnell
1993). They are nominally 'owned' (in the Javanese sense, or held in trust in perpetuity) by the
operating families who in fact do exercise full control over the garden or 'pekarangan' parts of their
farms. In three years out of four the field area of a farm will usually be under paddy, grown either by
the farmer himself or herself or more typically as a quasi-cooperative/mutual-help undertaking
('gotong royong') by the farmer and his or her neighbours. Thus in this first phase the boundary around
each farm system encompasses its household garden and some area of paddy shared with other farm
households. But in the fourth year the farm paddy fields, together with one fourth of all other village
fields (which lie within the command area of a publicly owned sugar mill) are pooled, put under
sugarcane, and the farmer becomes both a labourer (on his or her own land) for the mill and a
landlord (receiving rent from the mill). In short, the boundary of this farm system is fluid over time: it
always encompasses the pekarangan (household garden) lands; it sometimes encompasses the rice
field and other farm households which participate in the rice phase; and it sometimes exists only as
one part of the boundary of a much larger sugar-estate system. The situation is depicted
diagrammatically in the righthand side of Figure 3.1. The lefthand side of Figure 3.1 refers to an
isolated subsistence farm system. The central part of Figure 3.1 depicts a commercial system in
which the alternatives are to define the system as consisting only of the farm (implying boundary B')
or as also including input suppliers and output markets (implying boundary B). The contrast in
complexity between these three examples of farm-system boundaries is obvious.
At farm-household level, one obvious essential step in planning farm development programs and
investment projects is to accurately identify the relevant boundary of the subject system; yet it is a
step sometimes not taken. In the 1970s a project in North East Africa was aimed at village economic
development through introduction of modem rice-growing technology to Dinka tribespeople in the
hope of supplementing and if possible replacing their traditional millet (as well as supplying rice to
the urban population). Project planning overlooked the fact that, while millet was indeed important,
the real boundary of the traditional farming system encompassed also keeping cattle and catching
mudfish. Thus, while project planning management was directed at zealously maintaining field polder
embankments to grow rice, Dinka management was directed even more zealously at breaking them
down to get at the fish. (Planning that was less technically oriented would have aimed for the
integration of all three components in a locally acceptable system.) This and similar mistakes in other
farm projects also provide a caution that structuring or restructuring of a system must always start
with an understanding of the farm-household system as discussed in Section 3.2.
At the levels of subordinate subsystems, the boundaries of processes, activities and enterprises,
although set approximately by technical factors, are determined also by the purpose of the analysis or
by operational convenience (as discussed in Chapter 4).
FIGURE 3.1 - Boundaries of Three Contrasting Farm Systems
3.2 FARM HOUSEHOLD

'I grant indeed that fields and flocks have charms

for him that grazes or for him that farms;
But when amid such pleasing scenes I trace
the poor laborious natives of the place ...'
George Crabbe, 1754-1832
The household component of a farm-household system is a somewhat flexible concept. It can consist
only of the farm's nucleus family but more often includes extended-family members. It also commonly
includes some number of more or less permanent domestic and farm workers and miscellaneous
dependants. Most farm development projects are structured on the assumption that households are
headed by males, but this is often not so. In a typical Javanese village anywhere from 10 to 25 percent
of households might consist only of women and children. Where poverty exists, here and elsewhere, it
will generally be concentrated on such households.
The two roles of the household as resource manager and as system beneficiary were noted in Section
1.3.1. The household's role as resource manager is to provide purpose, direction, objectives and
management to the whole-farm system and its subsystems (Shaner, Philipp and Schmehl 1982). On
farms of Type 5 (large commercial family farms) and especially Type 6 (estates) the place of the
household might be taken by professional management. However, on small farms and within limits set
by the physical environment and available resources, the planning objectives are set by the
household's broad social-value structure and local tradition (though the latter is increasingly being
corroded by commercialisation and the influence of modem communications technology). The farm
economist is usually not well equipped to question these goals or the farm-planning objectives which
result from them. His or her role is to attempt to optimize the farm component of the system in terms
of whatever mix of socio-cultural, religious, traditional and material goals is relevant to the
household. He or she might also evaluate and convey the material consequences of alternative choices
and management strategies within the farm component.
Decision making within the household
It will often be necessary to enquire more deeply into who does what and who decides what within
the household if sound farm planning is to be possible (in Field A) or sound farm development
programs are to formulated (Fields C and D). Real power and responsibility for the farm component
might not always lie with the apparent or nominal household head. Thus in the Wadi Hadramant of
Yemen, as throughout most of Africa, although women might play an otherwise subservient role - or
seem to - they are in fact the effective farmers, doing everything except the heaviest chores from
planning the production system and executing most of it to preparing and allocating its output. Yet
farm 'development' projects here and elsewhere - with few more than token exceptions and invariably
conceived, planned, evaluated and managed by men - do not acknowledge this fact and remain
directed at the mirage of some unspecified but assumed patriarchal 'farmer'. Without at least an
insight into local culture, the conceptual models of how a system works which an analyst may have
can be unproductive if not dangerous things.
Gender analysis
As noted above, sound farm planning often necessitates an understanding of how resources,
responsibilities, tasks and benefits are distributed between the men, women and children of the farm
household. This is the subject of gender analysis as discussed and illustrated by, e.g., Feldstein and
Jiggins (1994), Feldstein, Flora and Poats (1990), Gondowarsito, van Stanten and Bottema (1995),
Quisumbing et al. (1995) and van Herpen and Ashby (1991). The need for gender analysis,
particularly in relation to the role of women in the farm-household system, is well illustrated by the
following response by a small farmer when interviewed in a farm survey (Gabriel 1995, p. 68):
Q. 'Does your wife work?'
A. 'No, she stays at home.'

Q. 'I see. How does she spend her day?'
A. 'Well, she gets up at four in the morning, makes the fire and cooks breakfast. Then she goes to the
river and washes clothes. After that she goes to town to get corn ground and buy what we need. Then
she cooks the midday meal.'
Q. 'You come home at midday?'
A. 'No, no. She brings the meal to me in the fields - about three km from home.'
Q. 'And after that?'
A. 'Well, she takes care of the hens and pigs, and of course she looks after the children all day. Then
she prepares supper so it is ready when I come home.'
Q. 'Does she go to bed after supper?'
A. 'No, I do. She has things to do around the house until about nine o'clock.'
Q. 'But you say your wife does not work?'
A. 'Of course she doesn't work. I told you, she stays at home.'
Tables 3.1 and 3.2 provide an illustration of some aspects of gender analysis (labour and task
allocation - but not the distribution of benefits) relative to a sample of small-farm households in
Northern Mindanao, Philippines.
TABLE 3.1 - Gender Analysis of Labour Use in Cassava Production in Northern Mindanao,
Philippinesa
a Based on a sample survey of 75 small farms (Rola 1995)
TABLE 3.2 - Gender Analysis of Participation of Farm-household Members in Household and
Other Activities in Northern Mindanao, Philippinesa
a Based on a sample survey of 75 small farms (Rola 1995)

On small farms the primary beneficiaries are usually the members of the household itself. However,
external beneficiaries are also often important. In Bhutan it is common for farm family members long
resident far from the farm and not dependent on its output to still retain important rights. They
exercise these by returning at harvest time and taking their benefits in the form of pork and grain. In
the villages of Central Java up to 30 percent of households might be landless and subsist as farm
labourers or - especially important in the case of households consisting of only poor women and
children - by harvesting their neighbours' paddy in exchange for retaining one seventh or one tenth of
the crop (the 'bawon' system). This latter can be the only significant source of income for otherwise
practically destitute people. They are perhaps proportionately more dependent on the farms of their
more fortunate neighbours than are the farmers themselves, although they are from a formal viewpoint
external to these systems. This is one common reason why some forms of Western-style farm
'development' do not occur. Javanese villagers have a strong sense of mutual economic and social
responsibility. 'Advanced' technology which would prepare land more quickly (tractors) or give
higher grain recovery rates (mechanical harvesters, grain dryers) or other technology whose first
effect would be to increase the comparative wealth of the few would be frowned on because of its
adverse effect upon the many. Thus external beneficiaries can impose constraints on the household, in
its first role as system manager and internal resource allocator, by exerting a collective moral
influence on how crops are to be produced and even which crops are to be grown.
Other obvious groups of external beneficiaries consist of landlords, village traders, local
governments as taxing authorities etc. and, at further remove, national governments deriving their
foreign exchange from the export of farm produce. These can also influence how a farm-household
functions in its first role as system manager, e.g., through tenancy agreements, provision of credit,
price policy etc.
3.3 FARM RESOURCES

The supply of resources in a whole-farm system can be examined from either an accounting or, more
fruitfully, an operational perspective.
From an accounting viewpoint with its ex post or backward-looking emphasis, farm resources fall
into two broad categories (Makeham and Malcolm 1986):
· Fixed resources provide services over a number of years or at least over a period longer than the production cycle of short-term
(seasonal, annual) crop or livestock enterprises. Common examples are land, machinery, an irrigation system. These services may
be used either by individual enterprises or to maintain the farm as a whole. In the very long run, of course, few resources are truly
fixed in supply. Even land and climate in their productive dimension can be created (as in the controlled-environment
greenhouses of Saudi Arabia, UAE and Qatar).
· Short-term or variable/operational resources are those that are usually entirely used up in the
annual production cycle, e.g., a supply of seed or fertilizer.
The essential difference between fixed and short-term resources is that the former provide a stream or
flow of services over time while the latter consist primarily of quickly exhaustible material things or
time-bound institutional sanctions (as noted in Section 3.3.3).
From an accounting viewpoint, both fixed and short-term resources have two other relevant
dimensions. In their economic dimension they become respectively fixed/long-term capital and
operating/short-term capital; and in their financial dimension these generate, respectively, fixed costs
and operating/variable/direct costs, viz.:
The distinction between these two cost categories of fixed and variable is discussed further in
Sections 4.3.8 and 9.
Discussion so far has concerned resources when used at whole-farm level, i.e., at Order Level 10. At
lower Order Levels of farm systems (processes, activities and enterprises), the above classification
of resources relative to capital and costs is parallel to that for the whole-farm: e.g., the resources
assigned to a paddy crop can be broken down and become fixed and operating paddy-crop capital,
generating fixed and operating/variable/direct paddy costs.
Variable/direct costs are discussed in Section 4.3 in the context of enterprises and activities, and
fixed costs in Section 5.3 relative to the whole-farm service matrix.
From an operational viewpoint the picture of farm resources is somewhat different. Here emphasis is
on the ex ante potential or planned use of resources rather than the results of their past use. In this
forward-looking context it is more fruitful to regard farm resources from a systems viewpoint.
Distinction between fixed and short-term resources is now less relevant. As shown in Figure 3.2,
more relevant are the flows of both short-term or operational resources and of services from fixed
resources to the farm's subsystems.
FIGURE 3.2 - Direction of Resource Flows within a Whole-farm System
Central to these resource flows is the farm's resource pool (a system of Order Level 8, discussed
from an operational perspective in Section 3.3.3 below). Short-term or operational resources flow
from the pool primarily to lower Order Level production subsystems (processes, activities and
enterprises) to generate the farm's intermediate and final outputs. Some also flow directly to the
whole-farm service matrix to maintain the structure (repair of fences, buildings etc.) or permit the
functioning of this subsystem (e.g., payment of land, water and road taxes).
The services of long-term or fixed resources flow both to maintain the service matrix and to the
resource pool from where they are assigned to the various production subsystems.
Any initial stock of both fixed and short-term resources must sooner or later be replenished from
output of the productive enterprises/activities, either as materials purchased with income from the
activities or as resources/intermediate products generated by these activities, e.g., oxen produced as a
by-product of the dairy activity. These resource return flows are indicated in Figure 3.2 as flowing
both to the resource pool and to maintain the stock of fixed resources or capital in the whole-farm
service matrix.
From a planning and operational viewpoint, farm resources fall into five categories. Discussion from
an operational viewpoint is focused on how specific resources might constrain or limit farm
production. The five resource categories are:
(1) Material long-term: This category consists of material things which yield their services over relatively long time periods. They
were referred to above as fixed resources or fixed capital. Land is typically the most important such item and will usually provide
its services indefinitely (but see system sustainability, Section 6.2.7.) On the other hand, land is relatively less important on many
capital- and labour-intensive specialist Type 3 farms such as those producing orchids, poultry and pigs. Other examples of
resources in this category are irrigation systems and farm sheds generating their services over 20 to 30 years or an ox pair
providing draught power over five or six years.
(2) Material short-term: This category, exemplified by such items as seed and other seasonal inputs,
was also discussed above. In a commercial environment where these items are purchased, the
production constraint is generally set by the amount of money available to buy them, not by the supply
of these items in themselves.
(3) Financial: This category consists of cash, debts receivable, and access to credit from formal
(banks, cooperatives) and informal (shops, traders, relatives) sources.
(4) Institutional: This category consists essentially of rights of access to materials, markets and
services. In its financial dimension, this category takes the form of land and road taxes, water-use
license fees, payments for production-quota rights (as sometimes prevail for sugarcane, milk, tobacco
etc.). They are termed 'institutional' because they consist of relationships between the farm family on
the one hand and institutions/agencies/persons on the other. Note that where they are transferable and
have financial value, these rights are assets as well as production resources.
(5) Labour: This consists of family labour available for general farm work or which might be
available only for specific tasks. For example, specific-purpose labour might consist of the very old
and young family members who can do only light work such as tending livestock; or a family member
who prefers and is especially skilled in tapping toddy palms etc. (Management ability is an important
attribute of family labour, sufficiently so as to sometimes warrant attempts at separate evaluation of
its productivity, but it is usually not possible to measure management as an ex ante input, only in
terms of what it actually achieves as discussed in Sections 7.2.3 and 5).
Resource inventory for planning
For planning (ex ante) purposes, specification of the farm resource pool simply amounts to making a
list of the availability of those resource items which might limit production in the planning period,
e.g., for the coming year the resource pool may include the following resources:
(1) Material long-term: land 0.25 ha irrigated lowland
0.60 ha eroded upland
oxen 1 pair
ploughs/harrows 3 units
pump 1 unit
buildings 1 house, 1 shed (100 m2)
(2) Material short-term: fertilizer 6 bags
cow feed 3 tonnes
(3) Financial: cash 1 500 Rs
expected from crops 3 500 Rs
(4) Institutional: milk quota 20 litres/day, sold in town
water license 0.25 ha
(5) Labour: work-age adults 490 labour days
children 180 days (cattle only).

Listing of the initial resources will depend partly on the resources actually available but also on the
uses to which they are likely to be put, i.e., the types of enterprises/activities which are likely to be
operated. The procedure, therefore, is somewhat circular and subjective but in practice it poses no
serious problems. The aim is to identify all the main farm resources, particularly those which are
likely to be production-limiting and relatively costly, rather than to compile an exhaustive list of all
resources which are present or which could conceivably act as constraints.
In almost all planning situations, the analyst will be able to form a general idea of the likely
production possibilities and therefore the types or categories of resources needed to exploit such
possibilities. If, as is often the case on many small farms, there exists two or three times the amount of
labour likely to be needed to execute the production plan, one would not devote much attention to this
resource. But, on the other hand, if the production system is likely, e.g., to include strawberries, and
this requires the deft fingers of children, then one would measure this particular category of available
labour with some precision.
Resource quality
For planning purposes it will often be necessary to quantify some resources according to the specific
uses to which they can be put. These uses may have a quality as well as a quantity dimension. The
example of child labour was noted above. If a farmer has 1.5 hectares of land and an asset statement
is being prepared (below), it would be enough to describe this as simply 'Land ... 1.5 ha', but for
planning purposes it would be necessary to list this in the resources pool as, e.g.:
wet-land (necessary for paddy) 0.25 ha
upland (suited to maize) 0.75 ha
orchard (inter-cropping possible) 0.50 ha

Similarly, if total crop-storage capacity is six tonnes, but only three tonnes of this is secure and
rodent-proof, this fact will be specified in the resource pool statement. Likewise, while 'Pasture'
might be an adequate description of a land parcel if all cattle/donkeys/buffalo/sheep are to run
together as a combined herd, it would not be adequate if the producing dairy cows are to be given
preferential treatment. The pasture might then be quantified as:
good fresh pasture (for cows only) 0.50 ha
rough grazing (all other stock) 0.70 ha

Resource-use time dimension
From the viewpoint of resource use in relation to time, resources fall into two groups: those which
provide a flow of services over time, and those which consist of a consumable store or stock of
materials or other farm resources. Land, family labour, oxpower, tractors and machinery, crop
storage space, fences and livestock housing are examples of service-generating resources which
provide their services as a flow over time. A store of seed or agricultural chemicals in a shed, a pond
of irrigation water, a contract to produce sugarcane, or a shed of animal feed are examples of store-
type resources, since once they have been consumed in the production process they cease to exist.
Flow resources have a time as well as a quantity dimension. Because agricultural production does not
occur instantaneously but over some time period of months or years, the resources allocated to such
production must be provided at discrete points in time or during specific time periods. Thus a one-
hectare paddy crop to be grown from March to June might require the following resources: one
hectare of irrigated land, 90 days of family labour, 21 days of ox work etc. This would be sufficient
information if the purpose is to compare the economics (input costs versus output value) of paddy
against other crops; but for planning purposes a more satisfactory statement concerning paddy-
enterprise resource needs would be as follows:
March April May June
land (ha) 1 1 1 1
labour (days) 40 10 5 35
oxen (days) 18 0 0 3
where 12 types of inputs are identified rather than just three, each having a quantity-time dimension.
In short, in planning where the resource inputs of an enterprise have to be time-scheduled, e.g., by
months, then the farm resource pool also has to be specified in these same quantity-time units.
Labour-days or ox-days in March are, respectively, different resources from labour-days or ox-days
in June.
However, the more closely that the resource needs of a production activity are specified (as
occurring in June, early June, first week of June etc.), the greater the degree of detail needed in
quantifying the resource pool. Obviously this could result in the specification of an unmanageably
large number of resource items. Some limits must be imposed on the procedure of adding a time-
dimension to flow-type resources; but, on the other hand, a resource pool constructed without noting
when resources are to be available would be of little use in planning. The problem is largely
removed by the fact that farm resource-using activities fall into four groups according to their relative
needs for precision in time-specification, viz.:
· Highly time-specific: These are farm production activities which must be operated according to some tight time schedule or
calendar of operations. Examples are found in the production of vegetables/fruit/poultry/fish to be marketed on festival days
such as Id or Chinese New Year. They must be ready on that day, not three days earlier or they would spoil, nor one week later or
there would be no demand. All production operations leading up to final sale will be according to a close timetable. Thus, for the
activities producing these commodities, the time dimension of inputs might be specified as closely as within a given week,
perhaps even on a given day, rather than within, say, a looser one-month timeframe. In planning this type of enterprise it follows
that the resource pool must also be specified in similar small quantity-time units: e.g., so many labour days or so much transport
in the first/second/third week of August etc. Note also that production on a continuous-flow basis throughout the year (as with
orchids, pigs, coconuts etc. on some specialist farms of Types 3, 4, 5 and 6) implies ensuring adequate resource availability on a
continuous basis throughout the year. Paradoxically, such production is thus highly time-specific on a continuing basis.
· Moderately time-specific: Typical of these activities is the planting/production of a crop under

seasonal irrigation conditions. The appropriate time-unit for operation scheduling will usually be as
broad as one month. In South India, the monsoon will probably arrive in late May and provide enough
rain for paddy fields to be ploughed during June. The irrigation channels will start flowing in July to
permit planting. July, August and September are the growing months, followed by harvesting some
time in September. In this case, crop input requirements, and farm resource specification, need be on
only a monthly basis (e.g., June: oxpower and labour; July: planting labour; September: harvest
labour etc.).
· Low time-specific: Activities in this category include some rain-fed crop production where the
agricultural year typically consists of only two periods, the wet and dry seasons. If the farming system
(in terms of the number of production activities) is not very complex, it may be sufficient to define
farm resources only on the basis of season. The timing of operations will be quite flexible and if a
wet-season crop is not to be followed by a dry-season crop it will not matter much, within limits,
how long some mature crops are left in the field (e.g., for cassava this might be many months).
· Not time-specific: A few agricultural enterprises are not managed with respect to time or the
seasons. A small two-cow dairy herd kept for household milk supply is one example. Commonly such
stock are run on the common lands along with the other livestock of the village. They receive no
special attention whether they are in milk or not, and inputs are the same throughout the year. The
resource supply for this type of activity need not be specified as having a time dimension.
Family labour resources
Obviously family labour is a very important resource on small farms. In some areas of high
population pressure, small farms might support nucleus and extended families of up to ten or more
persons, all of whom except the very young and very old can supply some labour. But the actual
available supply is often difficult to measure because family labour has quantity, quality, time and
often custom dimensions. Difficulty in measurement arises from the fact that the different family
age/sex population classes often generate different amounts of labour service (e.g., as differentially
provided by young men, women, older children and grandparents), and that some farm
operations/tasks are labour-type specific while others can use any class of labour. The custom
dimension complicates things further by insisting that certain tasks which could well be done by men
(or women) must in fact be done by women (or men) (van Herpen and Ashby 1991).
Assuming that such activity-specific available labour has been identified, the procedure for
quantifying the general family labour resource is to standardize all remaining labour according to
some common quality unit, frequently adult male equivalents (AME). In the following example, an
extended family of 18 persons shown in column (2) is converted to a general farm labour force of 9.1
'adult male equivalents' in column (4). When dealing with these typically heterogeneous populations,
this or a similar standardization procedure is necessary. But the limitations should be noted.
Col (1) Col (2) Col (3) Col (4)
Family Conversion
Age Group in Work Force in Adult Male Equivalents
Members Factor
Years
M F M F
0 to 8 2 3 0 0 0
9 to 15 2 1 0.5 0.5 1.5
16 to 55 3 4 1.0 0.8 6.2
56 and over 1 2 0.6 0.4 1.4
Total: 18 9.1
First, the conversion coefficients of column (3) are necessarily subjective and thus somewhat
arbitrary. Even if correct for one society, they might not be for another. In Nepal a man might be
considered old at 50; but in parts of Iran he might be thought still capable of hard work at 70. Second,
the male to female comparison suggested in column (3) is a generalization to which there will be
many exceptions. There are many field tasks which women can or do in fact do better than men (e.g.,
planting seed, plucking tea, weeding, harvesting paddy). In these tasks the 1:0.8 comparison ratio of
male to female labour is invalid (it possibly holds for heavy tasks). Finally, for some jobs children
might in fact be equivalent in labour supply to only half an adult male, but for such jobs as herding
livestock they are fully equivalent because one child can do as much as one adult.
Before leaving this topic, one further possible problem should be noted. The above procedure will
result in a farm labour supply that is (it can be assumed) arithmetically valid but in some cultures this
might still not be a measure of family labour actually available. Many families set a high value on
what in developed countries would be called 'leisure' but which in South Asia would be regarded by
even poor families as necessary participation in social and religious festivals, community affairs and
village politics. The labour supply standardization procedure might result, for example, in an
apparent labour supply of, say, 750 AME labour days; but when this is adjusted for non-farm socio-
cultural demands on the family's time, the actual supply might be only half this. The point could be of
obvious practical importance in the planning of new farms (on irrigation or settlement projects etc.)
where such planning should obviously be based on actual rather than apparent family labour
resources.
Resources versus assets
Quantifying a farm resource pool is roughly akin to constructing an asset statement for the farm family,
but there are differences. An asset statement is a listing of all property owned and its value. Value
might be based on the (imputed) productivity value of the assets (Section 7.2), but is more commonly
based on their current market value. An asset statement might be prepared periodically as an
instrument for measuring a family's economic progress over time (in terms of its changing 'net worth'),
or occasionally as evidence of security in applying for a bank loan etc. A family's asset statement
includes both farm and non-farm property and other items of financial value.
As indicated in Section 2.2.2, the main structural characteristic of small farms of Types 1 and 2 is
their orientation to internal generation of most of their needed resources. Farms of the other types
must also generate their resources, at least in the long run, but do so through the medium of cash sales
and external or off-farm purchase. The various ways by which small farms can obtain their resources
are discussed in Sections 4.4.1 and 9.3.1.
Perhaps the best way to conclude this discussion of resources is to summarize the relationships
between resources, capital and costs, as shown in Figure 3.3. This shows (a) how the five categories
of resources discussed above become, in their economic dimension, either fixed or operating capital;
(b) the various types of costs which are generated by the use of capital; and (c) how each
resource/capital category (1) to (5) of Section 3.3.3 contributes to final total farm costs. Figure 3.3
serves as a useful introduction to discussion of the whole-farm service matrix (Section 5.3) and the
evaluation of past whole-farm system ex ante operational planning (Chapter 9). While Figure 3.3
refers to a whole-farm system, the cost structure of an enterprise or activity subsystem would be
analogous.
In Figure 3.3, starting with the first resource category of long-term fixed material resources in column
1, row (ii) shows this as becoming farm fixed capital. In row (iii) this type of capital gives rise to
fixed capital costs. Then, going to row (iv), this group of fixed costs forms one component of total
farm fixed costs (the other component of farm fixed costs being those fixed costs which might arise
from institutional resources as discussed below).
Returning to column 1 and continuing down to row (v), this indicates the purposes for which total
farm fixed costs are incurred: replacement of capital resources as these wear out or become obsolete,
routine maintenance of these capital items, and (for some fixed capital or institutional items only)
costs of operating them so long as these costs cannot be allocated to any specific farm activity or
enterprise - if they could be so allocated, they would be included as a variable cost in column 3.
Continuing, row (vi) is a more precise description of each of the cost purposes of row (v):
'replacement' is achieved by setting up a depreciation fund (depreciation costs), 'maintenance' by
incurring costs for repairs or periodic servicing of capital items, while 'operating' remains as
'operating'.
FIGURE 3.3 - Relationships among Categories of Resources, Capital and Costs in a Whole-
farm System Context
Column 2 refers to institutional resources. In row (ii) these do not become either fixed or working
capital (although one could argue that such institutional resources as a licence to produce and sell
milk to a town market is a sort of capital). In row (iii) the use of institutional resources could
generate a type of farm fixed costs known as general charges (see Section 5.3.1), e.g., the payment of
a land or road or house tax which purchases the right to use or occupy these resources (as distinct
from the physical resources themselves). In row (iv) such general charges or institutional fixed costs,
if any, would combine with capital fixed costs from column 1 to give total farm fixed costs. In
addition, in row (iii) some of the institutional resources might also incur variable costs (e.g., payment
of an output-based tobacco production tax), in which case they would combine with the variable costs
of column 3 and (if relevant) column 4 to give total farm variable costs in row (iv) or total direct
production costs for the system in row (v).
Considering resource category columns 3 and 4 in Figure 3.3, these show short-term material and
financial resources as operating or working capital in row (ii), which generates variable costs in row
(iii); these in turn are totalled in row (iv) as total variable costs of the system. The specific purpose
of these could be shown in rows (v) and (vi). Note that, as indicated in the diagram, while financial
resources are used to pay for the fixed and direct farm costs, they may also incur variable costs
through such items as bank fees and interest charges.
Family labour is shown as a resource in column 5 of Figure 3.3. In contrast to hired labour, it is
usually not costed. Instead, family labour is taken as receiving income (for its services of labour and
management) as a beneficiary of the farm system. However, as outlined in Section 7.2.5 and
illustrated in Tables 7.8 and 7.9, family labour is costed on an opportunity cost basis in the
calculation of such total productivity measures as total factor productivity and return on capital or
equity.
Row (vi) of Figure 3.3 shows how all fixed and variable costs arising from the employment of the
various categories of resources become total farm costs.
The specific types/categories of resources used in and generated by individual farm enterprises and
activities are discussed in Sections 4.1 and 9.3.1.
3.4 REFERENCES
Dillon, J.L. (1992). The Farm as a Purposeful System, Miscellaneous Publication No. 10,
Department of Agricultural Economics and Business Management, University of New England,
Armidale.
FAO (1990). Guidelines for the Conduct of a Training Course in Farming Systems Development,
FAO Farm Systems Management Series No. 1, Food and Agriculture Organization of the United
Nations, Rome.
Feldstein, H.S. and J. Jiggins (eds) (1994). Tools for the Field: Methodologies Handbook for
Gender Analysis in Agriculture, Kumarian Press, West Hartford.
Feldstein, H.S., C.B. Flora and S.V. Poats (1990). The Gender Variable in Agricultural Research,
IDRC, Ottawa.
Gabriel, T. (1995). The Human Factor in Rural Development, Belhaven Press, London.
Gondowarsito, R., C. van Stanten and T. Bottema (1995). 'The Role of Women in Upland
Agriculture: Gender Issues Raised by Case Studies in Indonesia, Sri Lanka and the Philippines',
Palawija News (CGPRT Centre Newsletter) 12(2): 3-15.
Prabowo, D. and D.J. McConnell (1993). Changes and Development in Solo Valley Farming
Systems, Indonesia, FAO Farm Systems Management Series No. 4, Food and Agriculture
Quisumbing, A., L.R. Brown, H.S. Feldstein and L. Haddad (1995). Women: The Key to Food
Security, Food Policy Report, IFPRI, Washington, D.C.
Rola, M.F.M. (1995). 'Gender Roles and Attitudes in Upland Farming Systems in the Philippines',
Palawija News (CGPRT Centre Newsletter) 12(4): 1-12.
Shaner, W.W., P.F. Philipp and W.R. Schmehl (1982). Farming Systems Research and
Development: Guidelines for Developing Countries, Westview Press, Boulder.
van Herpen, D. and J.A. Ashby (eds) (1991). Gender Analysis in Agricultural Research,
Publication No. 204, CIAT, Cali.
4. FURTHER FARM-HOUSEHOLD
SYSTEM ELEMENTS: ENTERPRISES AND
ACTIVITIES AND THEIR BUDGETING
4.2 ENTERPRISES
4.4 ACTIVITIES
4.9 REFERENCES
'Every production of a specific crop on a specific soil type, in a specific season, with a specific husbandry technique, is a distinct
activity.'
Hans Ruthenberg (1976)
Structurally, a farm enterprise is a set of input-output relationships involving input resources which
are used to generate one or more final products. Such products are 'final' in that they are suitable for
consumption on or sale off the farm (beyond which point they might or might not be further
processed). Enterprises are subsystems and comprise the main structural building blocks of farm-
household systems.1 They are designated by their main product: e.g., a farm growing coconuts and
paddy operates coconut and rice enterprises. The main enterprise categories are field crops,
intensive-horticulture crops, tree crops, aquaculture and livestock.
1 In industrial and commercial economics, 'enterprise' is generally used to refer to a whole economic entity or system (e.g., plant,
factory, store) rather than to any of its individual subsystems.

In everyday language, the terms 'enterprise' and 'activity' are often used interchangeably. In technical
usage, however, these terms have quite distinct meanings. There are two reasons for this.
First, in systems analysis, many 'enterprises', if they are to be analysed as systems, must be broken
down into subsystems. 'Coconut enterprise' is a good enough description if it refers to growing and
selling coconuts; but if it includes selling some nuts, making coir from the husks of others, making and
selling yam from this coir, and making and selling charcoal from the coconut shells etc., it will
obviously be necessary to differentiate among these various subsystems of the coconut enterprise
(Figure 5.5). This is best done by referring to all coconut-based subsystems in aggregate as an
enterprise, and to each respective subsystem as a separate activity. Clearly, selling fresh coconuts
will involve different resource inputs, product outputs and management problems than spinning
coconut-husk yam. Thus a particular farm enterprise may involve one or more activities.
Second, some planning methods in farm-system analysis and construction (e.g., budgeting and linear
programming) require that the alternative tactics or technologies which can be used in the production
of some resource or final product be specified. To take a very simple example, production of one
hectare of paddy might require 100 kg of cattle manure or its equivalent as fertilizer. If the farmer has
no manure, possible alternative tactics for obtaining it might be for him or her to buy NPK fertilizer,
or to exchange labour with a neighbour for the needed manure, or to grow a green manure crop in the
future paddy field, or to buy a cow. These alternative tactics (or any particular mixture of them) are
each different activities which might be used to the same end of producing paddy. An important
aspect of farm planning is thus to determine which technological activity or activities might best be
used within each enterprise.
There are three ways in which 'activity' may be used as a technical term relative to the hierarchy of
agricultural systems of Figure 1.2. These are: (i) to refer to a particular technology or process used at
Order Level 1 or 2 in the farm system; (ii) to refer to a resource-generating subsystem of Order Level
3 in the farm system; and (iii) to distinguish at Order Levels 4 and 6 (a) a particular method of
producing the product of an enterprise from other methods of producing the same product or (b) the
production of a particular product in an enterprise which has a variety of possible products. In
general, which meaning should be given to the term 'activity' will be apparent from the context. Most
often it will have the enterprise-related meaning pertinent to systems of Order Levels 4 and 6. Fuller
consideration of the various types of activities relevant to the planning of farm systems is given in
Section 9.3.1.
4.2 ENTERPRISES

As noted above, an enterprise consists of a farm subsystem aimed at the output of final product. It may
involve one or more activities in terms of the technologies used and the form of the final product.
For farm management systems analysis in all Modes (as defined in Section 2.1.8), an essential
property of enterprises is that they be possible of identification and disaggregation from the whole-
farm system context in which they occur, i.e. that their boundaries be defined and that their input-
output relationships be measurable. Where the enterprises already exist as part of a mixed farm
system this proceeds by the three sequential steps of: (i) examining the farm system to identify the
main commodities being produced; (ii) identifying the resources used in relation to each commodity;
and (iii) identifying and quantifying enterprise input-output relationships in the form of an enterprise
budget table as outlined in Section 4.3 below.
In terms of their structure, enterprises may be broadly categorized as simple, composite or complex.
A simple enterprise is one which can be readily identified within and disaggregated from a whole-
farm system. The extreme case is found on mono-crop estates and single-enterprise farms (e.g., Sri
Lankan tea estates, Singaporean pig farms) where the enterprise subsystem is practically equivalent to
the whole-farm system. Only slightly more difficult to identify/disaggregate/quantify are the
buckwheat and apple enterprises on small Himalayan hill farms; but the disaggregation of, e.g.,
cotton/wheat/paddy/livestock enterprises on a mixed Sind farm is usually more difficult because of
the structural interrelationships among enterprises.
A composite enterprise is one which aggregates what structurally or logically should be considered
as two or more different enterprise subsystems into a single composite enterprise for one of two
reasons - the practical difficulties of disaggregation, or the fact that the work involved in
disaggregation might not be warranted for the purpose at hand. An example would be the 'livestock'
component of a typical Peshawar mixed farm where the livestock could consist of two cows (for
milk, butter, ghee, cheese sale and/or consumption), two oxen for ploughing (but sometimes rented
out), three young sale cattle, one camel and three donkeys for transport, six sheep for wool (to be sold
or spun) and five goats for milk and meat. All these classes of stock would usually be run/pastured
together most of the year, none receiving any special treatment. While the different classes of stock
can be readily identified, it would be very difficult indeed to disaggregate them into seven separate
livestock enterprises, i.e., to define meaningful boundaries for each livestock species subsystem. For
most planning purposes they could be regarded as a single composite 'livestock' enterprise.
A similar situation arises on the intensive vegetable farms in the mountain zones of Java, where five,
six, seven ... vegetable species occupy the same field simultaneously, some directly dependent on
others for shade, wind protection or live trellis support, and where each relay of each crop is partly
dependent on previous relays for residual fertilizer and pest reduction. It is very difficult to
disaggregate the species into separate bean, maize, sweet potato ... enterprises. Possibly the most
difficult of all systems to disaggregate are the highly mixed forest-gardens of the wet tropics, partly
because the mix of 15 or more species is continuously changing (McConnell 1992).
A complex enterprise is one in which there is more than one important product and where there is
some considerable degree of cycling of resources within the same enterprise, usually via the farm
resource pool. This is common in traditional agriculture as exemplified in Figures 2.4 and 2.5. By
comparison, most enterprises on modem commercial farms are structurally simple. Using wheat
production as an example, the situations are compared in Figure 4.1.
As shown on the righthand side of Figure 4.1, on modem commercial farms most resources are
typically purchased, wheat is often grown as a sole crop for grain which is sold, and that is the end of
the matter. On the other hand, in the traditional Asian situation, wheat is usually combined with
several other enterprises and most resources are farm-generated. Thus, in the lefthand-side example
of Figure 4.1, wheat both uses and produces some of its resources (retained seed and stockfeed
cycled as oxpower and manure fertilizer). The enterprise has several products: grain (for family
sustenance, sale and farm use), retained seed, bhoosa (straw, some of which might be sold for use in
brickmaking and some retained for use as livestock feed/bedding), and stubble which will be used or
sold to other villagers as grazing. In addition, the enterprise might provide the basis for kitchen-scale
food processing/marketing activities. Figure 4.1 clearly illustrates the complexity of what at first
glance might seem a fairly simple enterprise.

4.3.4 Unit budgets
An enterprise or activity is defined and quantified in terms of a budget table (as illustrated by Table
4.1) which, relative to some specific time span, defines the boundaries of the subsystem. A budget is
essentially a listing of all resource inputs to an enterprise or activity and their costs, and all outputs
and their values, both inputs and outputs being measured over some specified time period (Barnard
and Nix 1973, Ch. 14; Brown 1979, Chs 2 and 3; Makeham and Malcolm 1986, Chs 8 and 9; Upton
1987, Ch. 14). As appropriate, a budget can also show the difference between total costs and returns
which is enterprise net return or, if no allowance is made for fixed costs, enterprise gross margin
(as discussed in Section 4.3.8). The time span to which the budget refers needs always to be
specified and borne in mind. This period of time may be the production period of the output involved
or some other period such as an annual cycle.
FIGURE 4.1 - Comparative Structure of Traditional and Modern Wheat Enterprises
From a time perspective there are three kinds of budgets pertinent to enterprises and activities:
· Planning budgets are forward-looking projections of the resources which will be required to obtain some anticipated outcome.
They are prepared as a basis for formulating an activity, enterprise or whole-farm production plan, usually for the next crop or
seasonal or annual phase (if it is a long-term activity). They are concerned with what should happen.
· Control budgets are used to maintain a current check on and adjust resource supplies to an
enterprise or activity once it is in operation, i.e., once the planning budget has been activated. They
are not used on small farms but are an important management tool on estates. Usually they take only a
financial form and are designed to control fiscal expenditure in the activity or enterprise. The degree
of control can be very high: e.g., on Malaysian and Sri Lankan rubber, tea, oil-palm and coconut
estates the daily-prepared control budget can show on any day of the operating year the amount of
money spent on and the amount and value of product harvested from the enterprise, down to
practically the last cent and kilogram. Control budgets are used for these and other continuous
activities (such as on orchid, dairy, pig and poultry farms). Obviously, the shorter the production
phase, the less need there is for budget control.
· Evaluation budgets are backward-looking summations of what did happen. They are essentially
accounting documents intended to measure the past performance of an activity or enterprise over
some period, usually the prior phase. They are better referred to as operating statements because
they state with certitude the facts they contain (inputs, outputs etc.). These might be intended only to
offer an accounting of past results or, more analytically, to allow diagnosis of weaknesses within the
system (or subsystem) and prescription of remedial action (Chapter 7).
TABLE 4.1 - Example of an Enterprise Planning Budget: Inputs, Costs and Returns for 2.5
Acres of Dryland Ginger in the Wet Zone of Sri Lankaa
Input Real level (per 2.5 acres) Per acre or unit

level
Labourb
Family or
Operation Timing Labour days Labour days
hired
Clear land Mar 25 F 10
Clean drains Mar 38 F 15
1st fork Mar 125 H 50
2nd fork Mar 88 H 35
3rd fork, smooth Apr 63 H 25
Line for planting Apr 5 F 2
Hole, plant, Apr 38 F 15

mulch
Fertilize Apr 5 F 2
Weed once May 25 F 10
Fertilize Jul 5 F 2
Weed twice Jul 40 F 16
Guard Dec-Jan 75 F 30
Harvest Jan 40 H 16
Total family 256 102

labour
Total hired 316 Rs 3 160 Rs 1 264

labour
Materials
Item Amount Price Cost Cost
Seed Mar- 1 635 kg Rs 5.90/kg Rs 9 647 Rs 3 859

Apr
Tools Mar- 63 25
Apr
Watcher's hut Mar- 25 10

Apr
Fertilizer Jul 340 kg Rs 1.60/kg 544 218
Straw mulch Jul 3 750 bundles Rs 375 150

0.10/bundle
Crop storage Jan (already exists on 0 0

farm)
Transport Jan 63 25
Market sacks Jan 250 100
Total materials Rs 10 967 Rs 4 387
Total all costsc Rs 14 127 Rs 5 651
Output
Clean ginger 6 825 kg Rs 4.50/kg Rs 30 713 Rs 12 285
Gross margin Rs 16 586 Rs 6 634

a Constructed in February 1993 for the production period March 1993 to January 1994.
b Days are in AME.

cFor purposes of later analysis related to Figure 4.5, note that these total costs are all variable costs.
There are no fixed costs pertinent to this particular budget formulation.
Feedback
The use of evaluation budgets for diagnostic purposes implies feedback from evaluation to planning
budgets: adjustments to the system made on the basis of the former will occur via the planning budget
prepared for the next operating phase. Even if there are no weaknesses to correct, the actual
enterprise outcome will seldom be exactly as predicted in the planning budget. Referring to Table
4.1, hired labour actually used might be 270 rather than 316 days and actual achieved yield might be
6 270 rather than 6 825 kg. These results would now flow back as guides to hopefully more accurate
updated estimates in the next planning budget.
Data content
In addition to summarizing likely enterprise or activity results, planning budgets are primarily
concerned with identifying those resources and other factors which might limit or constrain
production. Accordingly, in constructing them, it should be kept in mind that what might be apparently
unimportant outputs (by-products) of one system might be critical inputs to another (especially on
Type 1 and 2 farms, as discussed in Chapter 9 and illustrated in Figures 2.4 and 2.5).
In control budgets the most important and often the sole data recorded might relate to flows of finance
to the activity or enterprise; alternatively, the critical factor to control might be labour, or in an
irrigated desert environment it might be water. Control budgets are not further discussed.
The basic data content of an evaluation budget depends on analytical circumstances, specifically the
operating objectives of the enterprise or farm (Chapter 6). If this is profit maximization, financial data
alone will enable the evaluation; but if the objective is subsistence food production, the data will
relate to resource inputs (mainly labour) in comparison with food outputs. But while either of these
might be sufficient for only an accounting of what did happen, they would in themselves be practically
useless if the analysis was in diagnostic mode. Evaluation budgets in this mode have the most
extensive and detailed data requirements. These are discussed in general terms in the remainder of
this chapter and are applied in examples of systems' comparative analysis in Chapter 7.
Hazards
The different types of budgets and their purposes will be clear enough. But budgets are not always
what they seem. On one group of commercial farms it was routine office practice, until recently, to
propose planning budgets for each enterprise in the approved way but then, at the operational stage,
as the need arose, to allocate cash which had been budgeted for the wheat to the sheep, fertilizer
budgeted for the pyrethrum to the wheat etc. In consequence, at the end-of-year evaluation stage, all
that could be said was that some total amount of cash, fertilizer, labour etc. had disappeared
somewhere into the farm, but as to the economics of the individual subsystems, or whether
uneconomic enterprises should be curtailed or replaced by more efficient ones ... 'O God I know not'.
For some purposes and modes of analysis (e.g., description), the budget might be specified at real
level (i.e., at the actual levels of the input and output variables) and refer to the enterprise as it
actually exists. But for most planning and evaluation purposes, since - e.g. - 2.5 acres of ginger cannot
be directly compared with 1.9 hectares of paddy, it will be necessary to standardize the various
enterprise or activity budgets by bringing them to a per acre or other unit level as in the righthand
column of Table 4.1. Further, for purposes of comparison between enterprises or activities, these
unit-level budgets must also refer to a common time basis such as, e.g., per hectare per year.
Use of land area as the basis for standardizing crop and livestock budgets is not always appropriate
or possible. The most appropriate standard unit to use will depend on the category to which the
enterprise or activity belongs and the particular farm type/situation. Most field-crop and tree-crop
budgets will in fact usually most appropriately be on a land-unit basis. So will budgets for the
horticultural crops on large farms. But for small farms growing tree or vine crops and some
horticultural crops (anthurium flowers in Sri Lanka, vanilla in Java), a more appropriate budget basis
might be costs and returns per 100 plants or even per single tree (as for the houseyard farms of Java
growing three or four high-value clove trees).
Further, in the wet tropics the tree crops in particular are often grown in highly mixed stands: a
forest-garden farm in Kerala or in Java might consist of six coconut palms, two breadfruit, three
jackfruit, four coffee trees, three pepper vines, etc. These species-enterprises would be best
budgeted/evaluated on a per tree or per vine basis (or for some purposes these various species might
be consolidated into a single composite enterprise budget).
In other situations, even though a species might be present in large numbers as part of the whole of a
farm system, it might be grown/managed/exploited without any regard to the land area it occupies - as
in the case of nipah palms along the rivers of Trengganu, sago palms on village lands in Irian Jaya
and palmyrah palms in the dry zone of Sri Lanka. Perhaps the extreme case is found in Kordofan
where the baobab trees, each of which is the property of some family or clan and an important basis
of their semi-nomadic existence, might be scattered along the clan's seasonal migration routes over a
distance of 600 kilometres or so. Budgets for these and similar tree-crop enterprises could obviously
not be expressed on a unit of land area basis; more appropriate would be inputs/outputs per family,
or, since labour is a common factor in their control or exploitation, per labour day of effort expended
in maintaining or exploiting some specified number of trees.
Livestock enterprises can present special problems in preparing standardized budgets for purposes
of comparison. In some cases it will be possible to construct budgets for dairy cows, sheep etc. on a
'per hectare of land use' basis; but the yak herders of Haa and the camel people of Wajir would not
have the slightest idea of the number of hectares over which they range.
Highly-mixed sedentary herds/flocks, as found in parts of Pakistan and North India, require
standardization of two kinds (assuming they can be disaggregated). First, the separate species (cattle,
camels, sheep etc.) and classes of each species have to be standardized in terms of some common
animal unit (AU) basis2 and then, as discussed above, the budgets for each of these (standardized)
species-enterprises might have to be compared on the basis of some common input or production
factor (such as land or labour or cash required or generated etc.).
2 In the following example, a farm's herd of five head of mixed cattle and a flock of 16 head of sheep are each standardized (on
the basis of approximate feed requirements) by taking one lactating cow as equivalent to one animal unit (AU). Each other type of
animal is then expressed as AU relative to a lactating cow, for example:
Cattle Equivalence Factor Sheep Equivalence Factor
milk cow 1.0 AU all types 0.15 AU
ox 0.9 AU
calves 0.3 AU
Population equivalence in AU:
cows: 2 head x 1.0 AU = 2.0 AU

oxen: 2 head x 0.9 AU = 1.8 AU
calves: 1 head x 0.3 AU = 0.3 AU
Total cattle in AU: = 4.1 AU
sheep: 16 head x 0.15 = 2.4 AU
Total cattle and sheep in AU = 4.1 + 2.4 = 6.5 AU.
It is important to note, however, that from a farm management as opposed to an animal nutrition point
of view, it may sometimes be more relevant to standardize livestock on some such basis as labour
required by each species/type, or cash inputs required by each, etc. The above nutrition-based
equivalence factors are for purposes of illustration only.
Long-term crop and livestock enterprises require a somewhat different approach in budgeting.
· Evaluation budgets of long-term enterprises (or activities) are usually prepared on an annual or seasonal basis and relate to the
performance of the enterprise over its most recent operating phase. These single-phase budgets are the same as budgets
prepared for short-term enterprises (or activities) and ignore what might have happened to the enterprise in earlier phases and
what might happen to it in future phases.
· Planning budgets of long-term enterprises (or activities) are a different matter. Here the
inputs/outputs of the enterprise (or activity) must usually be specified for each year of its future life -
over 20 to 25 years for coffee, 20 to 40 years for cardamom, 60 to 70 years for coconut, etc. Methods
for doing this are discussed in Section 10.11.
The necessary degree of detail in a budget is determined by the mode and purpose of analysis. Thus
Table 4.1 would probably be adequate for descriptive purposes and for planning purposes, but too
detailed for accounting evaluation and too vague for diagnostic purposes.
Table 4.1 offers much information relating to labour: jobs to be done, monthly timing of operations,
whether family or hired. This might or might not be necessary for planning. On the other hand, the
budget says nothing about the type or quality of labour to be employed. Are these workers to be men,
women or children?...or do the various tasks specifically require women or men? This would be
important on an estate; or if the budget referred to a crop such as coffee requiring school children as
harvesters. Regarding yield, 6 825 kg of good ginger is projected to be produced, but how much sub-
standard ginger will also be obtained?... and would this be enough to provide a basis for a small
kitchen-scale oil extraction activity?
Explicit and implicit information
In addition to offering explicit information, a budget should be amenable to implicit extension. Such
measures as activity gross margin should permit derivation of further, more detailed measures: gross
margin per hectare, per day of family labour, per Rs 100 of operating capital etc. This is essential in
comparative analysis (Chapter 7).
4.3.4 Unit budgets
In order to apply some important planning methods (e.g., linear programming, Section 9.6), Table 4.1
contains too much information and is too awkward to be used directly. For such purposes, unit-vector
budgets are needed. These are simplified, stripped-down versions of a budget such as that of Table
4.1 from which the following unit-vector budget for ginger has been developed:
Identification:
Date: Feb. 93
Crop: Ginger
Level: One acre
Production period: 11 months, Mar. 93 to Jan. 94

Gross margin: 6 634 Rs
Resource requirements:
Land 1 acre
Family labour
Mar. 25 AME
Apr. 19 AME
May 10 AME
Jul. 18 AME
Dec.-Jan. 30 AME
Operating capital
Mar.-Apr. 4 995 Rs
Jul. 369 Rs
Jan. 287 Rs
Other
Storage 1 acre equivalent

In the above unit budget, the inputs of Table 4.1 are consolidated under the headings 'land', 'family
labour', 'operating capital' and 'other'. Labour and capital are given a time-dimension (for reasons
discussed previously). Also, all information relating to costs versus returns is now consolidated into
a single item 'gross margin Rs 6 634', i.e., no separate cost or return items are shown. Information
relating to some items, e.g., hired labour, is now discarded because item-specific costs have been
consolidated in the single gross margin item.
All other data of Table 4.1 are now discarded (but of course they would not be discarded if the
purpose were detailed post-production evaluation rather than planning). On the other hand, some
resources which might only have been implied in Table 4.1 might now become important. Ginger
storage was listed in Table 4.1 but not the amount or quality of storage needed. If it is possible that
these aspects of storage could prove limiting factors to future or expanded production, they would be
entered as explicit resources in the unit-vector budget. Use of this type of budget in farm planning is
illustrated in Chapter 9.
In the conventional pre-systems approach to farm management, analysis was largely restricted to field
production operations; the post-harvest aspects of an enterprise such as on-farm storage, kitchen
processing, family consumption, transport and disposal of commodities beyond the farm gate were
regarded as falling within the province of specialists in other fields - marketing experts, processing
engineers, family nutritionists etc. Such a compartmentalized approach is possibly appropriate on
those farm types where products are simply sold off the farm, e.g., sugarcane to a mill or raw milk to
a bottling plant. However, as emphasized in Chapter 2, such a level of structural simplicity is rare on
small mixed Asian farms where products and by-products are used and disposed of in many different
ways. Here, to define the boundaries of an enterprise as encompassing only field production
operations would be quite inadequate. On Type 2 (semi-subsistence) farms it would tell only half the
story and on Type 1 (subsistence-oriented) farms only a fraction of it. More specifically, it would
overlook the possibilities for household income generation through on-farm or kitchen processing and
local marketing, and at village level of establishing cottage industries; moreover, where agricultural
resources are already exploited to their potential, local further processing of crop outputs often offers
greater scope for rural development than does farm development per se. A Bhutanese farm wife
making corn cakes in her kitchen and selling them around the streets of Tashigang is as important a
component of the farm's corn enterprise as her husband's corn-growing activity. In this hill country,
too, it is not unusual to come across a group of farmers who have each carried out a sack of oranges
or potatoes to the nearest road, maybe walking for two or three days, in the hope of selling them.
Farm production problems become rather irrelevant. The people know these things; why not the
universities?
Clearly, under these 'beyond-the-field' circumstances - and they are often the norm not the exception -
if an enterprise is not to lose its 'pith and moment', a systems approach to farming and farm
development is called for. But while systems concepts are increasingly accepted, their actual
application in farm development projects is not remarkable. It remains at best a field in which the
coconuts have only nodding acquaintance with the pineapples and the carp or tilapia, and at worst an
arena in which the 'two and forty jarring sects confute'.
Flowcharts
As an aid in understanding the often complicated post-harvest flows of products and byproducts from
an enterprise, a useful first step is to sketch them before attempting to incorporate their main elements
into a budget. An example is shown in the flowchart of Figure 4.2 for post-harvested paddy on a
small Type 2 farm in Bhutan. The chart emphasizes those portions of the paddy crop retained for
family use; quantities are not indicated.
FIGURE 4.2 Example of a Flowchart: Post-harvest Handling and Disposal of a Paddy Crop,
Bhutan
Source: Data from a survey by Nim Dorji and the senior author, Thimpu, 1988.
Table 4.1 is an economic budget. As a planning budget, it lists all the planned inputs and outputs and
their values whether or not these will incur actual out-of-pocket costs or will yield actual cash
receipts.
For some purposes a financial budget will be required. This is an edited version of an economic
budget which lists only actual financial outgoings (or costs) and income (or returns). For example, in
Table 4.1 if the seed is to be retained from the previous crop and the tools are on hand, these
economic 'costs' would now be excluded. Similarly if the cost of the watcher's hut in Table 4.1 is
based on providing an annual depreciation amount for this capital item and is not an actual cash
outlay, this also would be excluded from the financial budget.
On commercial farms, since all or most inputs are bought and outputs sold, economic and financial
budgets are practically equivalent (except that non-cash charges such as depreciation are excluded
from the latter). At the other extreme, on Type 1 farms there will be few if any purchased inputs and
few sold outputs so that financial budgets are hardly relevant.
In a non-commercial environment and where no actual markets exist for inputs or outputs, their values
must be imputed. In Table 4.1 straw was valued at 10 cents per bundle. If the farm had not actually
bought the straw this might be based on an estimate of what it was worth if used on some other crop
(such as mushrooms) or what the farmer could have earned by working for a neighbour instead of
staying home bundling straw for his ginger.
Usually the most important economic input is family labour. In Table 4.1 the 256 days of family
labour cost nothing in terms of cash outlay and the enterprise gross margin is Rs 16 586. This conveys
an accurate picture only if family labour in fact has no value; but if some realistic value for family
labour could be imputed (e.g., by what it could earn off the farm or from the farm's roadside boutique)
then Table 4.1 conveys a false picture of ginger economics: e.g., if family labour had a real value or
opportunity cost of Rs 10 per day, the gross margin would be decreased by Rs 2 560.
On Type 1 and 2 farms there is often no reference point for the valuation of resources. In
consequence, deciding on an appropriate opportunity cost or shadow price for labour, oxen, buildings
etc. can be quite subjective. The facts alleged in a budget relating to subsistence-oriented farming,
whether for a single farm or for the sector as a whole, will often not stand their ground. This is
especially so in Fields C (i.e., sectoral development) and D (i.e., advice for policy making), and not
infrequently - sad to say - the 'economies' of an enterprise (or farm or industry based upon it) are
nudged up or down to serve a non-objective purpose.
Depending on the purpose of analysis there are four ways in which an enterprise (or activity) can be
evaluated, or four measures of economic performance which can be applied3: enterprise net return
(NR), gross margin (GM), operational gross margin (OGM or 'activity price') and cost of production
(COP). NR, GM and OGM are outlined in this section. COP is discussed in Section 4.3.11 below.
3 With appropriate modifications to their calculation, the measures of performance outlined here can also be applied on a whole-
farm basis. Such applications and extensions to calculate total productivity measures are presented in Sections 7.2.1, 3 and 5.
Net return
Net return (NR) is the most appropriate measure if the purpose is an accounting evaluation of past or
future projected performance. It is obtained as NR = TGR - TC where TGR, total gross return of the
enterprise, is the sum of all outputs times their prices, real or imputed, and TC is total enterprise cost,
again real or imputed. TGR needs no further discussion.
TC consists of two components, total variable cost (TVC) and total fixed cost (TFC). Variable costs
(VC) are those of input items the amounts of which change (usually but not necessarily proportionally)
with the size or level of output of the enterprise: the labour, seed, fertilizer, etc. of Table 4.1 are
examples. Fixed costs (FC) are those of input items, usually services rather than physical things,
which remain constant regardless of the size or level of enterprise output: taxes on a barn used to
store grain or other produce, the interest costs on an irrigation system, a depreciation amount set aside
to eventually replace a cocoa dryer are examples. Fixed costs are further discussed in relation to
whole-farm systems in Sections 5.3 and 4.
The relationships between level of output and fixed, variable and total costs are exemplified in
Figure 4.3. With zero output, variable costs are also zero, but increase as output or size of the
enterprise is increased. Fixed costs are constant throughout (but some activities or enterprises will in
fact incur no fixed costs). TC at any point over the output range is simply the sum of VC and FC; here,
in this particular example, at an output level of 50 units, they are Rs 300 + 100 = Rs 400. Note that
while VC will always increase (i.e., slope upwards) as Y increases, the relationship between them
need not be linear. This is in contrast to FC which will always be linear and parallel to the Y or
output axis for a given configuration of the production system.
FIGURE 4.3 - Example of Relationships between Level of Output and Fixed Costs, Variable
Costs and Total Costs
Gross margin
Gross margin (GM) is the most appropriate measure of enterprise performance if the purpose is an
operational rather than accounting one, e.g., to make rapid comparisons among three, four, five ...
enterprises with a view to expanding some or contracting others. Gross margin is obtained as total
gross returns less variable costs, GM = TGR - VC, i.e., fixed costs are ignored since, by their nature,
they have to be met whatever is produced (and even if nothing is produced). Justification for this is
based on the fact that, by definition, farm fixed capital (ploughs, work oxen, irrigation, etc.) can
usually be used by alternative crops; if it is currently used for cotton and cotton prices drop, the
farmer will want to evaluate the possibilities of alternative crops - corn, sesame etc. In the short run
he or she is stuck with whatever supply of fixed capital is on hand and can budget out the alternatives
by considering only those input items which will change if he or she adjusts from cotton to wheat or
sesame. This would be an application of partial budgeting (as outlined in Section 4.6).
Operational gross margin
Operational gross margin (OGM) is somewhat equivalent to net return. OGM = TGR - DC, where DC
is enterprise direct costs and consists of variable and fixed components. The difference is only one of
emphasis; the DC of an enterprise consists of all costs directly associated with that enterprise, usually
calculated on a per unit of enterprise basis (per hectare, per head of livestock, etc.). The DC of an
enterprise thus consists of its variable costs plus some proportion of the farm's fixed costs which
should logically be borne by the enterprise. The concept of DC is a piece of terminology used with
linear programming (Section 9.6). An example is discussed in situation (d) of Table 4.2 below.
Assignment between fixed and variable costs
The assignment of input items to the fixed or variable categories of enterprise total costs will usually
present no problems. However, some items do not fall naturally into one or other category. Whether
they should be put in the fixed or variable baskets will depend on circumstances. Four situations are
shown in the budgets for a ginger enterprise in (a), (b), (c) and (d) of Table 4.2.
Situation (a) is taken directly from Table 4.1. There are no storage costs, variable or fixed, and no
other fixed costs, thus net returns of Rs 16 586 are the same as gross margin. In situation (b) the
enterprise budget now contains a storage cost of Rs 500. This consists of taxes paid on the storage
shed which remain constant regardless of how much ginger is stored or grown. It is thus included as a
fixed cost. In situation (c), at a cost of Rs 400, the farmer instead now rents storage on a cost per
cubic unit basis; thus ginger storage is here a variable cost because if yield was twice as high or
twice as many acres of ginger were to be grown, twice as much storage would be needed and its cost
would be twice as much.
Situation (c) is also different in that where previously hired labour was a variable input of 316 days
times Rs 10 per day, labour is now not casual but permanently attached to the household and must be
paid (at Rs 10 per day) regardless of how much ginger is grown or its yield. Thus in (c), labour
becomes a fixed cost to ginger. (This assumes that ginger is the only crop on the farm: if there were
other crops, the fixed labour costs would be shared among them.) This is offered as a caution that
some costs do not, simply because of their nature, always belong in either of the fixed or variable
categories. An important variation of this occurs on estates where the resident labour force incurs
both variable costs (a daily wage) and fixed costs (housing, clinic, worker insurance, retirement fund
etc.).
Situation (d) is different again. Here the budget is to be used for planning aimed at selection of the
best mix of several enterprises (ginger, corn, cotton etc.) and is on a per unit (acre) basis. Now the
storage fixed costs, which were assigned wholly to ginger in situation (b) because this was the only
crop on the farm or the only crop to need storage, are shared between ginger and some other storage-
using crop, say corn, one fifth or Rs 40 per acre to ginger and four fifths or Rs 160 per acre to corn
according to their likely relative use of this facility. In (d), the purpose of the budget is to obtain
enterprise direct costs and thence the operating gross margin (OGM or enterprise net 'price') for use
in formulating optimal combinations of several enterprises (e.g., by linear programming, Section 9.6);
the budget is not directed at the evaluation of past performance.
TABLE 4.2 - Evaluation of a 2.5 Acre Dryland Ginger Enterprise under Different Cost
Conditions
For 2.5 acres of ginger (a) (b) (c)
Variable Cost (Rs) (Rs) (Rs)
Hired labour 3 160 3 160 (fixed)
Seed 9 647 9 647 9 647
Tools 63 63 63
Watcher's hut 25 25 25
Fertilizer 544 544 544
Straw mulch 375 375 375
Storage 0 0 400
Transport 63 63 63
Marketing 250 250 250
Total Variable Costs (VC) 14 127 14 127 11 367
Total Fixed Costs (FC) 0 500a 3 160b
Total Costs (VC + FC) 14 127 14 627 14 527
Gross Returns (GR) 30 713 30 713 30 713
Net Returns (NR) 16 586 16 086 16 186
Gross Margin (GM) 16 586 16 586 19 346
For 1 acre of ginger (d)
Gross Returns (GR) 12 285
Total Costs without storage 5 651
Total Costs with storage of Rs 40 5 691

Total Direct Costs (DC) 5 691
Operating Gross Margin (OGM) 6 594

a For storage.
b For hired labour.

A whole-farm budget is the structural parallel of an enterprise or activity budget, with budgets of the
several enterprise and activity subsystems now aggregated. There is only one change: those fixed
costs which cannot be logically assigned to any activity are now included as whole-farm fixed or
common or overhead costs (OC). This latter term is preferable because it distinguishes more clearly
between fixed costs that are specific to an enterprise or activity and those fixed costs which are
accrued on a whole-farm basis. Overhead costs consist of land, water and road taxes, insurance and
repair costs of buildings which cannot be assigned to specific activities, and all other input costs
relating to the upkeep and maintenance of the whole farm. These overhead costs can be both fixed and
operational: e.g., both the annual tax paid on a farm water-supply pump and the fuel costs of operating
it. The relevant criterion is that they cannot be logically charged against specific activities or
enterprises.
The derivation of a whole-farm budget from the budgets of its constituent enterprises is shown in
Table 4.3. In this simple case there are only three enterprises on the farm. This might convey an
impression that enterprises can be clearly separated with well-defined boundaries. On small farms,
however, there will usually be a ragtag of miscellaneous activities directed to supplying feed for the
oxen, fertility for cash crops etc., so that budgets must be prepared for these also even if they do not
generate a final product or positive (imputed or real) gross margin; or they can be aggregated into a
smaller number of composite activities; or, when neither of these is possible, they must be included
as whole-farm costs. (Such resource-generating activities are discussed in Section 4.4.)
TABLE 4.3 - Whole-farm Evaluation Budget derived from Enterprise Budgets
Ginger Cotton Soybeans Whole Farm

Item 2.5 acres 1 acre 0.7 acres
(Rs) (Rs) (Rs) (Rs)
Gross Returns (GR) 30 713 20 000 10 000 60 713
Variable Costs (VC) 14 127 10 000 5 000 29 127
Fixed Costs (FC) 0 3 000 2 000 5 000
Overhead Costs (OC) 8 000
Total Costs (TC) 14 127 13 000 7000 42 127
Gross Margin (GR - VC) 16 586 10 000 5 000 31 586

Net Returns (GR - TC) 16 586 7 000 3 000 18 586
Household Non-farm Income 3 000
Household Total Income 21 586
Whole-farm budgets are used primarily as end-of-year accounting statements or summaries of farm
performance over that or some other operating period. They are generally not operational tools,
except when used to project whole-farm performance over several years, e.g., to project the total
results of a five, six, seven ... year farm-development program.
At the whole-farm budget level, one other important element can now be introduced, namely
household non-farm income. In Table 4.3 this is included to obtain household total income. Where the
purpose of analysis is not agro-technical farm planning but rather to ascertain the economic wellbeing
of the farm household, the next step would be to analyse household outgo or expenditure on non-farm
items necessary to keep the whole farm-household system operational - family expenditure on
medical services, food, clothing, children's schooling, meeting social obligations etc.
But this step is seldom taken in formal farm management economics (which by-and-large remains
preoccupied with agro-technics). This is a pity because it leaves one with an incomplete picture of
what is really going on within the most important system component, the household. In particular, it
gives no knowledge at all about which beneficiaries are getting what from the system ... and, at the
end of the day, this of course is what it is all about.
An enterprise or product-producing activity can also be evaluated in terms of its unit cost of
production (COP). This is obtained as its total costs divided by the quantity of output achieved with
those costs, i.e., TC/Y. In Table 4.1 ginger COP is Rs 14 127/6 825 kg or Rs 2.07 per kg.
COP is usually not a relevant measure of enterprise performance in providing advice to farmers
(Field A), especially on small farms where interest is in (net) income rather than what it costs to
obtain that income. But in other fields COP can be the most appropriate measure of performance. In
Field C the manager of a milk bottling plant will want to know what the milk COP is on farms
supplying raw milk to his or her plant (as a basis for setting his or her buying and selling prices and
consequent profit margin); he or she will not be interested in farmers' incomes as such. In Field D,
policy makers in Fiji will want to know what the copra COP is on Fijian farms in comparison with
Indonesia, Philippines, Malaysia etc. in order to establish sound industry research and development
policies - and perhaps even to decide if the Fijian industry has a viable long-term future.
One area where COP is important in Field A is on tea, rubber, cocoa, coconut and other estates
where COP rather than net returns or gross margin is the conventional measure of estate performance,
both on the same estate over time and in comparison with other estates (see Table 5.4). But taken by
itself COP conveys no information, except perhaps by implication, of the real economics on these
estates. In the 1960s, due largely to political instability, those tea estates of Sri Lanka with the lowest
COP per pound of made tea were the worst estates in every other respect, the low COP being due to
the fact that the green tea was just being harvested and processed, with little or no expenditure on
field or factory or other estate maintenance. In those circumstances COP was, if anything, only a
measure of the degree of current system exploitation. (The phase was only a temporary one.)
4.4 ACTIVITIES

As noted previously, an enterprise - or any parallel activities of Order Level 4 or 6 (Figure 1.2)
making up an enterprise - is a production process directed to the production of one or more final
products for sale or consumption (but it might also indirectly generate some amount of resources). All
other types of activities in a farm system (i.e., those of Order Level 1, 2 or especially 3) are directed
in one way or another to the generation of resources for use on the farm: but they might incidentally
also generate some final products. As with enterprises, these resource-generating activities have to be
quantified and evaluated. This requires the construction of activity budgets which are in all essential
respects similar to the enterprise budgets discussed above.
The main types of resource-generating activities are now briefly discussed. Their consideration in
whole-farm system planning is examined in Section 9.3.1.
Activities and enterprises as resource generators
Those enterprises or activities which also generate some amount of resources fall into three
categories: they might generate the resource (a) for their own use or (b) for the specific use of some
other enterprise or (c) for general non-specific use by other enterprises. In the sketch of Figure 4.4,
situation (a) exists in the dairy enterprise: enough of the female calves produced are retained to
replace the older cows as these are culled from the herd. In regard to this resource the enterprise is
essentially self-perpetuating. Situation (b) is represented by the second enterprise, maintaining a beef
cattle herd in parallel with the dairy herd in which the latter generates male calves for the specific
use of the beef enterprise. Situation (c) is represented by the flow of another resource, dairy cow
manure, into the general farm resource pool for use by any of the farm's crop enterprises/activities. In
all these cases the resource-generation aspects of the dairy cows are incidental to their main purpose
of producing a final product, milk. This type of enterprise (dairy cows) as a resource-generator
would be quantified/budgeted/evaluated as was the ginger enterprise of Table 4.1; its total output
value would include the value of calves and manure.
FIGURE 4.4 - Example of Enterprises as Resource Generators
Major resource-generating activities
On mixed farms, particularly livestock farms, most resource-generating activities consist of
substantial sets of inputs and operations which are similar in all respects to main enterprises except
that the main output now is a resource for use on the farm, i.e., an intermediate product rather than a
final product. Common examples are a fodder crop grown for dairy cows and a legume crop grown to
enhance fertility for some following cash crop.
When financial evaluation of such activities is required, this will proceed as in Table 4.1. If no final
product is generated, the results of this evaluation will be a negative 'net return' or gross margin
equivalent to activity total costs. If some final product is incidentally produced, the value of this will
partly offset activity costs; in some activities, costs might be fully offset by the value of final
products.
Budget evaluation of resource-generating activities will be necessary in two situations: first, when
the activity as a resource generator is to be compared with alternative activities generating the same
resource; and second, when cost and return data for all enterprises and activities on a farm, which
have been evaluated separately, are to be pooled to enable a whole-farm budget to be constructed (as
in Table 4.3 above).
As with enterprises, it is sometimes not possible to define the boundaries of resource-generating
activities, in which case several might have to be aggregated as a composite, or even accounted for
on a whole-farm basis.
Resource generation by purchase or barter-exchange activities
Probably the simplest way for a farmer to acquire a resource is to buy it or exchange some other
product or resource for it - payment of money for the use of a parcel of land, exchange of family
labour with a neighbour for use of her oxen etc. Such purchase or barter is equally an activity and is
quantified and evaluated in terms of a budget, albeit a very simple one. Budgets for two such
activities - renting land and exchanging labour for ox draught power - are shown in Table 4.4.
TABLE 4.4 - Example of Budgets for Resource Generation by Purchase or Barter-exchange
Activities
Activity
Budget Item
Rent land Exchange labour for oxpower
Input cost Rs 600 20 labour days
Output return 0.25 ha land 4 ox days
Gross margin -Rs 600 Rs 0
In the case of renting land, considering the activity only as an activity (and not considering any
eventual recovery of the rental cost from the future sale of the crop to be grown on the land), the
activity result is an annual cost of Rs 600 or a net return or gross margin of -Rs 600 per unit (0.25 ha)
of activity. This conceptual approach is appropriate where there are several alternative ways of
acquiring land (by rent, lease, outright purchase, clearing jungle etc.) and the costs or returns of each
have to be evaluated regardless of what might actually be produced on the land. However, if the
renting of land is a normal cost of growing some specific crop, 'renting land' would then be only a
normal cost item in the budget for such an enterprise and no separate land-renting activity budget
would be necessary.
This also applies to the case of exchanging labour for oxpower. The only difference between the two
cases is that the gross margin of exchanging labour is zero because it is assumed that the value of
labour exchanged is equal to the value of ox days received (at least in the eye of the farmer). Again, it
should be stressed that these resource-generating activities are here quantified and evaluated as
activities; if they happen to be integral parts of some specific enterprise then no separate budgets are
needed.
For one important method of analysis used in prescriptive farm planning, i.e., linear programming
(LP), such budgets as those of Table 4.4 are reduced yet further. As shown in Table 4.5, they are re-
written and couched, with slight rearranging, in relation to the farm's resource pool. The GM of each
activity is the same as before and indicates the money amount by which the value of output of the farm
system increases or decreases due to one unit of the respective activity being included as a resource
generator in the farm production plan: renting land as an activity in itself would reduce annual farm
income by Rs 600; exchanging labour would have no direct effect. (However, both would have a later
indirect effect when actually used in some enterprise.)
TABLE 4.5 - Example of Resource-generating Activity Budgets in Linear Programming Format
Activity
Budget Item
Rent land Exchange labour for oxpower
Gross margin (Rs) -600 0
Resource pool
Land (ha) -0.25
Labour (days) 20
Oxpower (days) -4
In LP format the input and output items of the budgets are stated in terms of their effect on or demand
for those corresponding resources shown in the resource pool. Renting land has a negative demand on
land in that it contributes to rather than detracts from whatever supply of land is available. Similarly
the budget of the labour exchange activity indicates that this barter activity would require or use 20
days of farm labour and that it would generate four days of oxpower. The above activities and their
'budgets' are of an operational nature; their use in whole-farm planning is illustrated in Section 9.6.
Resource generation by transfer activities
This again is a very simple kind of activity. It consists of transferring some particular resource now
assigned to one activity/enterprise to use by some other activity or enterprise. The ginger enterprise
of Table 4.1 provides an example. About seven tonnes of ginger storage capacity is needed. If this
does not presently exist on the farm, a storage shed could be built at some known cost, or
alternatively the farmer might be able to convert part of his or her cow barn to storage space for
ginger. This type of activity 'generates' resources by transferring them from activities/enterprises
where they are not needed or where they have a low use-value to other activities/enterprises where
they are needed or have a higher use-value.
Just as for other types of activities, for farm-planning purposes transfer activities need to be
quantified in budget form. Thus the above examples of transfer of space from use by cows to ginger is
quantified by the following simple budget of this transfer activity:
Budget item Transfer of space from cows to ginger
Gross margin of activity (Rs) 0
Cow shelter lost (m2) 1
Ginger storage gained (m2) -1
The above example might appear trivial, but resource generation by use-transfer is a powerful tool in
more sophisticated analysis of farm systems using linear programming (Section 9.6).
External activities
All the activities noted so far have been internal to the farm. In addition, where the opportunities
exist, households often supplement these on-farm activities by off-farm work and petty trading.
However, such off-farm activities are usually undertaken more to supplement household income than
to obtain resources for the farm component. An exception could be when an above-normal amount of
resources is needed temporarily for farm development as distinct from routine farm production.
Application and use of resource-generating activity budgets
Budgets for the various types of resource-generating activities outlined above are applied in whole-
farm planning in Section 9.6.

Using a budget such as that of Table 4.1 as a base, several extensions can be made from it for
operational purposes, viz.:
(i) preparing partial budgets;
(ii) developing conditional or parametric budgets, equations or graphs;

(iii) deriving further performance factors for use in comparative analysis of farm subsystems;
(iv) preparing long-term evaluation budgets for ex ante appraisal of investment in an activity or
enterprise which extends over many years; and
(v) assessing an enterprise or activity under conditions of uncertainty.
Uses (i) and (ii) are discussed in the following sections of this chapter. Further performance factors
are discussed in Chapter 7. Evaluation of activities over time is considered in Chapter 10 and under
uncertainty in Chapter 11.

As the term indicates, partial budgeting is concerned with the evaluation of only selected parts of a
system - only parts of a process, activity, enterprise, farm service matrix or of the whole farm (Dillon
and Hardaker 1993, Ch. 5). In farm management in Field A (i.e., advising farmers) partial budgeting
is most commonly used to test possible adjustments to farm-level systems which are presently not
working well. At higher levels it is widely used as the basis for structuring agricultural development
projects (e.g., to assess the impact which a dam or a new road system might have on some parts of
existing farm-level systems).
In Table 4.6 partial budgeting is applied to the 2.5 acre ginger enterprise of Table 4.1 to evaluate the
likely impact which irrigation might have on this presently dryland enterprise. Part (a) of Table 4.2 is
the base budget for this activity, the present 'without irrigation' situation. In making the comparison it
is necessary to consider only those few elements or parameters which might change if irrigation were
to be used. These changes comprise additional yield whose value is partly offset by additional
labour, transport and marketing costs and the lower price for irrigated ginger. As shown in part (b) of
Table 4.6, the net effect of these changes, i.e., 'after' versus 'before', is an increase in gross margin of
Rs 4 422.
TABLE 4.6 - Evaluation of Adjustments to an Enterprise or Activity by Partial Budgeting
(a) Base budget (unirrigated ginger) Change from (b) Partial budget (irrigated
(a) to (b) ginger) showing changes
in input costs and returns
Input cost (Rs) Extra labour (Rs)
Hired labour 3 160 + Water channels 700

Seed 9 647 0 Water application 600
Tools 63 0 Extra harvest 100
Watcher's hut 25 0
Fertilizer 544 0
Mulch 375 0
Storage 0 0
Transport 63 + Extra transport (Rs) 15
Marketing 250 + Extra marketing (Rs) 80
Total cost (Rs) 14 127 + Total extra costs (Rs) 1 495
Output Output changes
Ginger (kg) 6 825 + Extra ginger (kg) 1 500
Price (Rs/kg) 4.50 - Lower price (Rs/kg) 4.40
Gross value (Rs) 30 713 + New gross value (Rs) 36 630
Gross margin (Rs) 16 586 + New gross margin (Rs) 21 008
Extra gross margin (Rs) 4 422

At process level, partial budgeting is a useful tool for adjusting systems of Order Level 1 and 2 which
do not warrant the use of more sophisticated methods of response analysis (Chapter 8). At whole-
farm level, partial budgeting can be used for adjusting the enterprise or activity mix of a system if this
does not warrant the use of the more powerful programming methods of Chapter 9.

The type of budget of Table 4.1 refers to an activity as formulated at some specific point in time.
Obviously it will cease to be valid if there are changes to prices, costs, technology or the underlying
production conditions. In most situations such changes will in fact occur either through the influence
of the market on prices and/or costs, through changes in technology introduced by the farmer causing
changes in yield and cost, or through yield changes due to Nature. If the budget is going to be needed
for future or continuing use, it is desirable to anticipate likely future changes in the relevant critical
parameters and build these into the base budget. This is done by adding, at the time of budget
construction, conditional (or parametric) extensions to it (Dillon and Hardaker 1993, Ch. 5). On the
other hand, one might not be concerned with future change but only with considering several
alternative cost/price/technical scenarios at the moment of budget construction (e.g., What if costs
were to double? What if yields increase/decrease by 20%?...). Conditional extensions to reflect such
possible changes in critical parameters are a useful analytical tool.
The steps in conditional extension of an enterprise or activity budget are straightforward, viz.:
(i) Identify the parameter which is most significant in terms of its impact on budget results and likelihood to change. In the
example of Table 4.1 this is, say, ginger sale price.
(ii) Specify the range over which this parameter is likely to change; here, say, from Rs 3 to Rs 6 per
kg.
(iii) Specify the budget result (e.g., ginger gross margin per acre) as a simple equation in terms of the
parameter of interest (e.g., ginger price). Thus, for the particular example being used here, from Table
4.1:
Gross margin/acre = Total revenue/acre - Variable cost/acre,
i.e.,
GM = TR - VC = (Yield/acre)(Price/kg) - VC = (Y)(py) - VC = 2 730py - 5651
where Y denotes ginger yield per acre in kg and py denotes ginger price per kg in Rs.
(iv) Graph the single-parameter equation from (iii) as shown in Figure 4.5 and use this graph as an
alternative to solving the equation for all likely prices.
Such a parametric-budget approach permits the validity of the base budget to be maintained even
though the variable of concern might change over time.
FIGURE 4.5 - Example of a Single-variable Parametric Budget showing Ginger Gross Margin
per Acre for Ginger Prices ranging from 3 to 6 Rs/kg
Parametric budgeting is readily expanded to accommodate two or three variable parameters - beyond
three variables, graphical presentation of results is usually too complicated. As an example, Figure
4.6 shows a three-variable parametric budget relating to the growing and chipping of cassava tubers
by an estate for sale as livestock feed. The objective of this budget is to allow rapid determination of
total estate net returns.
The unstable parameters of concern to management on this particular estate are: annual field yield of
fresh cassava tubers per acre, the percentage of dry processed chips recovered per ton of wet tubers,
and the sale price of dry chips. The graphs of Figure 4.6 show how the results of a base budget for
this enterprise might be parametrized in order for the budget to be 'solved' for any combination of the
three unstable parameters: e.g., at a yield of 10 tons per acre, with a chips recovery rate of 35% and a
chips sale price of $13 per picul4, total annual net returns for this particular West Malaysian estate
would be approximately $330 000. Analogously to the case of single-variable parametric budgeting
(Figure 4.5), the graphs of Figure 4.6 are based on the net returns equation
NR = TR - TC = (Y)®(py) - TC
where Y is tuber yield in tons per acre per year, r is the dry chips recovery coefficient and py is chips
sale price per picul. Y, r and py are variable (i.e., parametrized in the budget over their relevant
ranges) and total costs are given.
4 Picul is a common weight measure in parts of South East Asia. It equals 100 catties of 1.3 pounds each and is thus equivalent to
about 60 kg.
FIGURE 4.6 - Example of a Three-variable Parametric Budget showing Annual Net Returns on
a Cassava Estate by Tuber Yield, Dry Chips Recovery Rate and Sale Price

Because budgets are the basis of most of the subsequent farm-household system analysis presented in
the following chapters, it is appropriate to conclude this chapter with a list of do's and dont's as a
guide to good budget construction. This applies specifically to enterprise and activity budgets but
also generally to other types of budgets such as whole-farm, partial, evaluations-over-time etc.
(1) The first requirement is for an accurate and unambiguous heading or title to the budget, e.g., not just Tea' if reference is to
some specific type of tea grown at a specific location and elevation.
(2) Specification should be given of the level or size of the enterprise or activity to which the budget
data refer - e.g., whether to one, two, six, seven or 100 hectares; or whether to one or 100 head of
camels. It is not helpful when one encounters budgets headed simply 'Sheep' or 'Economics of Goats'
etc. without indication of the number, type, breed and location of the animals under consideration.
(3) If the budget contains standardized or converted units - e.g., when all the cows, calves, oxen,
buffalo etc. on a farm are converted to common animal units - the basis of standardization should be
specified in the table or a footnote.
(4) Where appropriate, a listing should be given of the separate operations performed within the
budgeted enterprise or activity. When input requirements are time-specific, this should be
accompanied by a calendar of operations showing when each operation is performed.
(5) A list should be given of all inputs and, if appropriate, as in an enterprise budget, their unit costs
and the total of these costs.
(6) A similar listing of all products generated (by-products as well as main products), their
respective unit prices if they are sold and the total value of the output should be given.
(7) The manner of disposal of the items mentioned in (6) above, together with the quantities disposed
of by cash sale and various other means, should be shown. This is particularly important in traditional
farming systems.
(8) Grade and price classifications should be given for that part of products sold. Sometimes a
farmer will sell his or her produce ungraded, in which case an average price is appropriate. Often,
however, he or she will do his or her own grading, resulting in two or more product categories, each
with its respective price. This level of detail is often necessary in diagnostic analysis where it will
serve as a pointer to specific system weaknesses (Chapter 7).
(9) An aggregation should be given of cash sales to obtain total cash returns for the budgeted
enterprise or activity.
(10) An aggregation should be given of all economic returns (cash receipts plus the real or imputed
values of those parts of production which are used on the farm or bartered or given away). If only
produce which is sold is considered, this might grossly underestimate actual total production and thus
the real efficiency of an enterprise or activity.
(11) A statement should be given of enterprise net returns or gross margin, preferably both.
(12) A budget prepared for diagnostic purposes must obviously contain sufficient detail to permit
problem diagnosis. To describe an input as 'fertilizer - 100 kg' might be sufficient for some purposes
but it would be near useless if a specific fertilizer problem exists on a farm and there is a need to
know what kind of fertilizer is being used and perhaps even when it is being applied and how
frequently.
(13) Although budgets prepared for financial accounting or economic evaluation need not contain
physical data (e.g., ox days used, kg of crop harvested etc.), it is generally desirable that they do so,
especially if they might be used at some future date to chart the progress of the farm over time.
Because of price changes, inflation and devaluation, the information - drawn from comparison of
'then-and-now' budgets - that a Javanese paddy farmer advanced from having an income of Rp 80 000
in 1975 to one of Rp 300 000 in 1995 would be spurious; what would be important is whether his or
her paddy yield increased or actually declined, and by how much.
(14) The time-dimension of the enterprise should be noted, preferably in the budget heading. This is
often essential information in relation to short-term and annual field crops: a one-hectare bean crop
which occupies the land for six weeks will, on an annual basis, require only one quarter of the land
resources required by one hectare of a six-months paddy crop; and, other things equal, where land is
in short supply a four-month paddy crop will be superior to a five-month crop. Some livestock
enterprises also will require a time specification, e.g., 10-, 11- or 12-week batches of broiler
chickens. Most livestock enterprises, however, will be operated on a continuing indefinite basis and
with these, as well as tree crops, the time basis of most budgets will usually be a one-year period.
(15) Finally, the date of construction and the sources of data contained in a budget should be
indicated - whether from records of a single farm, from a survey of several farms or simply the
analyst's estimates. This can serve two purposes. It will provide the opportunity for independent
check of the data should this become necessary, and it will indicate at least implicitly the degree of
reliability that can be attached to the respective data items and to the results of any analysis based
upon them.
4.9 REFERENCES
Barnard, C.S. and J.S. Nix (1973). Farm Planning and Control, Cambridge University Press.
Brown, M.L. (1979). Farm Budgets: From Farm Income Analysis to Agricultural Project Analysis,
published for the World Bank by Johns Hopkins University Press, Baltimore.
Ruthenberg, H. (1976). Farming Systems in the Tropics, 2nd edn, Oxford University Press.
Upton, M. (1987). African Farm Management, Cambridge University Press.
5. FURTHER FARM-HOUSEHOLD
SYSTEM ELEMENTS: PROCESSES,
STRUCTURAL COEFFICIENTS AND THE
WHOLE-FARM SERVICE MATRIX
5.2 INTERRELATEDNESS OF SYSTEM COMPONENTS: STRUCTURAL COEFFICIENTS
5.6 REFERENCES
'So soon as the time for ploughing is arrived, then make haste ... and bestir yourself early in the morning.'
Hessiod, c. 8-th Century B.C.
This chapter continues the discussion of the elements of the farm-household system as listed in
Sections 1.4 and 5.

Structurally, the main building blocks of a whole-farm system are its enterprises; these provide its
income in cash or kind. Enterprises may stand alone or be supported by resource-providing activities.
The building blocks of both enterprises and their enabling activities are agro-technical processes.
A process is the specific way in which a production operation is done, together with the levels or
amounts of resources used in doing it. A process specifies or implies the use of some operational
technology, the types and levels of resources used in its operation, and the types and levels of its
resultant outputs.1 Processes present two general analytical or management problems: (a) to select the
'best' or most appropriate technology, and (b) to operate the selected technology at its optimal level in
terms of inputs and outputs.
1 According to the general definition of a system in Chapter 1, a process is itself a system. Its components are its technology, its
inputs and its outputs. Interaction exists in the relationships among these inputs and outputs. The (management) objective
relative to a process system is to operate it (i) at a level which is optimal for the enterprise to which it contributes and (ii) in such
fashion as to achieve maximum efficiency in resource use, i.e., maximum output for a given input, or minimum input for a given
output. The boundaries of a process are defined by the specification of the process itself and, at least conceptually, a given
process system can usually be disaggregated into successively smaller subsystems.
Before looking further at these problems it will be useful to consider some examples. Table 5.1
presents a typical activity budget for maize production. While Table 5.1 (and similar activity or
enterprise tables such as Table 4.1) would be adequate as the basis for financial evaluation of this
activity, it is nonetheless a highly generalized statement: it offers little or no structural information
about any of the seven operations which comprise the activity, e.g., no information is given regarding
how the field was ploughed (mule, bullock or tractor?), the number of times it was ploughed or the
degree of tilth aimed at. Or considering operation (2), planting, no information is offered regarding
the several variations possible in this operation which could affect crop yield, e.g., depth of seed
placement, in-row and between-row plant spacing, seasonality of planting, the moisture level of the
field at planting time etc. Nor is any agro-technical information offered regarding any of the other
operations beyond the bare fact that they were done and the inputs which were used in doing them.
TABLE 5.1 - Budget of a Maize-growing Activity
Inputs per acre

Labour Oxen Costs/returns
Operation Materials
(days) (days) (Rs)
1. Ploughing 5 4 200
2. Planting seed: 7 kg 1 240
3. Irrigating water: 1 ac ft 1 50
4. Fertilizing fert: 40 kg N 1 200
5. Weeding 3 30
6. Harvesting sacks: 50 6 280
7. Drying wood: 2 m 3 400
Total inputs and costs 20 4 1 400
Outputs per acre
Item grain: 1 t 3 000
The seven operations shown in Table 5.1 represent seven operational subsystems, one each relating
to ploughing, planting, irrigating etc. However, due to the absence of data, it is not possible to judge
whether or not these subsystems consist of the best of all possible technologies to use in each
operation, nor whether each one is operated at maximum efficiency. Seven kilograms of seed are
specified but would it be better to use six or nine kilograms? Forty kilograms of fertilizer are
budgeted but might this better be 60 or 80 kilograms? In short, Table 5.1 tells nothing about the
optimality of resource use. Clearly, if interest is in ascertaining whether there are better ways of
performing these separate operations and whether the resources to be used in them are being used to
their maximum efficiency, these operations have to be examined in greater detail.
The operations of Table 5.1 are listed again in Figure 5.1. This also shows, for each operation, the
alternative technologies that might be employed in performing it and a range of input intensity levels
at which each technology might be implemented. For example, the 'ploughing' operation could be
done by tractor, by oxen or by spade; and whichever ploughing technology is chosen could be applied
one, two, three or four times: further, whichever of these possible combinations is used, it could be
applied to cultivate to a depth of 7, 14 or 21 centimetres. Alternative possible technologies and
alternative possible technology intensities or input levels are also shown for the other of the seven
production operations.
Each specific technology (i.e., the way or means by which an operation is performed), operating with
the application of one or more different kinds of (variable-level) inputs, is a process. The technology
operates either as a single-variable input process or, if it is used with two, three, four ... different
kinds of inputs, as a multi-variable input process. In Figure 5.1, three possible technologies are
shown for the planting operation: using two-row or four-row machines or doing this operation by
hand. Whichever of these planting technologies is chosen by the farmer, it can be used with a variable
amount or intensity level of seed (20-, 30- or 40-thousand seeds per acre). Each of the three planting
technologies is thus a single-variable input process where the variable input is seed used in
conjunction with a fixed input consisting of either a two-row, four-row or hand planter.
FIGURE 5.1 - Example of Alternative Technologies in Maize Production (per Acre Basis)
On the other hand, the operation 'ploughing' is a two-variable input process: each of the three
possible ploughing technologies (doing this operation by tractor or by oxen or by hand) can be done
or combined with some variable range of the two 'inputs' shown - depth to which the field can be
ploughed and number of times it can be ploughed. Thus there are a total of three by three by three, or
27, combinations of ploughing technology by depth by frequency which could be applied to this
operation. Each of these 27 combinations could at least potentially result in a different crop yield as
illustrated relative to number and depth of ploughings by the graphs in the lower half of Figure 5.2.
For analytical purposes, such processes are usually described and quantified in mathematical terms
as response functions or production functions which relate the output of a crop, livestock activity
etc. to the inputs used in producing it (Dillon and Hardaker 1993, Ch. 7). For a single-variable input
process, they are specified by the general functional form Y = f(S) which states that, e.g., crop yield
per ha (denoted by Y) is some function of the amount of seed used per ha (denoted by 5); or more
realistically as a multi-variable input process, Y = f(S, C) which states that yield is significantly
determined by both the amount of seed and the number of times the field is cultivated (denoted by C);
or more realistically still, Y = f(S, C, N, P) where the effects of nitrogen fertilizer (N) and phosphate
fertilizer (P) are also considered.
Figure 5.2 illustrates the concept of production functions. In the top figure, crop yield per ha (for
some particular set of climatic, soil and husbandry conditions) is graphed as a function of the single
input-variable seeding rate over the range from 10 000 to 50 000 seeds per ha. This depicts the
response or production function Y = f(S). In the lower figure, crop yield per ha (again for some
particular set of climatic, soil and husbandry conditions - including seeding rate but excluding
ploughing frequency and depth) is shown as a function of the two variable inputs depth (7, 14 or 21
cm) and number of ploughings (1, 2, 3, 4 or 5). This depicts the production function Y = f(D, K) by the
set of three single-variable input-output graphs Y = f(K/D) (for D respectively equal to 7, 14 or 21)
where D and K respectively denote depth and number of ploughings and the symbolism K/D indicates
that K is variable while D is fixed at some specified level (Dillon and Anderson 1990, Ch. 2).
The number of separate operations in an activity can be large (seven in the above example of Figure
5.1). The number of processes in an operation can also be large; e.g., in Figure 5.1 three technologies
are shown for the irrigation operation (the alternatives of splash, furrow or sprinkler application),
each of these operating over some intensity range (days' interval between water applications). One
could, if it were appropriate, list many more factors relating to irrigation which would have an effect
on crop yield, i.e., other dimensions in which irrigation could operate such as the amount of water
applied at each irrigation, the quality/salinity of water used, the time of day at which irrigating is
done etc.
However, such an examination of an operation in greater-and-greater detail by defining more-and-
more dimensions in which it exists, if done for all operations in all farming environments, would be
not far short of defining the total scope of all agricultural science. (Thus relative to the operations
listed in Figure 5.1, 'planting' would encompass the realm of agronomists; 'harvesting' and 'drying'
that of engineers; etc.) From the very large number of physical processes which do exist, farm
management is concerned with selecting and analysing only that relatively small number which are of
practical economic significance; or, in terms of the example of Figure 5.1, only those processes
which have or are thought to have a significant (economic) impact on crop yield. Just which of these
processes are 'significant' and thus to be subjected to economic analysis via response analysis, as
outlined in Chapter 8, will depend on analytical circumstances. As a practical guide, farm
management analysis would be confined to processes which have a relatively large effect on output or
which are expensive or which require large inputs of scarce resources. In short, in such activities as
those shown in Table 5.1 and Figure 5.1, there might be only two or three processes of sufficient
importance to warrant full economic analysis. (But whether such analysis is in fact to be done at all
will depend on the Field and Mode of the farm management analysis as discussed in Section 2.1.)
FIGURE 5.2 - Graphical Examples of Single-variable Input and Two-variable Input Production
Functions
Types of processes
In agricultural production there are three broad types of processes:
(i) Biological processes are those in which the stimulus or input operates directly on the biological system of an animal or plant.
Irrigation water applied to a crop root zone, feed digested by a dairy cow are examples. The great majority of economically
relevant agricultural processes are of this type.
(ii) Bio-mechanical processes are those in which the stimulus or input operates indirectly on a plant
or animal, its direct or immediate effect being on the physical micro-environment in which the plant
or animal exists. An example is the number of times a field is tilled. This will affect subsequent plant
growth/yield, but will operate indirectly through changing the physical/mechanical conditions in
which plant growth occurs - e.g., through changes to clod size, particle cohesiveness, water
infiltration and soil moisture storage capacity etc. At a practical farming level these bio-mechanical
processes are usually not distinguished from direct biological processes.
(iii) Mechanical processes do not exist as part of a plant/animal biological system, but act as
physical/mechanical support to such systems. Some of them can be operated at varying intensity
levels, such as the delivery of feed to a herd of cows by a conveyor belt driven at varying speed.
Here there might be some non-linear relationship between, say, fuel consumption and amount of feed
conveyed by the belt, and the relationship would appear as in Figure 5.3 B. In consequence,
determination of optimal belt speed would be economically relevant. Many types of mechanically
powered equipment - tractors, trucks, stationary motors - have such an optimal operating rate. But
with others the relationship between fuel input/speed and work output is linear over some specified
range (Part C of Figure 5.3); if the machine is operated beyond this range, it might simply
disintegrate.
The discussion of response analysis of processes in Chapter 8 is limited to biological processes.
Shape of input-output process relationships
Figure 5.3 illustrates some differently shaped production functions for the case of a single-variable
input production process. Each graph shows the physical input-output relationship or total physical
product curve as the level of the single variable input is increased with all other input factors held
constant. Note that, in graphs A and B, the law of diminishing returns (sometimes called the law of
variable proportions) prevails - beyond some point, as the level of the variable input increases with
no change in the level of other input factors, increases in output occur at a diminishing rate (the
marginal product is decreasing) and eventually, beyond the point of maximum output, output declines
in absolute terms (the marginal product becomes negative).
In graph A of Figure 5.3, which is the classical depiction of a production function, all three
theoretical stages of production are to be seen as the level of the variable input increases: first, an
initial stage of increasing returns with output increasing at an increasing rate - maximum efficiency
would never be achieved by operating the process in this stage since the marginal product is
increasing; second, a middle stage of decreasing returns with output increasing at a decreasing rate
(i.e., marginal product is positive but declining); and, third, beyond some maximum level of output, a
final stage where output declines (i.e., marginal product is negative). Graph B is more realistic in
showing only the second and third stages since increasing returns are not found in practice. Graph C
depicts a process in which output increases at a constant rate (i.e., linearly) up to some maximum
level and then remains constant (at least within the input range covered by the graph - it must be
expected eventually to decline). While graphs A and B are non-linear, graph C is of (segmented)
linear form. Note also that as depicted in all three graphs, output is zero when none of the variable
input is used. This indicates that the variable input under consideration is essential to production (as,
e.g., seed in crop production). Often, however, some output may still be obtained when none of the
variable input is used. In this case the input is to some extent optional (as, e.g., chemical fertilizer in
crop production).
FIGURE 5.3 - Stylized Production Functions or Input-Output Relationships for a Single
Variable Input
The reasons why the use of some kinds of inputs beyond a certain level causes a decline in output
will be intuitively obvious. As more and more water is added to a crop, keeping all other inputs
constant (especially drainage capacity), the field will eventually become waterlogged and yield will
decline; or as a crop is weeded with greater and greater frequency (i.e., weeding labour inputs are
increased), the point will be reached where workers will cause more damage in trampling the crop
than benefit. Or again, as in curve C of Figure 5.3, offering more feed to a cow than she can consume
might not cause actual milk output to decline, but beyond a certain level feed will be uneaten and go
to waste (and the value of output will decline relative to the increasing cost of inputs).
As noted, processes present two problems in farm systems analysis: (a) selection of the 'best'
technology or method of performing an operation, and (b) determination of the optimal level of
intensity at which the process should be operated.
· Technology: Usually, selection of the 'best' process is not difficult on small farms; it is often determined by the socioeconomic
environment. Selection of a two-row or four-row mechanical grain planter might present a real management problem in South
Dakota, but there would be no such alternative possibilities among the hill farmers of Nepal where all field operations are done
by hand. Where a farmer does face a management choice in technology selection, decisions can be made by budgeting out the
costs and returns of the alternatives (partial budgeting, Section 4.5).
· Optimization of input or intensity levels: The second problem of operating the 'best' process at the
'best' level can be a little more complicated. If the choice is between two, three, four ... discrete
alternatives, this also can be made on the basis of comparative partial budgets. If the process is of
such economic significance as to warrant more precision and if it relates to a continuous type of input,
then it might require the more sophisticated methods of response analysis (Chapter 8).
5.2 INTERRELATEDNESS OF SYSTEM COMPONENTS: STRUCTURAL

COEFFICIENTS

The next element of a system to be examined is its structural coefficients. From the system definition
of Section 1.1, a necessary property of any system is the existence of interrelatedness among its
components. In farm-household systems this takes the form of structural coefficients of two kinds.
Internal structural coefficients exist and operate within individual production activities or other
subsystems and define explicitly or implicitly the interrelatedness which exists between/among their
several sub-components. External structural coefficients are also often relevant for a given system,
such as a production activity/enterprise, but their function here is to quantify relatedness between the
given system or subsystem and other related systems or subsystems.
Internal structural coefficients are those which 'tie together' or link the various components of a (crop,
livestock or whole-farm) system. They are an especially important characteristic of those farm
production activities which generate part of their own resources, e.g., mixed livestock activities
(Makeham and Malcolm 1986, Ch. 12). The budget for one such activity, annual maintenance of a
sheep flock to produce wool and lambs, is shown in the lower section of Table 5.2.
This flock budget would probably be adequate if the purpose is to evaluate the financial results of
sheep production and perhaps compare such results with those of other activities.
TABLE 5.2 - Internal Structural Coefficients and Annual Activity Budget for a Self-sustaining
Flock of 100 Ewes
Internal Structural Coefficients
Ewe mortality rate 0.05
Lamb birth rate 0.90
Lamb mortality 0.09

Rams: ewes 0.05
Ewe and ram replacement rate 0.20
Activity Budget (Rs)
Input costs
Replacement rams (bred)
Replacement ewes (bred)
Feed 8 000
Veterinary 300
Shearing 1 900
Labour 2 000
Marketing 800
Total direct costs 13 000
Output returns
Culled ewes and rams 9 000
Male lambs 10 000
Female lambs 5 000
Wool 8 000
Total returns 32 000
Gross margin 19 000

On the other hand, however, the budget tells practically nothing about the techno-economic structure
of the activity, i.e., how the various internal components of ewes, lambs and rams, and the technical
performance of each, impact on each other and on the overall budget results.
Such interrelatedness must, of course, exist but in the budget itself it is only implied. To be of use for
some analytical purpose - e.g., analysis in problem-diagnostic mode - the implied coefficients have to
be stated in explicit form. The relevant internal structural coefficients are shown in the upper section
of Table 5.2. They indicate that in this example the normal death rate among ewes is five per cent
annually; 90 lambs are born annually per 100 ewes; nine per cent of lambs do not survive; five rams
are run per 100 ewes; 20 per cent of the (old) ewes and rams are replaced (by retained lambs)
annually.
There are three main reasons for taking explicit note of these internal structural coefficients. First,
they show at a glance the techno-economic conditions under which the results of an activity are
achieved (in this example the results are a self-sustaining flock of 100 ewes yielding an annual gross
margin of Rs 19 000). Second, they are essential information if the budget is intended for activity
problem diagnosis. Third and of primary interest here, they provide a basis for closer examination of
the interrelatedness of the various components of the activity - in this example between populations of
and returns from the lamb and ewe components of the activity.
The objective might be to quantify the separate effects of sheep losses and sales on the annual
maintenance of a constant-size flock of 100 ewes. Assuming that this particular aspect of the sheep
system is to be examined, a flowchart such as that of Figure 5.4 would be prepared showing the
interrelatedness of components, in terms of animal numbers, in more detail. This is an elaboration of
the relevant parts of the budget table. As noted, this flowchart refers to sheep populations. If some
other aspect of interrelatedness is of interest (e.g., feed supply versus births/deaths etc.), a flowchart
based on the relevant parameters would be prepared analogously to Figure 5.4.
From the previous discussion of parametric budgets (Section 4.6), most of the internal structural
coefficients of Table 5.2 will be recognized as 'critical parameters'. Such coefficients as lamb birth
rates etc. are critical in the sense that any significant change from their present level would obviously
flow through to alter the final result (sustainability and gross margin) of the activity. Such changes
would be evaluated by making conditional or parametric extensions to the budget of Table 5.2.
However, it might be noted that while internal structural coefficients are similar to critical
parameters, they are not exactly equivalent. Some factors, e.g., lamb mortality rates, are both
coefficients and critical parameters. But others, such as the amounts of wool cut from ewes and lambs
and the prices received for these, are critical parameters although they might not involve
interrelatedness among the several elements of the system and are therefore not internal structural
coefficients. Similarly, some activity input/cost items such as labour and feed might be economically
'critical' but would have nothing to do with interrelatedness among activity components.
The second type of system interrelatedness, that which is quantified in terms of external structural
coefficients between activities or enterprises, is illustrated by the example of Figure 5.5. This
presents budgets for four integrated coconut activities: growing coconuts for direct sale, or
alternatively using the nuts and by-products as inputs to the three related activities of making copra
(from coconut flesh), making charcoal (from shells) and making coir fibre (from husks). From the
previous discussion (Section 4.1), it is here useful to distinguish, on the one hand, between this set of
four coconut-based operations as an enterprise and, on the other hand, each of the product-specific
subsets of operations as activities. Note also that Figure 5.5 encompasses only a small set of the great
number of coconut activities that are possible (Grimwood 1975).
FIGURE 5.4 - Flowchart depicting Role of Internal Structural Coefficients in Annual
Maintenance of a Self-sustaining Flock of 100 Ewes
Activity budget 1 of Figure 5.5. which relates to producing 4 000 nuts annually from the standard 64
palms per acre, would apply if these nuts are simply to be grown and sold. Budgets 2, 3 and 4 apply
if instead the nuts are retained on the farm for further processing. They show operations, inputs, costs,
outputs, gross returns and net returns for the respective activities (all of which are possible
components of the overall coconut enterprise). For present purposes what are of interest are the
external structural coefficients which link these several activities together. These coefficients are
briefly explained.
FIGURE 5.5 - Structural Relationships among Four Coconut Activities (per Acre Basis)
In Figure 5.5 the first relationship between Activities 1 and 2 is through the output of nuts in Activity
1 as this determines the raw material available for Activity 2, copra making. This linking coefficient
has a value of 1.0 - all nuts are used for copra. But on some other farm the coefficient could be, say,
0.75 if three quarters of the nuts are processed and one quarter sold as nuts. This would occur, e.g., if
25 per cent of nuts are first grade and sold as food, while all second-grade nuts are processed as
copra.
In Activity 2, copra making, the first operation is husking 4 000 nuts, which produces 4 000 clean nuts
plus 4 000 husks. The clean nuts are then split into halves which contain the (future) copra. These are
first sun-dried for half a day then loaded into a fuel-burning drier for five days. After unloading, the
cured copra is hand-extracted from the half-shells and is then subjected to the further operations
shown under Activity 2. The now-empty shells, on the other hand, are disposed of in two ways: two
thirds of them are recycled back as dryer fuel for curing some future batch of copra (i.e., two thirds of
the shells from any batch would provide sufficient heat to dry the copra of that batch). This is shown
in Figure 5.5 as a coefficient value of 0.67 (of 8 000 half-shells) so that 5 360 half-shells are cycled
back as fuel into the copra-drying operation.
The remaining output of one third of the half-shells, i.e., 2 640, now becomes raw material input for
the next activity, charcoal making. Thus the output of charcoal in Activity 3 is governed by the (shell)
fuel requirements of Activity 2. Note, however, that these coefficients of two thirds and one third are
subject to some variation. Firing the half-shells in copra curing requires considerable skill, the
degree of which can be directly measured in terms of the copra-shell fuel requirement coefficient: if
this rises much above 0.67, say to 0.75 (leaving only 25 per cent of the shells available for
processing to charcoal), it would indicate unskilled workmanship or poor management.
To return to the second structural linkage between Activities 1 and 2: the size of this external
structural coefficient is determined by the degree of maturity of the nuts at harvest time (coconut is
usually harvested at two-month intervals). Maturity level affects the rate of recovery of copra. The
recovery or nuts-to-copra conversion rates for three maturity conditions are shown as structural
coefficients in Activity 2, but they are determined by the timing of the harvesting operation in Activity
1: the conversion rate, denoted here by C, will be 1 330 nuts required to make one 'candy' of copra
(480 pounds, 218 kilograms) if nut harvesting is at the still-green stage. For ripe nuts, C = 1 150, and
for brown nuts, C = 1 100. (Of course, other management practices such as palm fertilizer applied,
weeding etc. will also affect the absolute level of nuts-to-copra conversion, but the relative recovery
rates due to the maturity effect will still hold.)
Finally, Activity 4 consists of making coir fibre from the husks by-product generated in Activity 2.
The structural coefficient relating these two activities has a value of unity since each nut for copra
production provides one husk for making coir fibre. This is the only structural coefficient shown as
providing linkage between Activity 4 and the other activities. However, on most estates another
important linkage does in fact operate between Activities 1, 3 and 4: it consists of the common
practice of retaining some proportion of the husks generated in Activity 2 and using them (buried in
trenches alongside the palms) as a source of potash in the fertilizing operation of Activity 1. The
proportion of husks used for this purpose, which will depend on the potash status of soils on any
particular estate, will then govern coir production output in Activity 4. (The deciding factor between
using husks as fertilizer or as an input to coir production will be the cost saved by using husks instead
of artificial potash versus the income foregone by not making coir.)
In summary, the external linkage coefficients shown in Figure 5.5 quantify the interrelatedness which
exists between the various activity subsystems of this total coconut enterprise. Any full analysis of
any one of these activities would also require consideration of its effects on the others.
Both internal and external structural coefficients become very important in linear programming
(Section 9.6).

The next element of a farm-household system to be considered is its farm service matrix. This is a
system of Order Level 9 (Figure 1.2). It consists of some set of material, labour and cash inputs
which enables the farm-household to function as a system, but which is not specific to any particular
production activity of the system. (If they are activity-specific they will be treated as costs of the
appropriate activity.) This section thus continues the discussion of resources, capital and costs begun
in Section 3.3.6.
In their financial dimension these system-service inputs are referred to as whole-farm overhead,
common or farm fixed costs. Such fixed costs must be included in the economic evaluation of a
farming system for the two reasons that they influence both the profitability and the sustainability of
the system.
There are two broad categories of overhead costs, general charges and capital fixed costs. General
charges represent an actual money expenditure and as such, when included together with the
aggregated direct costs of the production activities of the system, will determine its profitability. The
second category of fixed costs (capital fixed costs, below), while not directly affecting immediate
system profitability, will affect its sustainability.
General charges are those fixed costs incurred in obtaining service inputs of an institutional nature
(see Section 3.3.6) which logically cannot be assigned to any specific production activity of the farm
and which must be paid for. They represent an actual financial outlay. Some examples are: land tax
payments, taxes on general-purpose livestock, water licence fees and vehicle registration fees.
Table 5.3 provides an example of the general charges and capital fixed costs making up the annual
fixed costs on a rubber estate. It illustrates the importance and wide range of general charges.
Capital fixed costs are those costs incurred in the use of non activity-specific physical farm capital
and pertain to its maintenance, operation (in non activity-specific uses only) and provision for its
eventual end-of-life replacement. Their payment is not mandatory - at least in the short run - and
'payment' is often in the nature of an internal bookkeeping transfer which is designed to provide - if
indeed at all - only for the long-term sustainability of the system's fixed capital and for the eventual
replacement of such capital as it wears out. Typical of these capital fixed costs are 'payments' or
provisions for the maintenance, use and replacement of farm buildings, houses, fences, pumps and
other machinery, paths, roads, farm bridges, ponds, irrigation and soil conservation systems, and
breeding livestock.
TABLE 5.3 - Example Listing of Annual Total Fixed Costs on a Rubber Estate
Cost Item Amount
General charges
Salaries Rs 18 757
Staff allowances 600
Staff provident fund 1 012
Workers provident fund 11 462
Wage adjustment 673
Labour holidays 6 454
Food for functions 708
Medical, sanitation 3 635
All insurance 2 852
Office supplies 1 771
Commissions paid 1 005
Visiting agent 750
Rents, taxes 2 231
Vehicles 10 000
Watchmen 1 373
Retirement gifts 2 478
Capital fixed costs

Housing upkeep Rs 9 264
Labour-lines upkeep 1 857
Minor buildings 390
Roads, water supply 950
Total fixed costs Rs 78 222
Total production (pounds weight) 276 600
Fixed cost/pound weight Rs 0.28

The essential difference between general charges and such capital fixed costs is that payment for the
former is mandatory, while cash disbursement for the latter is not. (The alternative, if no funds are put
aside for capital replacement and no replacement is carried out, is to run down the farm capital
structure, in which case it sooner or later ceases to be sustainable.)
On most small farms, expenditure on fixed costs of both categories will be very low. Most fixed-
capital costs (i.e., for capital maintenance) will be 'paid' for in the form of family labour used to
repair and eventually reconstruct fences, ponds, ox gear etc. Obviously, the importance of farm fixed
costs will thus vary according to the farm type under consideration (Chapter 2). The forest-garden
farms of Kerala and Sri Lanka can practically ignore capital fixed costs (but not general charges such
as land taxes), simply because apart from their trees, a few hand tools and the land itself, they have
little capital. On the other hand, on the intensively operated mixed crop-livestock farms of Sind,
capital fixed costs (and total farm fixed costs) might be significant - but still much less so than on
commercial estates. The typical situation on these latter is summarized in Table 5.4 which shows
annual average direct, fixed and total cost of production (COP) per pound of made rubber on a
sample of 148 Sri Lankan estates. Total estate fixed costs (general charges plus capital costs, but, as
noted, mainly general charges) are very significant and amount to about one third of total costs.
TABLE 5.4 - Average Annual Costs of Production of Made Rubber for a Sample of Sri Lankan
Rubber Estates (Rs per Pound Weight)
Direct costs Fixed costs Total COP

Sample
Cultural Factory Marketing
36 sheet estates 0.402 0.088 0.030 0.249 0.769
86 crepe estates 0.430 0.147 0.045 0.293 0.915
26 mixed estates 0.450 0.117 0.049 0.289 0.905
All 148 estates 0.430 0.138 0.044 0.289 0.901

Source: Data from a survey by the senior author and Marshall Perera for the Office of the Rubber Controller, Colombo. 1989.
Obtaining the general charges of a farm or estate will usually present no problems. Table 5.3 is an
adequate guide to the types of cost items which may occur. On the other hand, the evaluation of
capital fixed costs will require a little arithmetic as exemplified by the illustrative schedule of Table
5.5. This is for a particular farm whose operating budget is presented later in Table 7.1. The schedule
consists of a complete inventory listing of all the farm's fixed-capital items (including land) and their
capital values as shown respectively in columns (1) and (2). The inventory is then assessed to obtain
the annual fixed costs of replacing/repairing/operating these capital items as respectively listed in
columns (4), (5) and (6). These costs are aggregated in column (8) to give the annual fixed cost of
each capital item. Column (7) is not needed immediately; it will be required later for economic
evaluation of a farming system (Section 7.2).
In calculating annual capital fixed costs as per Table 5.5, the values given to capital items in column
(2) and hence their depreciation charge in column (4) should not be based on their initial purchase or
construction cost. Capital items should be valued at their current expected replacement cost. The
tractor, e.g., might actually have cost Rs 8 000 five years ago, but today the cost of replacing this
five-year-old machine with a new one might be Rs 10 000. Whatever the initial cost or value might
have been is of historical interest only and, except by coincidence, will no longer be of relevance in
making provision for eventually replacing the item when that becomes necessary. Column (3) contains
estimates of the years of useful service which can be expected of each item from the time of its initial
purchase by the farmer. These are inevitably somewhat arbitrary. Land is assumed to be used
sustainably and thus to have an indefinite useful life. Columns (4), (5) and (6) of Table 5.5 show the
individual components of the fixed-capital cost of each capital item, based on columns (2) and (3).
TABLE 5.5 - Farm Capital Investment Inventory and Schedule for calculating Annual Capital
Fixed Costs
Capital investment
inventory
(1) (2) (3) (4) (5) (6) (7) (8)
Item Useful Deprec'nb Repairs Operating Interest at Total (4)+

Capital item value a life costs 10% (5)+(6)
(Rs) (years) (Rs) (Rs) (Rs) (Rs) (Rs)
Land 50 000 - - - - 5 000 -
House, sheds 15 000 25 600 200 - 1 500 800
Tractor 10 000 10 1 000 400 - 1 000 1 400

Hand thresher 2 000 6 330 50 - 200 380
Cultivators 3 000 5 600 100 - 300 700
Livestock 600 3 200 50 - 60 250

gear
Barn 5 000 20 250 200 - 500 450
Fences 6 000 30 200 400 - 600 600
Dam/pond 8 000 40 200 - - 800 200
Water pump 4 000 8 500 200 800 400 1 500
Oxen 3 000 5 600 - 200 300 800
Total 106 600 4 480 1 600 1 000 10 660 7 080
a Value based on current expected replacement cost. 'Land' includes all fixed improvements (drains, levelling, earthworks etc.)
and permanent tree crops in house yard.
b Here the straight-line method of calculating depreciation is used and it is assumed that the capital
items have no residual value at the end of their useful life (see Section 5.5).
Depreciation is the decline over time in the capacity of a capital item to provide the service expected
of it and for which it is held (Barnard and Nix 1973, pp. 85-88; Doll and Orazem 1984, pp. 263-
266). Depreciation is usually measured on an annual basis. Ideally, it would be measured in actual
terms but that would involve inordinate record keeping. Instead, simpler rule-of-thumb procedures
(as outlined in Sections 5.5 and 10.10) are used. As well as through wear and tear arising from use
of the capital item, depreciation also occurs through technical obsolescence as more modem
machines or other items providing the same service but at lower unit cost become available. While
wear and tear may be relatively predictable and approximately constant over time, technical
obsolescence may not be. In consequence, to be as correct as possible, annual depreciation should be
calculated afresh each year. Whether or not it is economically worthwhile to do so will depend on the
particular situation.
Column (4) of Table 5.5 provides some examples of depreciation. These amounts are measured as a
constant pro rata proportion over each item's life of its expected replacement cost (i.e., straight-line
depreciation, Section 5.5). In percentage terms this depreciation cost corresponds to 100 divided by
the item's useful life. Consider the tractor valued at Rs 10 000 which is estimated to have a total
useful life of ten years. On a pro rata basis, the annual cost of obtaining this service is Rs 10 000/10
or Rs 1 000. Or to put it another way, based on the present assessment, if the farm's fixed capital is
not to be depleted, a sum of Rs 1 000 per year should be set aside to enable the tractor to be replaced
when it finally wears out or, due to technological advance, becomes obsolescent (in five years' time
if it is five years old now). (Hence the preference for basing item value in column (2) on current
expected replacement cost rather than past/historical actual cost.) There are several methods by
which annual depreciation cost can be calculated; they are discussed in Section 5.5 below.
Repairs refers to the sum of money which normally must be spent on maintaining each capital item in
good repair. Note, however, that in spite of such repairs, the item (tractor, thresher, fences etc.) will
sooner or later wear out or become obsolete, hence the need for its depreciation as well as its routine
repair.
Operating costs are the costs of actually operating a capital item: e.g., Rs 800 for water-pump fuel,
Rs 200 of feed and veterinary expenses for the oxen. In Table 5.5 the tractor would, of course, also
incur fuel and oil operating costs but in a typical farm situation it would be fairly easy to charge these
against the separate specific activities (paddy, cotton etc.) which actually consume this fuel or incur
such operating costs, rather than against the farm as a whole. Often, however, some capital operating
costs will be incurred in nonspecific farm work, in which case they should be included in column (6).
This was discussed in Section 3.3.6 in relation to Figure 3.3.
Interest as listed in column (7) of Table 5.5 is a bookkeeping charge levied on total farm capital -
i.e., all capital invested in the farm - in order to assess the productivity of the several farm resource
categories or factors of production (land, labour and capital) as outlined in Section 7.2.3. It is not a
'cost' in the conventional sense of detracting from the value of farm net output (or family income). For
the moment, only the procedure of obtaining this item is of concern. Briefly, some appropriate rate of
interest, here ten per cent, is applied to the assessed value of each/all of the capital investment items
as listed in column (2) of Table 5.5. The appropriate interest rate is determined by the opportunity
cost of similar capital items in their 'most likely most profitable most prudent' other use. Thus if the
best alternative investment available were for the family to sell the farm and put the proceeds in a
bank earning ten per cent annual interest, this would determine the opportunity cost of farm capital
and thus the appropriate interest rate to apply in obtaining column (7).
To conclude, one might state the obvious: these procedures for obtaining the interest costs of column
(7) are full of practical pitfalls. Is the replacement value of an item a proper representation of its
value in use? If not, a more appropriate value should be used. Or, as is not infrequently done, items
might be valued at the mid-point of their useful life with allowance made for any residual value they
may possess at the end of their useful life. Thus the oxen of Table 5.5 with a useful life of five years
and, say, a residual value of Rs 800 as meat would be valued at Rs (3 000 + 800)/2 = Rs 1 900. Is it
sufficient to apply some rate of interest (say ten per cent, as above) uniformly to all capital items, or
are the opportunity costs of some capital items greater than of others? (Probably yes, but precision
must be weighed against the extra work required in obtaining the necessary information.) What is an
appropriate alternative investment? Could money, instead of being put in the bank at ten per cent, be
lent to a speculator in the souk at 50 per cent and still be regarded as a prudent alternative
investment? (Probably not.).... Anyway, do realistic alternative uses of farm investment capital
actually exist? (Certainly yes in South India, possibly not in the hills of Bhutan.)
On the other hand, when concern is with small farms, these practical difficulties are largely removed
by the facts that (a) the value of farm capital itself, and thus the level of fixed costs associated with it,
will not be very great; and (b) most capital can be maintained and replaced by using (largely 'free')
farm-generated resources and family labour. In short, while the effort that goes into the construction of
Table 5.5 must inevitably be considerable if later analysis based on it is to be technically
satisfactory, the actual data obtained within and subsequently from Table 5.5 might not make much
practical difference in the evaluation of small-farm performance (Chapter 7).
To summarize the discussion of Table 5.5: total capital fixed costs on this farm, column (8), are the
sum of the previous cost-item columns (4), (5) and (6) - in the example Rs 7 080 annually. Total farm
fixed costs are the sum of this item and the farm's general charges.
It was noted previously that payment of general charges is usually mandatory. Taxes and licence fees
must be paid. It was also noted that payment of capital fixed costs might be considered optional, at
least in the short run. In fact it would be more correct to say that some of the components of capital
fixed costs (Table 5.5) are mandatory while others are optional. Again referring to Table 5.5,
payments for running repairs and operation (e.g., fuel, feed for the oxen) are in a large sense
mandatory, while provision for depreciation is largely optional (in any one operating year).
Depreciation (column 4) might be provided for in practice by setting aside Rs 4 480 per year; or such
an amount might be re-invested in the business to generate the necessary future replacement funds; or
no provision at all might be made. This third case implies that: (a) farm capital stock is being run
down or depleted; (b) to the extent that such depletion is occurring, real farm net income is being
overestimated; and (c) sooner or later the farm will cease to be a sustainable system (Norman and
Douglas 1994. Ch. 2).

Ideally, the annual depreciation of a capital item should be measured in terms of its actual wear and
tear due to usage and the change in its degree of obsolescence (Barnard and Nix 1973, pp. 85-88;
Doll and Orazem 1984, pp. 260-267). This would imply much inconvenient record keeping not worth
its cost. Instead, it is assumed that usage of the item and the accompanying flow of service from it
follow some regular pattern and depreciation is estimated accordingly (Thuesen 1959, Ch. 5). Thus
there are four main methods of calculating annual depreciation 'cost' or the annual rate at which a
capital item provides services and is thereby used up or becomes obsolete. Three of these methods
are briefly discussed here. (The fourth, which involves the application of amortization procedures, is
outlined in Section 10.10.)
Straight-line depreciation assumes a constant flow of service with consequent wear and tear and/or
obsolescence during the item's useful life so that the value of the item will therefore decrease by a
constant uniform annual amount over its life (Makeham and Malcolm 1986, Ch. 13). This is the
easiest method to apply. The formula for obtaining the annual straight-line depreciation amount,
denoted by D, is
D = (PV - SV)/L
where PV is the item's present value (i.e., its expected future replacement cost at the moment of
analysis), SV is its expected salvage or residual value at the end of its useful life and L is its expected
total years of life. Thus annual depreciation on a cultivator with an expected replacement cost of Rs 1
000 and which has an expected total useful life of ten years and can then be sold for Rs 200 as scrap
would be
Rs (1 000 - 200)/10 = Rs 80;
or if it has no salvage value,

Rs 1 000/10 = Rs 100.
This method is probably most appropriate for small Asian farms where most capital items (pumps,
cultivators, livestock gear etc.) suffer no loss in value due to obsolescence, are entirely used up in the
production process, and do not have a residual value (e.g., as the basis for trade-in on a new model).
The results of applying this method are shown in the lefthand-side graph of Figure 5.6.
FIGURE 5.6 - Example of Straight-line, Declining-balance and Sum-of-integers Depreciation
Methods applied to an Item with an Initial Value of Rs 1 000, a Ten-year Life and Zero Salvage
Value
Declining-balance depreciation assumes that the item is used up and/or becomes obsolescent at a
constant percentage rate of its annual starting value. Applied to the previous example with a zero
salvage value and assuming an annual depreciation rate of ten per cent, this would result in an annual
depreciation amount of Rs 1 000 x 0.10 or Rs 100 in year 1 of the item's life, Rs (1 000 - 100)0.10 or
Rs 90 in year 2, Rs (1 000 - 100 - 90)0.10 or Rs 81 in year 3, Rs 73 in year 4, etc. With declining-
balance depreciation, the formula for calculating the annual depreciation 'cost' in year t, denoted by
Dt is:
Dt = (V t-1)d
where Vt-1 is the item's depreciated value at the end of year t-1, Vt-1 = (Vt-2 - DT - 1), V0 = (PV0 - SV)
where PV0 is the item's expected replacement cost at time zero, i.e., at the start of the depreciation
analysis, and d is the constant annual rate of depreciation.
Obviously, as illustrated by the middle graph of Figure 5.6, the procedure results in a lower and
declining depreciation loss in later years. This is appropriate where the value loss is due to the
'planned obsolescence' of modem equipment (which is reflected in dealers' secondhand machinery
and vehicle-buying price schedules as these are used in the West). Such obsolescence hardly exists
on small Asian farms where, as noted, value loss is due to wear and tear from physical use. Note also
from Figure 5.6 that, unlike straight-line depreciation, the declining-balance method does not lead to
full write-down of an item at the end of its useful life.
Sum-of-integers depreciation is essentially similar to the declining-balance approach in that this
method also results in less than full write-down and in a declining annual depreciation 'cost' which
eventually becomes zero in the last year of life as shown by the righthand-side graph of Figure 5.6.
Compared to the declining-balance and straight-line methods, the sum-of-integers method implies a
more rapid initial depreciation but slower later depreciation. The formula for sum-of-integers
depreciation is:
Dt = (PV t=0 - SV) [Remaining years of life/Sum of integers of total life]
where Dt is again depreciation for year t and PVt=0 is the expected replacement value of the capital
item at time zero, i.e., at the start of the depreciation analysis. Using the previous example of the
cultivator with a life of ten years and no salvage value (i.e., SV = 0), the 'sum of the integers' would
be (1 + 2 + 3..... + 10) or 55, and the annual depreciation 'cost' would be:
Year 1: Rs 1 000 x 9/55 = Rs 164
Year 2: Rs 1 000 x 8/55 = Rs 145
Year 3: Rs 1 000 x 7/55 = Rs 127
... ...
Year 9: Rs 1 000 x 1/55 = Rs 18
Year 10: Rs 1 000 x 0/55 = Rs 0.
Comparing the three depreciation methods outlined above, the straight-line approach is generally to
be preferred on the grounds that it is less complicated and ensures full depreciation or write-down of
the item.
5.6 REFERENCES
Barnard, C.S. and J.S. Nix (1973). Farm Management and Control, Cambridge University Press.
Dillon J.L. and J.B. Hardaker (1993). Farm Management Research for Small Farmer
Doll, J.P. and F. Orazem (1984). Production Economics: Theory with Applications, 2nd edn,
Wiley, New York.
Grimwood, B.E. (1975). Coconut Palm Products: Their Processing in Developing Countries, FAO
Agricultural Development Paper No. 99, Food and Agriculture Organization of the United Nations,
Rome.
Norman, D. and M. Douglas (1994). Farming Systems Development and Soil Conservation, FAO
Farm Systems Management Series No. 7, Food and Agriculture Organization of the United Nations,
Rome.
Thuesen, H.G. (1959). Engineering Economy, 2nd edn, Prentice-Hall, Englewood Cliffs.
6. HOUSEHOLD GOALS, FARM
PLANNING OBJECTIVES, SYSTEM
PLANNING AND PERFORMANCE
CRITERIA
6.3 REFERENCES
'There is always a reason why farming is carried out in one way rather than another.'
Hans Ruthenberg (1976)
Subsequent chapters are concerned with the evaluation of farm-system performance and with the
planning of new or improved systems. Backward-looking or ex post evaluation requires standards
against which past performance can be measured. Forward-looking or ex ante farm analysis, i.e.,
planning, requires specification of system objectives as a basis for selecting the most appropriate
subsystems (enterprises, activities, processes and technologies).
For farms of Types 1, 2, 3 and 4 (i.e., for all except large commercial farms and estates), standards
and objectives arise in the system's household component and are largely determined by family values
and goals. In most cases these can be accepted as given. The main task is to identify what the most
desirable attributes or properties of such a farm system are likely to be, from the farm family's
viewpoint, and to develop criteria for measuring the degree to which a given or proposed system
possesses these properties, or lacks them (Harwood 1979, Parts 1 and 2). It is also necessary to have
criteria by which the lower Order Level subsystems - specifically the separate production activities
of Order Levels 3, 4 and 6 (resource generation, crops and livestock respectively) - can be measured
in terms of their contributions to the achievement of whole-farm objectives, and thus household goals.

There are basically two major farm-operating objectives, profit maximization on market-oriented
farms and household sustenance on subsistence-oriented farms (Collinson 1983, Ch. 2; Makeham and
Malcolm 1986, Ch. 3). By profit maximization is meant maximization of net gain measured as total
benefit less total cost. As discussed in Section 6.2.2, profit is usually but not necessarily measured in
money terms.
Profit maximization measured in money terms can generally be taken as the planning objective on
farms of Type 5 (large commercial family farms) and Type 6 (estates) but this is increasingly
constrained by external factors such as labour laws, health and safety regulations, and national
policies to produce crops which will generate foreign exchange or serve as a basis for local
industrialization. Internal constraints can also exist on such farms and take the form of management
jealousy in protecting the 'mark' of their product even when production of lower quality produce
might yield more profit, and spending more than the necessary amount of money on estate upkeep to
maintain estate appearance and status. Profit maximization measured in money terms can also be the
primary objective of some subtypes of Type 3 (small independent specialized) and Type 4 (small
dependent specialized) farms.
Note that, strictly speaking, when uncertainty is present (as is usually the case), profit maximization is
not a feasible objective. Under uncertainty, rather than a variety of sure profit options, the farmer
faces a set of profit probability distributions corresponding one-to-one to the available decision
options. For each of these risky choices with its corresponding probability distribution of profit, it
can be argued that the farmer will have some equivalent sure profit or certainty equivalent such that
he or she would be indifferent between (a) taking the risky option with its uncertain profit or (b)
receiving the sure profit amount. The operating objective is thus to discover and implement that
option which has the highest certainty equivalent. Thus, under uncertainty, profit maximization
translates to certainty equivalent maximization. These matters are elaborated in Chapter 11.
Household sustenance through the production of food and fibre for household consumption provides
the primary objective of all other (predominantly Type 1 and 2) farms but, as noted previously, it is
increasingly modified by the need also to generate some level of cash income.
Only some Type 6 farms (estates) and some Type 1 (subsistence) farms are found, respectively, at the
extremes of aiming only for profit maximization or full subsistence; most farms of these and other
types are located somewhere between these extremes along the profit-sustenance continuum as shown
schematically in Figure 6.1.
FIGURE 6.1 - Stylized Representation of Relative Importance of Financial Profit and
Subsistence by Farm Type
Beyond the priority objectives of ensuring sufficient food and cash for the farm household, Type 1
(subsistence) and Type 2 (semi-subsistence) farmers generally have a number of secondary
objectives. These are likely to include such things as having security in their livelihood, having the
opportunity to observe socio-cultural customs and obligations, and having a satisfactory amount of
leisure time (Clayton 1983, Ch. 4).
6.2.1 Productivity
6.2.2 Profitability
6.2.3 Stability
6.2.4 Diversity
6.2.5 Flexibility
6.2.9 Summary
Given the two extremes in planning objectives and the relative weight which will be accorded to
each for the various farm types (Figure 6.1), there are eight main properties a system might possess,
thus requiring a set of eight criteria by which these system properties may be assessed. Each of these
properties and its associated assessment criterion apply both to whole-farm Order Level 10 systems
and to constituent activity/enterprise subsystems of Order Levels 3, 4 and 6. They are relevant to
evaluation and planning both at farm-household level and at a broader social level. The eight
properties of farm systems and activities which need to be assessed are:
(1) Productivity
(2) Profitability
(3) Stability
(4) Diversity
(5) Flexibility
(6) Time-dispersion
(7) Sustainability
(8) Complementarity and environmental compatibility.
At farm level
These system properties can be quantified (at least conceptually; see below). They are all 'desirable'
or at least neutral in the sense that an individual farm which ranks highly with respect to productivity,
profitability, stability etc. represents a superior system to a farm on which productivity, profitability
etc. are low. When applied to a specific farm only a few of these properties might be thought relevant
by the farm family or other decision maker. One farm manager might have a purely profit objective
and would wish his or her performance to be evaluated in terms of profit. A second manager on the
same type of farm might seek some balance of profitability-stability-sustainability so that criteria
relating to these three properties would then need to be applied.1 In both cases, however, the
analyst/adviser would be wise to review the farm system relative to all eight criteria and to proffer
relevant advice.
1 It would be an error to evaluate both farms according to a criterion which is relevant to only the first farm. However, at a more
general level, this is what often happens when Asian farms are evaluated by Western-oriented economists as being 'inefficient',
'backward' and of low productivity etc. on the grounds that they perform poorly according to the conventional Western criterion of
money profitability.
In following sections the above eight properties and their associated performance criteria are
examined in relation to their use at farm level in Field A (i.e., on-farm problem solving and advice) -
which is not to say that they are not relevant to farm management analysis in Fields B, C and D;
indeed, particularly in Fields C (industry and sector-level analysis) and D (policy-making), they will
often be highly relevant.
At social level
Apart from farmers, society has a vital long-term interest in how rural resources are used and how
farm systems perform. When farm management operates in Fields C and D (i.e., playing a contributing
role in planning new settlements, irrigation projects etc., and in government policy guidance), it must
also be cognizant of the several properties of farm systems. Some of these, particularly productivity,
stability and sustainability, might well be more important from a social than from a private-household
viewpoint. To plan new farms which will be profitable is one thing; to plan profitable farms which
also make optimal sustainable use of what are finally social resources is yet another. Even more
difficult and increasingly important is to develop new or restructured farm systems which have all
these desirable properties and, in addition, are compatible with the social environment and not
destructive of the physical environment. Also, as evidenced by increasing interest in gender analysis,
equity within farm-household systems may also be important from a societal view.
6.2.1 Productivity
Productivity is primarily a measure of the relative suitability of a system or activity in a particular
agro-ecological environment. On commercial farms it is an indicator of relative efficiency of
resource use and management performance. It is an underlying condition for profitability but should
not necessarily be taken as a desirable attribute or objective in itself. On non-commercial farms,
productivity is a necessary condition for achieving family sustainability - but only to a limit.
Production beyond what a family can consume or store or barter becomes irrational and may even be
undesirable. At some places in the Himalayan hills (such as eastern Bhutan) up to 30 per cent of the
maize crop is surplus to food requirements and, in the absence of programs to develop alternative
uses, is converted to alcohol with the most unfortunate social consequences. Yet in such areas it is not
uncommon to find programs to 'improve' maize production as a road to farm 'development'. A more
rational approach would be based on crop development within the context of the farm and village
socioeconomic system. This might well aim at reduction in some crop production and reformulation
of the whole farming system.
Productivity is conventionally measured in terms of such units, e.g., as tons, kilograms or litres of
output respectively per acre, hectare or animal unit employed over some relevant time unit (typically
a year). Or, if desired, it may be measured in financial terms over some relevant timespan as the ratio
of total revenue to total cost, i.e., the value of output per unit of cost. Productivity is an appropriate
measure of system and activity performance when applied to single-output enterprises or mono-
product systems. However, as one moves from commercial to increasingly sustenance-oriented farm
types (Figure 6.1), there is a tendency for by-products to become more numerous and important. This
introduces difficulties in the measurement of output and productivity. In a commercial situation,
productivity can usually be measured as a single variable. However, in a quasi-subsistence situation,
it might require the construction of a table of output items such as exemplified by Table 6.1. This
shows the main and secondary products of a range of tree and vine crops commonly grown on forest-
garden farms in Sri Lanka and Java and the relative importance of the products of each crop as
assessed by the household head (which assessment, of course, would vary among individual
households).
Product mixes from some single crops can be considerably more complex than Table 6.1 suggests,
e.g., as with rubber. On an estate there will be no measurement difficulties: output is simply the
number of kilograms of made rubber per hectare. In contrast, on a mixed smallholding, 'output' might
consist of latex for sale, seed from the annual seedfall, shade for associated crops such as cacao and
yams, live supports for such climbers as pepper, and prunings or branchfall for household fuel. Some
of these products can be measured (latex, seed, fuel), some cannot (the shade effect and the
contribution of rubber trees to the micro-environment for use by associated crops). If the separate
outputs and inputs can be measured, the crop can be evaluated as a set of (related) activities; if they
cannot, it will need to be evaluated as a composite enterprise (Section 4.2.2).
TABLE 6.1 - Relative Importance of Main and Secondary Products of Some Crops on a Sample
of Forest-garden Farms
Crop Main product Secondary products
Pepper pepper 1.00
Rubber latex 0.78 timber 0.09 fuel 0.10 seed 0.03
Tea leaf 0.85 fuel 0.15
Coconut copra 0.78 husks 0.15 charcoal 0.06 leaf 0.01
Cloves buds 0.74 stems 0.25 dust 0.01
Nutmeg nutmeg 0.70 mace 0.30
Jackfruit flesh 0.75 seeds 0.20 fuel 0.05

Source: Data from a 1978 survey by the senior author.
6.2.2 Profitability
Financial profitability of activities/enterprises is discussed in Chapter 4 and of the whole-farm
system in Chapter 7. As shown in Figure 6.1, financial profitability becomes a less important
performance criterion as analysis moves towards the subsistence end of the farm-type continuum. In
particular, money profit or gross margin as a measure of performance of activities is typically both
not possible and largely irrelevant for Type 1 (i.e., subsistence) farms and often, to a significant
degree, for Type 2 (i.e., semi-subsistence) farms because of their lack of market interaction.
Financial profit as a criterion for measuring the performance of farm-household systems is often
unreliable. This is because, on small farms, money profit is often generated at the expense of
weakening or distorting the system through such factors as increasing household exposure to debt for
purchased farm inputs, the danger of fostering an exploitative and non-sustainable rate of resource use
(causing soil degradation), reduction in the level of reliability of household food supply and
increasing risk. Nevertheless, many farm-economic surveys undertaken to measure the performance of
small farms continue to gauge this solely in terms of financial profit, even when these farms might
have quite different objectives (as discussed below).
Profit is normally measured in money terms as gross financial revenue minus total financial cost per
period. Note, however, that it may - if need be - also be assessed subjectively in qualitative terms as
net gain, i.e., as total benefit less total cost however measured. Such an approach might be used in
assessing the performance of Type 1 (subsistence) farms having no significant market interaction,
leading to qualitative assessment of a Type 1 system as, e.g., profitable or not profitable.
Associated with profitability, however measured, is the matter of farm-size adequacy. Clearly, a
prime requirement of any whole-farm system is that it be of sufficient size to satisfy the farm-based
needs of its primary beneficiaries. Small farms should thus be assessed in terms of income adequacy,
i.e., their ability to sustain the farm household's need for income in cash and/or kind without causing
resource or environmental degradation. Income adequacy is thus an important aspect of profitability.
6.2.3 Stability
System stability refers to the absence or minimization of year-to-year fluctuations in either production
or value of output. (The latter also implies either stability in input costs, yields and prices or
counterbalancing movements in these influences on value of output.) Where conditions are
favourable, price and production instability can often be countered by more careful activity selection
(e.g., of drought-tolerant varieties, pest-immune crops); by diversification of activities; by seeking
greater flexibility in product use or disposal; by multiple cropping over both space and time; and by
increasing on-farm storage capacity and post-harvest handling efficiency.
In some situations the most direct strategy for stabilization is simply to increase production/income to
a level which allows an annual surplus to be retained/invested in good years to cover deficiencies in
poor years. This is generally possible on farms of Type 5 (large commercial family farms) and Type
6 (estates). (The classical tea/rubber/oilpalm/coconut estate systems are generally production-stable
but price-unstable.) Many variations of such a strategy are possible. Around Ponorogo in the Madium
Valley of Java a common practice among farmers growing sugarcane, paddy and palawija crops (i.e.,
food crops other than rice) is to invest the proceeds of the (relatively stable) sugar crop in gold, then
later sell this to finance the following (relatively unstable) subsistence food crops. In the hills of
eastern Bhutan where mono-crop maize is the stable food and monsoon rains are erratic, the common
stabilizing strategy is to plant 30 to 40 per cent more maize than will actually be required if the
season turns out well which - since the crop requires no cash inputs but only family labour and oxen -
is an insurance premium willingly paid. (But the social costs of this were noted in Section 6.2.1
above.)
The magnitude of year-to-year variation in yield varies widely among crops and locations. The data
presented in Table 6.2 for two common Sri Lankan tree crops, clove and coconut, illustrate the
magnitude of possible yield fluctuations and the importance of the production (and income) instability
problem which faces the smallholder clove growers in comparison with the growers of such stable
crops as coconut. While cloves are often a very profitable crop, this would be partly offset by their
high level of price and income instability. As would be expected, this usually leads small farms to
combine cloves with other lower value/lower risk crops in order to achieve greater stability in the
farm system as a whole.
Measuring stability/instability
Price/yield/income stability is most conveniently measured in terms of the coefficient of variation,
denoted by CV, which expresses the standard deviation, denoted by SD, or positive square root of the
variance (V) of a sample of observations on a variable X as a percentage of the sample's mean value
. Thus
where n is the number of observations, Xi is the i-th observation and S denotes the sum of the
following values for i from 1 to n. The set of observations X1, X2...Xn may come from a sample
generated across time or space or both. Thus the lower section of Table 6.2 gives an annual time-
series set of observations on copra yield per acre on a particular estate in Sri Lanka for the 13 years
1960 to 1972. In Table 6.3 the CV of this sample of copra yields is calculated as 11.5 per cent. The
upper section of Table 6.2 gives a set of data on clove yield which is both of a time-series (years
1967 to 1972) and spatial (cross-section for five farms) nature.
TABLE 6.2 - Year-to-year Variation in Clove and Coconut Yield
Clove: Relative annual yield over time on five farms, Sri Lanka (1972 base)
Farm no: (1) (2) (3) (4) (5)
1972 100 100 100 100 100
1971 0 2 0 40 600
1970 70 457 5 13 0
1969 93 71 30 100 0
1968 70 - - - 75
1967 0 - - - -
Coconut: Annual copra yield from astand of mature trees, Sri Lanka (piculs per acre)
1960 14.0 1967 11.6
1961 12.4 1968 12.5
1962 10.8 1969 10.0
1963 10.5 1970 12.6
1964 9.5 1971 11.4
1965 9.9 1972 11.8
1966 10.5
Because CV is a pure number, it can be used to compare the relative stability of different
activities/systems. For example, the CV of clove yield based on the 23 observations of Table 6.2 is
157 per cent, implying that income on a clove-only farm - due to the yield effect and ignoring possible
price effects and differences in flexibility of product use after harvest - is some 13 times less stable
than income from coconuts. Likewise, comparison of the CV values for the clove-yield sample data of
farms (1) and (5) of Table 6.2 indicates yield is twice as unstable on farm (5) with a CV of 163 per
cent as on farm (1) with a CV of 80 per cent.
Of course, a stable system or activity is not necessarily superior to an unstable one. Depending on
relative costs/prices, an unstable activity may still be preferable to a stable one on grounds of long-
run relative profit. But, other things being equal, stability will usually be chosen over instability,
especially in subsistence situations where the goal is food rather than money, and where a high CV for
yield might be synonymous with recurring famine.
In addition to the inclusion of system-stabilizing activities in the farm system, there are other means of
reducing system instability. These relate to the diversification of activities and their products; to
achieving flexibility in the post-harvest use/disposal of products; and to the time-pattern of income
receival. These approaches to mitigating instability are discussed below.
TABLE 6.3 - Calculation of the Coefficient of Variation (CV) for the Copra Yield Data of Table
6.2
Copra Yield Deviation Squared deviation

Year
(Xi) (Xi- ) (Xi- )2
1960 14.0 2.65 7.02
1961 12.4 1.05 1.10
1962 10.8 -0.55 0.30
1963 10.5 -0.88 0.72
1964 9.5 -1.85 3.42
1965 9.9 -1.45 2.10
1966 10.5 -0.85 0.72
1967 11.6 0.25 0.06
1968 12.5 1.15 1.32
1969 10.0 -1.35 1.82
1970 12.6 1.25 1.56
1971 11.4 0.05 0.01
1972 11.8 0.45 0.21
Mean
Standard deviation
Coefficient of variation
6.2.4 Diversity
Diversity corresponds to 'not having all one's eggs in a single basket.' It refers to a strategy of
increasing the number of activities in a system and/or their separate products in order (i) to reduce
overall system risk of income or family-sustenance failure and/or (ii) to increase overall
production/profit (averaged over time) through a better use of available resources. A high diversity
level is conducive to system stability (but diversity might conceivably be achieved at the cost of a
reduction in average profit).
Activity diversity
As noted in Chapter 2, in terms of activities the most diversified farms are the small subsistence and
semi-subsistence farms of Types 1 and 2, respectively - e.g., the irrigated crop-livestock-orchard
farms of North India and Pakistan, the clover-wheat-barley-sheep-rabbit-poultry-scorpion-vegetable
farms of China's Loess Plateau, the small mixed crop-livestock farms of South China and Taiwan, and
the forest-garden farms of the wet tropics. The possibilities for diversification are relatively limited
on Type 3 farms, i.e., small specialist farms growing a single traditional crop such as the paddy farms
of Java. Diversification is not a strategy generally available on Type 4 farms, i.e., those growing an
industrial crop under conditions dictated by a landlord or factory. While the possibility often exists
(e.g., growing food/cash crops within sugarcane rows or in the intervals between cane plantings, as in
Mauritius), it is negated by such practical factors as lack of equipment suitable for other than the main
crop; or lack of markets; or simply by opposition on the part of landlords or sugar mills to
developments which would diminish their power over the tenants.
The three elements contributing to the overall diversity of a farm system are: (i) the number of
tree/crop/animal species present; (ii) the number of their respective products; and (iii) the number of
ways in which these products can be used or disposed of (i.e., the degree of flexibility they provide
as discussed in Section 6.2.5 below). These three elements of diversity exist in both physical and
economic (value) dimensions, either or both of which might be relevant to a particular analysis of
farm-system diversity.
If diversity comparisons are to be made within or between groups of farms, it may sometimes be
sufficient to express diversity level as simply the number of species of trees, crops and livestock
present. Generally, however, this would be a poor measure: the species (and their associated
activities) need somehow to be weighted according to their relative importance, e.g., in terms of the
number of individuals within each species, or of the areas occupied by the various crops, or the
amounts or values of outputs from the various activities. One relatively simple measure suited to such
assessment is Simpson's diversity index (Simpson 1949; Kumar, George and Chinnamani 1994). This
is defined as
where S is the number of species or activities that are present; ni (for i = 1 to S) is the number of
individuals in the i-th species, or area devoted to the i-th species or activity, or income or value of
the i-th species or activity; and N (=S ni) is the total population of all individuals, or total area across
all activities, or total farm income or value across all species or activities. For a farm system with no
diversity (i.e., having only a single species or activity so that S = 1 and n1 = N), DI is zero. As farm
diversity increases, DI approaches unity, e.g., for a farm with 20 crops each occupying three units of
the total farm area of 60 units, DI = 0.95.
Calculating DI values for a small South East Asian mixed-farm system is exemplified by the
worksheet of Table 6.4. DI values are calculated (A) on a species/activities or physical diversity
basis and (B) on an income basis. In its physical or structural dimension the farm system of Table 6.4
is dominated by tree crops, particularly coffee. Note that difficulties arise in the calculation of DI on
a species basis for those crops and livestock activities for which the individuals cannot be
enumerated, e.g., as with rice and other field crops or with pond fish. DI might then best be
calculated, from a physical perspective, on the basis of the area devoted to each species or its
associated activity.
Product diversity
This refers to the number of separate final products of a system or activity. Some of the main outputs
from a range of tree/vine crops were listed in Table 6.1. These can be disposed of in a range of ways
which also represent an avenue to diversification (see Section 6.2.6 below). As shown in Figure 6.2,
a single simple crop such as maize in Java can be significantly product-diversified by the way in
which it is managed.
TABLE 6.4 - Worksheet Calculation of Simpson's Diversity Index for a South East Asian
Mixed-farm System in Terms of (A) Species and (B) Income
Farm structure (A) Species or physical diversity (B) Economic or value diversity
Species/activity No. of individuals (ni) (ni/N)2 (S = 9) Annual income ($ni) (ni/N)2 (S = 11)
1. Areca 50 0.0151 50 0.0011
2. Jackfruit 25 0.0038 70 0.0022
3. Coffee 200 0.2415 50 0.0011
4. Pepper 20 0.0024 50 0.0011
5. Coconut 50 0.0151 50 0.0011
6. Banana 30 0.0054 50 0.0011
7. Cloves 15 0.0014 100 0.0046
8. Papaya 15 0.0014 40 0.0007
9. Vegetables ? ? 70 0.0022
10. Cows 2 0.0000 850 0.3298
11. Fishpond ? ? 100 0.0046
Sum N = 407 0.2861 N = $1480 0.3496
0.7139 0.6504
Considering the range of crops grown on many small farms (e.g., up to six or more on Sind farms, up
to 25 or 30 or more on forest-garden farms), the possibilities of integrating different classes of
livestock with these crops, and the total number of crop or livestock products which can be
generated, it is apparent that diversification can reach very high levels. Small farms producing 40 or
50 or more final outputs are not uncommon.
In general, large commercial family farms and commercial estates (farm Types 5 and 6, respectively)
remain undiversified in spite of the many opportunities which theoretically exist. Partly this is
because it is felt that diversification would divert management effort away from the traditional main
commercial crop. It is probably also due to an overly technical orientation in research to date, e.g.,
cattle grazed among young rubber trees (failed because of neglect of cattle management practices to
prevent damage to the trees); cacao-under-rubber (failed because of crop competition for labour and
unwillingness to sacrifice some rubber population); food and grain crops among coconut and rubber
trees (agronomically successful but has not reached its potential because of lack of a marketing
element in the research program).
FIGURE 6.2 - Product Diversification from a Maize Crop on a Javanese Farm
Income diversity
As exemplified in Table 6.4, Simpson's DI can also be calculated relative to income. The calculated
DI values of 0.7139 for species and 0.6504 for income indicate the farm is more diversified in
physical terms than it is in economic terms. The (ni/N)2 values also indicate that while the farm's
physical structure is dominated by coffee, its economic structure is dominated by the herd of two
cows. Another convenient measure of income diversity is given by the income diversity ratio
where Ri (i = 1 to n) is the income from the i-th activity. Note that 1 £ R £ n for Ri³ 0; and the larger
the value of R, the higher the degree of income diversity. For the farm of Table 6.4, Ri = 2.86. In
contrast, if the 11 enterprises of this farm had contributed equally to total income (i.e., Ri = $134.55),
then the level of the income diversity index R would have been 11.
6.2.5 Flexibility
The property of flexibility of product use provides a second dimension to diversification: it refers to
the availability of alternative ways of product disposal. There are a maximum of four ways:
consume/use, sell/barter, store or process. A product for which all of these possibilities exist is
intuitively preferable, other things equal, to one which can only be eaten or must be immediately sold.
Further, the quality of processability permits repetition of the consume-sell-store-process alternatives
at second, third or higher degree, but very few agricultural products are in fact farm-processed
beyond a second-degree stage. Thus, e.g., the sap of coconut/palmyrah/nipah/kital palm is drawn off
and farm-processed to make palm sugar or fermented to be drunk as toddy and, less frequently, toddy
is further distilled to arrack, but hardly ever is arrack processed beyond this second stage. This also
applies to such animal products as milk/butter, skins/leather and wool/cloth.
Farms of Type 4 (small dependent specialized family farms growing a cash crop such as cotton,
tobacco, commercial sugarcane etc.) have least flexibility in product use since they have no
alternative other than sale. Small subsistence and semi-subsistence farms of Types 1 and 2 usually
have the highest overall system flexibility because of the type and number of items produced.
Flexibility is well illustrated by the range of ways by which jackfruit are commonly disposed of on a
Kandy farm. The family will consume some (as the carbohydrate staple in place of bread or rice),
sell some for cash, barter some in the village for a chicken, then clean and dice the remainder to
smoke-cure and store for use over the off-season. Further, they will probably extract the seeds and
consume these; or sun-dry and barter them for some other food item; or water-store them (for up to
eight or nine months) for eventual consumption or sale or barter. Even greater flexibility is possible
in the disposal of the many products of the coconut palm (Grimwood 1975).
Time-dispersion of production or income refers to the degree to which a given production or income
pattern is predictably dispersed (or, conversely, concentrated) over time - over a season or, more
usually, the operating year. It is a measure of the uniformity of within-year production/income flow.
(Production, price or income stability, discussed in Section 6.2.3 above, refers to the riskiness or
unpredictability of these variables between years or locations.) Time-dispersion is a basis for
distinguishing systems from which the product or income is received as a lump amount at one point in
the operating year (e.g., in a single harvest month) from systems which yield a uniform flow over the
operating period. The two extremes are (a) a product/income which is perfectly dispersed (e.g.,
received as 12 equal monthly amounts over the operating year each equivalent to 8.3 per cent of the
annual total amount) and (b) a product/income which is all received as a single quantity in only one
month of the year.
Table 6.5 shows the monthly time-dispersion of total annual production for a sample of tree and vine
crops at selected locations in Malaysia and Sri Lanka. As these data show, such crops as tea and
rubber are highly time-dispersed; others, e.g., kapok, have far more time-concentrated patterns of
production. The table could obviously be extended to include annual or short-term crops and
livestock products.
When grown at any particular location, all crops fall into one of three categories in terms of time-
dispersion of their products, viz.:
(i) naturally time-dispersed crops (e.g., rubber, tea, cacao, cinnamon)
(ii) naturally time-concentrated crops (e.g. most fruit, vegetables, field crops)
(iii) crops which are time-dispersed by management (e.g., relay-planted, stored-in-ground).
On small farms a high level of time-dispersion (or low level of time-concentration) of

production/income is usually desirable for the following four reasons:
(1) Regularity/reliability of food supply: The important dimension in production of perishable food items (fruits, vegetables,
animal products) on family-sustenance farms and where storage is not possible is regular and reliable availability rather than the
amount of total annual product. A sufficient quantity of jackfruit, milk, eggs etc. available in each month of the year is superior to
a larger volume of produce occurring in only one month of the year.2
(2) Avoidance of storage costs and losses: Although simple storage methods are generally available
(e.g., rodent-proof granaries, anti-weevil grain treatment, smoke-curing or drying), they have not been
adopted by many farmers and post-harvest losses continue to range from 20 to 40 per cent. Lack of
household supervision of stored produce continues as a major cause of food loss, even when the cost
of safe storage is within the reach of the poorest families. Thus, where it can be achieved, a time-
pattern of food production which avoids or minimizes the necessity for storage is an alternative which
is superior to storage as a means of ensuring food availability.
(3) Minimization of family debt: Near-perpetual indebtedness of small-farm families to landlords,
moneylenders and/or shopkeepers is a serious problem throughout much of Asia. Debt is incurred for
two main purposes: as credit for family food and material requirements during lean periods, and to
meet socio-cultural obligations (weddings, deaths, festivals). Farm systems which yield income only
once a year (e.g., tenant-operated sugar farms) are almost sure to enter into indebtedness - all too
often under onerous conditions - as a normal part of their existence. Other things being equal, the need
for such indebtedness is minimized if the farm system can be so structured as to generate a uniform
flow of food and cash income throughout the year.
(4) Technical and economic efficiency: The relatively high efficiency of farm Types 5 and 6 (i.e.,
large commercial family farms and estates) is due largely to their organization along industrial lines.
Processing is an integral part of their operations and efficient processing requires continuous-flow
rather than batch-type operation. To achieve this, estates deal with crops (e.g., tea, rubber, cocoa,
coconut) which naturally give a highly time-dispersed flow of production over the year.
Alternatively, they produce crops which, although each production unit (e.g., hectare) may be time-
concentrated, can be sequentially harvested to give a uniform flow of operations and product over the
year (e.g., cassava, sisal).
2 The annual time-pattern of production (as opposed to total annual production) has been a largely
neglected aspect of agricultural research in relation to small-farm development.
TABLE 6.5 - Monthly Distribution of Production of Some Tropical Tree and Vine Crops at
Selected Locations in Malaysia (M) and Sri Lanka (SL)
Source: Data from 1978 surveys by the senior author.
These considerations of estate structure indicate corresponding weaknesses on many small farms,
e.g., dryland crop farms operated according to seasonal conditions. While such small farms must have
a basic set of farm equipment, this equipment might not be used for more than two or three months of
the year. (Such excess capital cost will not be great in absolute terms since most equipment will be
farm-made, but nevertheless the highly time-concentrated production pattern will require a capital
stock some three or four times greater than would be the case if the production pattern was time-
dispersed.)
A related weakness consists of the time-bunching in the demand for farm labour relative to the
supply-flow of family labour. Total labour supply on a small farm on an aggregated annual basis
might commonly exceed annual requirements by 50, 100, 200... per cent, but labour shortage at
critical periods (planting, weeding, harvesting) might still impose a major production constraint. This
is commonly the case on dryland hill farms operated without draught animals, especially where there
is only a short wet-period for land preparation and planting. But the problem is also serious across
the great monsoon paddy lands where little can be done in the fields before the irrigation channels
begin to flow.
The Western answer to similar problems with highly time-concentrated farm systems - almost an
automatic reflex - has been farm mechanization. But other counter-strategies are possible on Asian
farms: cooperative work-sharing (such as the 'bawon' harvest system of Indonesia); changing the
structure of the system to include crops less sensitive to harvest or planting date; or by making more
fundamental structural changes to the system by which labour-intensive field crops are at least partly
replaced by near zero-labour food tree crops (such as jackfruit, breadfruit and coconut).
Measurement of production/income time-dispersion
There is no generally recognized measure of time-dispersion of production or income but a useful
index of relative dispersion can be constructed on the basis of the dispersion of individual monthly
values of production or income relative to their annual totals. Examples are offered below.
Time-concentration of single activities
Table 6.6 shows monthly relative production (or income) for four crops - kapok, pepper, rubber and
single-crop rice - on a Sri Lankan farm. For each crop, the coefficient of variation (CV) of its monthly
production (obtained in the usual way as per Section 6.2.3) is also shown. As shown for single-crop
rice, annual production (or income) of any crop which occurs wholly within a single month represents
complete time-concentration and has a CV of 347 per cent. With production (or income) measured on
a monthly basis, a relative time-concentration (RTC) index of any other crop relative to such a
perfectly concentrated crop can be obtained as the ratio of its CV to the CV of 347 per cent for the
perfectly concentrated crop. This results in an RTC index of 0.81, 0.28 and 0.09 for kapok, pepper
and rubber, respectively, while the completely time-concentrated single-crop rice has an RTC index
of one. The relative time-dispersion (RTD) of production (or income) from an activity or system can
then be measured as one minus its relative concentration, i.e., RTD = 1 - RTC, giving RTD values of
0.19 for kapok, 0.72 for pepper, 0.91 for rubber and zero for single-crop rice. (Note that a perfectly
time-dispersed crop would have a CV of zero, an RTC index of zero and an RTD value of unity.)
Time-concentration of systems
Discussion so far has related to separate crops or activities which are the components of systems.
The relative time-dispersion of a whole-farm system also can be obtained as the sum of the relative
time-dispersion (RTD) values of the productive components which comprise the farm system,
weighted according to their individual importance. Consider the pepper-rubber-kapok growers
around Matale in the Kandy Hills of Sri Lanka. Here a common crop system consists of old rubber
trees thinned out to about 40 per acre which are tapped for latex, but the main function of which is to
support pepper vines, at two vines per tree. The rubber-pepper fields are enclosed by live fences of
kapok, yielding floss and (oil) seed, which also support pepper. Thus the products of this system are
latex, pepper, kapok floss and seed (and kapok pods for household fuel). The annual value of
production per acre is some Rs 2 400 for pepper, Rs 800 for rubber and Rs 600 for kapok or Rs 3
800 in total. Using these product values as relative weights for the three crop components of the
system and taking their relative time-dispersion values from Table 6.6, the system as a whole would
have a relative time-dispersion (of production and income) index value of: 0.72 (2 400/3 800) + 0.91
(800/3 800) + 0.19 (600/3 800) = 0.68.
For the several reasons discussed above, the time-dispersion of income, and especially of sustenance
food production, is an important dimension of small-farm system performance. Note also that the
time-dispersion of more complex systems generating even more products can be quantified by the
method outlined.
TABLE 6.6 - Relative Monthly Production (or Income) and Relative Time-concentration and
Relative Time-dispersion of Four Crops on a Sri Lankan Farm
Monthly production (or income) as a percentage of annual total by crop

Month and statisticsa
Kapok Pepper Rubber Single paddy
January 0 20 13 0
February 0 10 9 100
March 0 2 9 0
April 0 0 8 0
May 80 2 5 0
June 20 7 4 0
July 0 12 7 0
August 0 3 7 0
September 0 2 11 0
October 0 3 7 0
November 0 14 11 0
December 0 25 9 0
8.33 8.33 8.33 8.33
V 542.42 64.61 6.61 833.39
SD 23.29 8.04 2.57 28.87
CV 280% 96% 31% 347%
RTC 0.81 0.28 0.09 1.00

RTD 0.19 0.72 0.91 0.00
a The statistical measures , V, SD and CV are derived as explained in Section 6.2.3.

'.... but in this way the collectors of Kittul (fibre) kill a large number of the palms yearly, to the great sorrow of the toddy and jaggery
makers who depend on the Kittul to afford them a living.'
Colonial Secretary of Ceylon to Madras Government, 1890.
By sustainability is meant the capacity of a system to maintain its productivity/profitability at a

satisfactory level over a long or indefinite time period regardless of year-to-year fluctuations (i.e., of
its short-term instability). In an agricultural production context, sustainability is relevant to farming
systems of whatever composition, but not necessarily to the individual production phases of short-
term crops. The concept involves the evaluation of farm activities and systems in terms of their
(interrelated) ecological, economic and socio-cultural sustainability over long time periods of many
years.
From a national or agroecoregional perspective, reference may be made to the spiral of
unsustainability (CGIAR 1995). As depicted in Figure 6.3, under the pressure of increasing
population and inappropriate policies, this is a downward spiral of diminishing resource availability,
deteriorating environmental quality and increasing poverty leading to economic, social and political
instability. Farmers, especially small farmers of Types 1 (subsistence) and 2 (semi-subsistence), are
both possible victims of and contributors to this spiral of unsustainability should it occur. Conversely,
through the management of their resources in a sustainable way, farmers can help to prevent its
occurrence. Sustainability is thus a very important criterion in assessing the performance of existing
and potential farm activities and systems.
FIGURE 6.3 - The Spiral of Unsustainability
Sustainability is a multidimensional concept. In the context of farm systems it may relate to physical,
biological, economic and social attributes. Assessment of the sustainability of a particular farm
system from both a private and a public view might involve judgements as to its merits in terms of
such characteristics as listed below (National Research Council 1993, Ch. 3). To exemplify this
listing, intensive cropping (1) on good soil/slope conditions and (2) on poor soil/slope conditions
and (3) perennial tree crop plantations are rated high (H), medium (M) or low (L) relative to each
attribute.
(1) (2) (3)
Biophysical attributes:
Nutrient cycling capacity M L H
Soil and water conservation capacity M L M
Stability to pests and diseases L L L

Level of biodiversity L L L
Carbon storage L L M
Economic attributes:
Requirement for external inputs H M H
Provision of employment H M M
Generation of income H L H
Social attributes:
Health and nutritional benefits M M L
Cultural and communal viability H M M
Political acceptability H M H
Sometimes, because of the dominance of some particularly negative attribute, the unsustainability of a
farm system will be clearly evident. Often, however, because of its long-term perspective and
multidimensional nature, assessment of a farm system's sustainability must be a matter of judgement or
degree involving tradeoffs between such various attributes as listed above. Nonetheless, all farm
activities and systems are either sustainable - i.e., capable of indefinite continuation - or exploitative,
or restorative. By-and-large, most tree crops, pasture-based livestock (at reasonable stocking rates),
sawah or terraced paddy at reasonable intensity, and field-crop systems with a long ley phase are
sustainable. Many nutmeg groves in the Kandy Hills are still producing at about 100 years of age. In
the same area there are clove plantations which naturally regenerate themselves from seed-fall and
thus have an indefinite life. In the Sri Lankan mid-country there are many tea fields with over 100
years of recorded production (but many others, exploitatively managed in the past, have long since
had to be abandoned). Vacancies which occur in row-planted cinnamon are usually in-filled to give
this crop an indefinite life. Old coconut is commonly inter-planted with young palms with the same
result. At their traditional levels of land-use intensity, these are sustainable systems.
On the other hand, as evidenced by Byerlee (1992) and Gill (1995), 'continuous' intensive rice or
wheat systems even on good irrigated lands are probably not indefinitely sustainable. The more or
less continuous row cropping of clean-cultivated cassava, maize, oilseeds and cotton on lands of
significant slope, with the now common annual 'fix' of urea, is certainly not sustainable. In Java, most
of the remaining farm systems of the Slendro hills, the southern flanks of Mt Lawu and the karst tracts
skirting the Indian Ocean are little more than monuments to systems that have been pushed beyond
their limits, as are many of those in the Himalayan foothills. The people cling to them because they
have no choice ... 'Bare ruined choirs where late the sweet birds sang'.
Causes of farm unsustainability
Beyond the pressure on the agricultural resource base induced by such primary social causes as
population pressure and poverty, farm systems may become unsustainable due to many factors of
which the following are probably the most important:
(1) Soil loss due to sheet/rill/gully erosion if unchecked will remove the physical base for plant production. By and large,
engineering methods of soil conservation (terraces except on wet paddy lands, contour drains, diversions, strip cropping etc.)
have not been successful in Asia (or Africa) except when installed by estates or authoritarian governments. Thus the only practical
approach to sustainable land use is through less intensive crops and less demanding (but not necessarily less productive) farming
systems combined with a farming systems development approach to soil conservation and sustainability as argued by Norman and
Douglas (1994).
(2) Soil structure deterioration and nutrient loss through leaching and over-cropping, especially
when combined with actual soil loss (above), will also necessitate eventual abandonment of the
system (or the land), or the application of ever-increasing quantities of artificial external inputs -
leading eventually to the same consequence, often together with adverse downstream effects and
watertable pollution.
(3) Declining terms of trade or long-run adverse movements in agricultural commodity prices
relative to input costs (especially of imported inputs) are increasingly a prospect facing much tropical
and sub-tropical produce (e.g., some oilseeds, cassava chips, sisal and the other coarse fibres). At
some point some of these crops might well not be sustainable in marginal producing areas and will
have to be abandoned for economic reasons. To the extent that they form components of systems, these
systems will have to be restructured. Increasingly, economic pressures for change are also reinforced
by socio-political factors: e.g., the liquidation of sub-marginal tea estates in Sri Lanka for village
settlement; the growing pressures for inter-row production of food crops on Malaysian cash-crop
estates; pressures for diversification and food production on Mauritian sugar estates.
(4) Government failure through the introduction of inappropriate policies affecting agricultural
production and resource use or the failure to introduce appropriate policies for the protection of
natural resources and the environment (Pinstrup-Andersen and Pandye-Lorch 1994; Scherr and Yadav
1996). Historically, many developing countries have had food price policies favouring urban
consumers at the expense of producers. This has undoubtedly engendered poverty and resource
degradation among marginal producers. Likewise, many countries have inadequate controls on the use
of agricultural chemicals whose indiscriminate use has often caused widespread environmental
pollution.
(5) Biological factors (disease, pest outbreaks) have on more than one occasion led to the decimation
of crops and farm systems, and to the impossibility of restructuring these under conditions which
could be economically sustained (e.g., the abandonment of coffee in Sri Lanka in the 1880s due to
rust).
(6) Inequitable research and development relating to a crop in one geographical area not
infrequently reduces its economic sustainability in other areas. To the extent that improved varieties
and technologies are developed in countries already enjoying a comparative advantage (e.g., coconut
in Philippines, rubber and oil palm in Malaysia, specialist tea in China), this will force changes in the
farming systems of less advantaged countries, and lead either to their abandonment of these particular
crops or, more likely, restructuring of their farming systems into mixed and probably more complex
systems.
(7) Regional interrelatedness, whether physical, economic or political, provides another set of
factors that can lead to unsustainability. The causes of unsustainability noted in (1) and (2) above,
soil erosion and degradation, arise on an individual farm or local group of farms and can (at least
theoretically) be removed by local action on a watershed basis. But the non-maritime floods which
increasingly devastate the delta farms of Bangladesh arise from causes (e.g., deforestation of the
Himalaya chain) located in other up-stream provinces - or indeed in other countries - and these, not
amenable to local action, are affecting the sustainability of downstream agriculture on a vast scale.
Acid rain in Europe, chemical pollution in the Mississippi, the Nile, the Vistula and some of the East
African lakes - these are other examples of real threats to farm system sustainability which arise
beyond the farm gate and often beyond the national border.
Alternative system-management strategies
Alternative on-farm management strategies towards resource and system sustainability or exploitation
are illustrated by the examples depicted in Figure 6.4. Curve (1) represents a uniform and sustainable
system producing 60 income units ($) annually. A second possible system such as depicted by curve
(2) might yield an initially higher income, but one which declines over time (e.g., because of erosion,
salinity, chemical pollution) until the system becomes economically unsustainable after 20 years.
From a social viewpoint, farmers should select system (1). However, there are many reasons why in
fact they often choose the exploitative system (2): (i) the need for maximum current income (as
distinct from future income) because of poverty, debt, exploitation by landlords, conditions of tenancy
etc.; (ii) limited length of farmer planning horizons (although system (2) is not sustainable, it yields
greater total returns if the planning horizon is less than 20 years); (iii) failure to recognize the fact that
the declining income of system (2) is caused by resource exploitation (environmental damage is
gradual and insidious); (iv) ignorance that a better, sustainable system might exist; and (v) selection
of system (2) as a deliberate choice, but with the intention to later - when family income needs
decline, when the sons are educated, when the daughters are married - switch to a less exploitative
system, or even to a restorative one as represented by curve (2a).3
3 Consideration of these factors leads to the conclusion that the presence of a conservation program in a particular area and the
engineering knowledge this implies in designing contour banks, grass waterways, avoiding soil salinity etc. is only one of several
necessary conditions for sustainable land use. The others are economic, social and political. The rate of accumulation of
engineering and scientific knowledge needed for conservation has long exceeded rates of change in public attitudes towards
conservation, without which such engineering solutions cannot be applied - or, if imposed, maintained.
The third alternative, adoption of a restorative system or one which improves the initial productivity
conditions, is represented by curve (3). The accumulated income from system (3) exceeds that from
system (1) but only after 20 years. System (3) is therefore a rational farmer alternative only for farm
families with long planning horizons or who are free from the exploitative pressures of debt and
poverty and can afford to wait. From a social viewpoint, however, system (3) will always be a
superior alternative.
FIGURE 6.4 - Illustrative Income Flow of Sustainable and Unsustainable Systems
There are many examples of restorative or resource-creating systems - though not as many as of
exploitative systems. They are of two broad types. The first seeks to convert some previously
exploitative system into a sustainable one. In Sri Lanka several thousand families have been settled
on farms growing restorative soil-protecting tree crops in dense stands on old eroded tea lands
(without removing the tea). There also some of the estates have been converting from (exploitative)
tea to (regenerative) cloves and cardamom, again by interplanting. The second type of restorative
system seeks to so improve some initially poor resource base that a sustainable fanning system
becomes possible. In Malaysia and Philippines, the nipah palm, planted in saline mud flats, is used to
dry out these coastal tracts and create conditions under which more productive species - coconut,
breadfruit, jackfruit - can be established as the basis of future settlement farms. Likewise, in the arid
zones of Australia and southern Africa, the shrub saltbush is used to remove soil salts and up-grade
grazing systems.
Sequential exploitation-restoration
Sustainability does not necessarily require uniformity of production or income over time in all phases
of the system. As depicted in Figure 6.5, the vegetable farmers on poor clay soils in Johor
successively plant restorative groundnuts, the vegetative parts of which after harvest are placed in a
trench, and then a following exploitative crop such as sweet potato is planted to grow largely on the
peanut residues. Following three or four cycles of this successively exploitative-regenerative phase
system, an exploitative subsystem is introduced: when vegetable prices are particularly high, several
exploitative crops might be planted in succession; then, when prices drop, the system is again put into
a restorative legume-phase until fertility is recovered.
FIGURE 6.5 - Successive Exploitative and Regenerative Phases in a Johor Vegetable
Production System
Another well-known example of sequential exploitation-restoration is provided by the shifting
cultivators of Sarawak who slash-and-bum bush to grow their main subsistence maize crop, following
this with upland paddy (and then possibly with a cassava crop if land is scarce) before abandoning
the site to bush for a six-, seven-, eight-... year restorative phase. The system consists of the
successive exploitation or run-down of six, seven, eight ... individual land parcels which comprise
the 'farm'. In former times it was stable.
Exploitation over all parcels was in balance with restoration. But now, in many areas of Sarawak,
Kalimantan, the southern dry zone of Sri Lanka and parts of Africa, the restorative phase in such
slash-and-bum systems has had to be reduced due to population pressure and the exploitative phase
prolonged. Because also of the temptation to produce cash crops in addition to food, the exploitative
phase is now much intensified and such slash-and-bum systems are breaking down.
When applied to activities, this last of the eight properties requires that any crop or livestock
component of a system be capable of structural integration with all other components of the system
and its environment in terms of management practices, resources and technologies used, and disposal
of products/by-products. Such structural integration is especially important in relation to long-term
activities where bad decisions made regarding one activity and their adverse effects on other
activities might not be easily rectified. This probably is a statement of the obvious. However, the
more that is learned about the residual effects of herbicides and pesticides and their further effects
lower down the food chain, the more apparent it becomes that this property of systems and their
components has been neglected in the past.4
4 One does not have to go to Asia for concrete examples. In many sugarcane producing areas, including Australia, fields have in
the past been so liberally sprayed with weedicides that they are not now permitted by health authorities to be used for animal
production, possibly for a 'detoxification' period of 20 years. As a consequence, those farmers who would otherwise have
adjusted out of sugar to other products, because of low sugar prices, find themselves locked into this crop indefinitely. What was
previously accepted as good sugarcane management is now seen to be incompatible with other alternative activities.
When applied to whole-farm systems the requirement is that each of these be at least not incompatible
with other systems in the village or area (for purposes of obtaining inputs and disposing of products).
Although the subject has been largely ignored in farming systems development research, there are
often close complementary relationships between groups of systems, e.g., in neighbouring villages,
without which each would be weaker (Prabowo and McConnell 1993). In the Solo Valley of Java,
the farms of Batan (where they grow only paddy and cannot keep cattle) complement those of nearby
Boyalali (where they cannot grow paddy but have many draught cattle). This type of inter-village or
inter-system complementarity is common throughout Asia. Where it exists, development efforts which
remain preoccupied with the optimization of systems in technical and social isolation are likely to
achieve no great success.
A related requirement is that both whole-farm and activity systems be compatible with the wider
physical, biological and socio-religious-cultural environment. This desirable system property of
environmental friendliness appears so obvious that examples should not be necessary. Nevertheless,
it is frequently overlooked in farm-systems development. The tale of rice schemes (planned for non-
rice eaters) - of mechanization schemes (in areas without a mechanical tradition and with surplus
labour) - of chemically-based crop production projects (which would obliterate a thousand years of
village culture) .... if it were told, this tale would be a long and doleful one.
6.2.9 Summary
It now remains to consider how the criteria outlined in Sections 6.2.1 to 8 above would be applied in
specific analytical situations. Such application of the criteria might be considered from the
perspective of, first, necessity and, second, desirability.
In terms of necessity, to use only one criterion to assess system performance will sometimes be
sufficient. Commonly, for commercial farms, this is some aspect of profit, e.g., gross margin or net
farm income. But the use of some other single criterion may also sometimes be necessary: e.g.,
analysis in support of planning a farm credit program in Field C (i.e., analysis oriented to systems
above the farm-household level) might be concerned primarily with the need for credit as determined
by the time-dispersion patterns of farm income, and for this purpose most of the other system
properties might be ignored. More often it will be necessary to work with some subset of the eight
properties and their criteria. In relatively few situations it might be necessary to consider all eight
factors, e.g., in planning comprehensive general-purpose rural development projects. As noted above,
most of the properties are capable of quantitative measurement; those that may present difficulties,
such as sustainability and compatibility, might have to be assessed subjectively.
So much for what may be necessary. In terms of desirability, from both a private and a public view,
the aim should be to have farm-household systems that are sustainable and environmentally friendly.
Any farm management analyst worth his or her salt will therefore always endeavour to appraise farm
and farm-household systems in terms of these two overarching criteria.
Operating objectives, system properties and criteria for their measurement are summarized in Table
6.7.
TABLE 6.7 - Summary of Farm-household System Objectives by Farm Type and Performance
Criteria
OPERATING OBJECTIVES
Farm type Primary objectives
(1) Small, largely subsistence, family Subsistence
(2) Small, part commercial, family Sustenance and some cash income
(3) Small, independent, specialized, family Cash-based sustenance

(4) Small, dependent, specialized, family Cash-based sustenance
(5) Large, commercial, family Mainly profit
(6) Commercial estate Profit
SYSTEM PROPERTIES AND CRITERIA FOR MEASUREMENT OF PERFORMANCE
Property Criterion
1. Productivity Yield per land unit or animal unit or other unit of resource or the value of output
per unit of cost.
2. Profitability In financial terms or measured subjectively as net benefits.
- of activities Gross margin.
- of whole farms Measures discussed in Chs 5 and 7.
- over time Measures discussed in Ch. 10.
3. Stability Coefficient of variation (CV).
4. Diversity Simpson's diversity index (DI)
- of activities Number of activities in system.
- of products Number of products of system.
- of income Income diversity ratio (R).
5. Flexibility Number of first, second ... degree uses to which products can be put (sold,
consumed, processed, stored).
- of a single
product
- of all system
products
6. Time- Relative dispersion of generation over the operating period (usually year) on a
dispersion daily/weekly/monthly/quarterly basis as measured by the relative time-dispersion
index (RTD).
- of production
- of income
- of whole-farm
system
7. Sustainability No single general quantitative measure. Measurement would relate to physical,

biological, economic and social factors with reference to the number of years
over which a given system may be operated before its continuation becomes
infeasible or inadequate.
8. No cardinal measure but an ordinal measure ranking activities or systems on a

Complementarity scale of high, low, neutral or negative relative to their physical, biological,
and socioeconomic, cultural and religious environmental friendliness could be used.
environmental
compatibility
- of activities in
a system
- of systems in
the environment
6.3 REFERENCES
Byerlee, D. (1992). 'Technical Change, Productivity and Sustainability in Irrigated Cropping Systems
of South Asia: Emerging Issues in the Post-green Revolution Era', Journal of International
Development 4(5): 477-496.
CGIAR (1995). Report of the Task Force on Sustainable Agriculture, Document No. MTM/95/10,
CGIAR Secretariat, World Bank, Washington, D.C.
Clayton, E. (1983). Agriculture, Poverty and Freedom in Developing Countries, Macmillan Press,
London.
Collinson, M. (1983). Farm Management in Peasant Agriculture, Westview Press, Boulder.
Gill, G.J. (1995). Major Natural Resource Management Concerns in South Asia, Food, Agriculture
and the Environment Discussion Paper 8, IFPRI, Washington, D.C.
Grimwood, B.E. (1975). Coconut Palm Products: Their Processing in Developing Countries, FAO
Agricultural Development Paper No. 99, Food and Agriculture Organization of the United Nations,
Rome.
Harwood, R.R. (1979). Small Farm Development: Understanding and Improving Farming Systems
in the Humid Tropics, Westview Press, Boulder.
Kumar, M.B., S.J. George and S. Chinnamani (1994). 'Diversity, Structure and Standing Stock of
Wood in the Home Gardens of Kerala in Peninsular India', Agroforestry Systems 25(3): 243-262.
National Research Council (1993). Sustainable Agriculture and the Environment in the Humid
Tropics, National Academy Press, Washington, D.C.
Norman, D. and M. Douglas (1994). Farming Systems Development and Soil Conservation, FAO
Farm Systems Management Series No. 7, Food and Agriculture Organization of the United Nations,
Rome.
Pinstrup-Anderson, P. and R. Pandya-Lorch (1994). Alleviating Poverty, Intensifying Agriculture,
and Effectively Managing Natural Resources, Food, Agriculture and the Environment Discussion
Paper 1, IFPRI, Washington, D.C.
Ruthenberg, H. (1976). Farming Systems in the Tropics, 2nd edn, Oxford University Press.
Scherr, S.J. and S. Yadav (1996). Land Degradation in the Developing World: Implications for
Food, Agriculture, and the Environment to 2020, Food, Agriculture and the Environment Discussion
Paper 14, IFPRI, Washington, D.C.
Simpson, E.H. (1949). 'Measurement of Diversity', Nature 163: 688.
7. ECONOMIC EVALUATION OF FARM
SYSTEMS: MEASURES FOR
EVALUATION AND COMPARATIVE
ANALYSIS
7.4 DATA SOURCES
7.6 REFERENCES
'We must study the present in the light of the past for the purposes of the future.'
John Maynard Keynes (1883-1946)
Evaluation of whole-farm systems consists essentially of measuring how adequate and how effective
an existing system has been in achieving its planning objectives over some past operating phase
(season or operating year). In the simplest case assumed here, the objective will be generation of
maximum net income in money terms - either directly for market-oriented farms or indirectly by
imputation of values for subsistence-oriented farms. Thus the evaluation will refer primarily to
financial criteria and to factors directly underlying those criteria; it reduces largely to a systematic
consolidation of the costs and returns which arise in various parts of the system - i.e., in the separate
enterprises or activities and in the whole-farm service matrix. This emphasis on costs, returns,
income and profitability, however, is not meant to decry the importance of those other farm system
properties - such as sustainability, environmental compatibility etc. - outlined in Section 6.2. Their
achievement, too, must be evaluated. Some criteria for such evaluations are suggested in Table 6.7.

The demarcation lines which set 'whole-farm evaluation' apart from other areas of farm management
analysis are somewhat elastic and shift according to the intended purpose of the evaluation. The
sequence and scope of evaluative analysis are shown in Figure 7.1. The basic component of
evaluation is a farm operating statement (or a whole-farm budget in financial terms). As indicated,
this draws on and consolidates data from each activity as well as data on unallocatable overhead or
common costs relating to the farm service matrix or farm system as a whole. An important
requirement is that, whatever data are selected for executing the evaluation, they must be consistent
with the system's actual planning objectives (Section 6.1). The next step is to derive measurements of
performance from the base whole-farm budget table. This defines the scope of what is here termed
limited evaluation. The purpose of a limited evaluation is simply to report on the condition or record
the performance of a farm system. In itself it is not intended as a basis for further analysis or action. It
is intended only to report the facts. Nor does it explicitly consider the farm's household system.
In contrast, an expanded evaluation is intended as a basis for further analysis and action, and it does
consider the household system as well as the farm system. As shown in Figure 7.1, expanded
evaluation most commonly takes the form of a comparative analysis of (A) the individual resources
used by the system and (B) the individual activities which comprise the system. Such a comparative
analysis requires the same kind of farm data as used in a limited evaluation. However, for
comparative analysis the data must be in disaggregated (i.e., activity-specific) form and be
supplemented by data relating to the farm household. Comparative analysis is intended to diagnose
problems and provide a basis for remedial action. This defines the scope of an expanded evaluation.
Its purpose is not only to record the achievements of a system but also to look for evidence of under-
achievement and to identify structural weaknesses within the system. It leads to the kind of follow-up
action indicated in the bottom section of Figure 7.1. If weaknesses within either the A stream
(resource allocation) or B stream (activity mix) of the diagram are minor, they might be corrected on
the basis of a little partial budgeting. If they are major, they might require, in the A stream, the
application of response analysis (Chapter 8) and, in the B stream, restructuring of the system using
allocation budgeting or simplified programming or linear programming (Chapter 9). System
weaknesses might also be due to factors not directly related to the activities or the resources used in
the activities, e.g., they might arise in the service matrix with too many/not enough work oxen, too
much/not enough investment in soil conservation, drainage, farm storage, etc.
FIGURE 7.1 Sequence of Analysis in Farm-system Evaluation
To summarize Figure 7.1:
· Limited evaluation is concerned with reporting the facts relating to the past performance of a whole-farm system (the operating
statement).
· Expanded evaluation using comparative analysis is concerned with diagnosing system

weaknesses which might be implied by those facts.
· Subsequent analysis (Chapters 8 and 9) is concerned with prescribing remedial action to remove
weaknesses, possibly to the extent of restructuring the whole system.

As noted, a limited evaluation of a farm system consists of preparing an operating statement (or
whole-farm budget) covering an appropriate period (usually a year) such as exemplified by Table 7.1
and deriving appropriate measurements of whole-farm performance from this operating statement as
shown in Table 7.2.
Table 7.1 records annual costs/returns performance on a whole-farm or pooled basis for a 2.8 ha
farm having paddy, maize, bamboo and cattle activities. (The capital investment inventory of this farm
was presented earlier in Table 5.5.) The data of Table 7.1 are shown as they would typically be
recorded on the farm or received as memory estimates from members of the household. The
production/disposal data are activity-specific but the inputs/costs are 'all mixed up' with no
indication of which inputs have been used by which activities. For the moment, since the intention is
to obtain a picture of the farm as a whole, this does not matter. (But it will cause problems later;
Section 7.2.2.) Also, on small farms it will usually be necessary to distinguish between different
classes of outputs: final products that are sold or consumed by the household, and intermediate
products to be used in the next farm operating phase as resources. For subsistence-oriented farms it
will be necessary to impute prices/values to products which are not sold for cash.
In section A of Table 7.1, the values of all final products are consolidated to a total gross return of Rs
44 000. In section B, all direct input costs of all activities are also consolidated to give a total direct
cost of Rs 12 500. This excludes the value of farm-generated resources (family labour with an
imputed value of Rs 4 500). Section C records farm fixed costs (discussed in Sections 5.3 and 5.4).
Depreciation is shown as a separate item in line D.
TABLE 7.1 - End-of-year Operating Statement for a Mixed Farm of 2.8 Ha
A. All Outputs/Returns (Pooled) (Rs) Notes
Paddy grain sold 25 000
food 6 000
(seed) (1 500) Used on farm
straw (used) (900) Used on farm.
Maize grain sold 7 000
food 800
stover (used) (700) Used on farm.
Bamboo poles sold 1 200
(used) (200) Used on farm.
Cattle calves sold 2 500
milk sold 600
food 900
Total Gross Return: 44 000 Omits (items) used as resources on farm.
B. All Purchased Activity Direct Inputs (Pooled)
HYV seed 2 000

Fertilizers 3 000
Agricides 1 400
Veterinary 1 200
Irrigation fuel 1 900
Transport 600
Hired labour 2 400
Family labour (450 days) (4 500) Based on opportunity cost of Rs 10/day.
Total Direct Costs: 12 500 Omits family labour and farm-generated inputs.
C. All Farm Fixed Costs (except Depreciation) Depreciation recorded in D below
General Charges
Land tax 500
Livestock tax 90
Road-repairs levy 250
Water-use fee 70
Capital Equipment Repairs, Operation
All capital-item repairs 1 600 From Column (5) of Table 5.5.
All capital operating costs 1 000 From Column (6) of Table 5.5.
Total Fixed Costs: 3 510 Omits depreciation.
D. All Capital Depreciation 4 480 From Column (4) of Table 5.5.

Based on the data of Table 7.1, six whole-system evaluation measures are presented in Table 7.2.
The origin of and conditions attached to these factors are noted in the table. There is usually a degree
of subjectivity involved in selecting the 'best' evaluation factors, i.e., in determining the 'best' way of
measuring performance. But which of the measures E, F,... J are most appropriate will, to some
degree, be indicated by analytical circumstances.
TABLE 7.2 - Derived Measures for Annual Whole-farm Evaluation
Value
Measure Calculationa Notes
(Rs)
E. Farm Gross Margin A-B 31

500
F. Farm Net Actual Returns E-C 27 Depreciation not yet charged.

990
G. Farm Net Sustainable Returns F-D 23 Depreciation charged; system now

510 sustainableb.
H. Family Farm Available H=F 27 But only if depreciation is not covered.

Income 990
I. Family Farm Sustainable I=G 23 Long-term Sustainable farm income.

Income 510
J. Total Available Family (H or I) + S 23 S is non-farm income, here assumed to be

Income 510 zero.
a A, B, C and D are as specified in Table 7.1.

b Sustainable refers to the non-depletion of capital assets excluding land.
Measure E, total farm gross margin (of Rs 31 500), is the easiest to derive: it consists of the sum of
all activity gross margins (Chapter 4), or if activity costs/returns have been pooled, the total value of
farm output of final products less total farm direct costs (A - B in Table 7.1). This is a good measure
of performance if the evaluation is needed for making comparisons between similar farms and if their
capital structures (levels of fixed costs, item C in Table 7.1) are similar or relatively unimportant.
An alternative is to use factor F (farm net returns, i.e., TGM - FC). But this level of 'income' (of Rs
27 990) is not yet stable over the long term because it makes no provision for replacing capital
equipment as this wears out. If this is an important consideration, measure F should be decreased by a
depreciation charge to obtain measure G (farm net Sustainable returns) in which case the 'income' (of
Rs 23 510) would be Sustainable over the long run in terms of its fixed-capital requirement. Note that
this is clearly a narrow use of the word Sustainable. It says nothing about the possible loss of
sustainability due to degradation of the land base, falling terms of trade etc.
As discussed in Chapter 5, a depreciation 'charge' is not one involving actual cash; it is only a
bookkeeping convention which, at the farmer's discretion, may or may not be covered by putting
money aside either in a depreciation fund (not a common choice) or (more commonly) as an
investment in the farm to assist future replacement needs. If no money is actually put aside to cover
depreciation, the amount of cash income actually available as family income would be H (the same as
F). This would imply the (temporary or continuing) run-down of farm capital. Finally, concern might
be with total family income from all sources rather than only income from the farm. If so, factor J
would be relevant in that it measures the performance of both the farm and the household components
of the system. (But, here again, income of the farm component might or might not be defined to be
sustainable over the long run depending on whether it is measured as H or I.)
Inter-farm comparisons
The evaluation analysis so far has presented records of performance for a single farm in terms of
measures E, F,... J. However, at the risk of contradicting what was said previously regarding the flow
of analysis in Figure 7.1, it might be noted that comparisons with other farms can also be introduced
at this present stage. Such comparisons would involve comparing measures E, F,... J on the subject
farm with those of similar farms. In fact this would be desirable because although E, F,... J are
adequate as a record and over time would indicate the degree of variability in performance of the
subject farm, in themselves they do not provide any basis at all for judging if the farm's income levels
are good or bad or better or worse than those of other farms. If introduced at this point, whole-farm
comparison would take the form shown in Table 7.3. Thus the subject farm (Farm 1), primarily a
paddy farm, would be compared in terms of the measures E, F,... J with other farms of the village or
area having similar size, soils, water supply etc. but not necessarily the same mix of activities.
Indeed, for farm management analysis in other fields (e.g., Field D, policy guidance), it might be
necessary to compare the subject farm with, e.g., livestock or cotton farms.
TABLE 7.3 - Example Format for Whole-farm Comparative Analysis (Annual Basis)
Farm 1 Farm 2 Farm 3 Farm 4

Evaluation measure
(Rs) (Rs) (Rs) (Rs)
E. Farm Gross Margin 31 500 33 600 30 100 28400
F. Farm Net Actual Returns 27 990 28 100 27 200 22900
G. Farm Net Sustainable Returns 23 510 26 700 25 050 21 700
H. Family Farm Available Income 27 990 28 100 27 200 22 900

I. Family Farm Sustainable Income 23 510 26 700 25 050 21 700
J. Total Available Family Income (I + S) 28 510 28 300 29 600 22 090
All farms differ to some degree in their size, available capital, labour force etc. If more detailed
inter-farm comparisons are to be made, it will be necessary to convert the measures E, F,... J to a
comparable unit basis such as per ha, per Rs 100 of capital, per labour day etc. Thus, if factor I
(sustainable family income) is of interest, further measures would be derived from this such as:
I (Sustainable Family Income) for the subject farm:
(Rs)
Per ha of land 23 510/2.8 = 8 396
Per Rs 100 of capital 23 510/1 066 = 22
Per family labour day 23 510/450 = 52
Per family member 23 510/6 = 3 918
Per adult family member 23 510/3 = 7 837

For some purposes, e.g., when it is intended to use the evaluation as a basis for further comparative
analysis, it is necessary to present the operating statement in both aggregated whole-farm terms and in
terms specific to each separate activity. Here, as previously, the analyst might possibly obtain the
necessary data from farm records. Usually, however, no useful records will be kept and the needed
information must be obtained via intensive discussions with members of the farm household. In either
case, farmers tend not to distinguish clearly - or as clearly as the analyst would like - either between
inputs going to the various activities or between outputs flowing from each of them. The task of
disaggregating the data into activity-specific subsets is often a major but necessary one if later
comparative analysis is to be possible.
Disaggregation will result in a set of operating statements, one for each activity and another for the
farm as a whole, as shown in Table 7.4 (which again refers to the same farm as discussed in the
previous section). Because of their nature, it will usually be possible to 'son out' the direct
inputs/costs belonging to each activity. Some fixed costs also will be obviously activity-specific;
where they are not, because they are of an overhead nature, they are assigned or charged to the farm
as a whole as shown in the second-last column of Table 7.4. Data for the whole farm, aggregated
across the activity and farm-as-a-whole columns, are given in the last column of the table.
The several components of fixed costs were defined in Chapter 5 and would be obtained as in Table
5.5. To calculate these and also depreciation, it is usually necessary to construct a separate worksheet
in which all capital investment is allocated item-by-item to the respective activities on the basis of
relative use of each item by each activity. Such a worksheet is illustrated by Table 7.5 in which, e.g.,
based on the farmer's advice or other knowledge, the tractor is allocated in the proportions of
3:2:0.5:4.5 to paddy, maize, cattle and the farm as a whole, respectively; but all of the thresher to
paddy; and since farm fences serve each of the activities, these are allocated to the farm as a whole.
In Table 7.4 the evaluation measures E, F, G have exactly the same definitions when applied at
activity level as when applied to the whole farm in Table 7.2. However, measures H, I, J apply only
to the whole farm. The evaluation then continues, on a per unit of resource basis, in line items K and
L of Table 7.4. Measure K repeats measure E but now on a per ha, per Rs 100 capital and per family
labour-day basis, while L repeats G but now on a similar per unit of land/capital/labour basis.
TABLE 7.4 - Farm-system Evaluation using Activity-specific Data (Annual Basis)
TABLE 7.5 - Worksheet for Allocation of Farm Capital Investment to Specific Activities or
Farm as a Whole as required for Analysis of Table 7.4 (Annual Basis)
Fixed Capital Inventorya Specific Activities Farm as a Whole
Value Paddy Maize Bamboo Cattle

Item
(Rs) (Rs) (Rs) (Rs) (Rs) (Rs)
Land 50 000 - - - - 50 000
House 10 000 - - - - 10 000
Sheds 5 000 2 000 2 000 - 500 500
Tractor 10 000 3 000 2 000 - 500 4 500
Thresher 2 000 2 000 - - - -
Cultivators 3 000 2 000 1 000 - - -
Ox gear 600 300 200 - - 100
Barn 5 000 1 000 1 000 - 1 000 2 000
Fences 6 000 - - - - 6 000
Dam 8 000 3 000 - - - 5 000
Pump (house) 4 000 - - - - 4 000
Oxen 3 000 - - - 3 000 -
Totalb 106 600 13 300 6 200 0 5 000 82 100
a As per Table 5.5.

b As transferred to Table 7.4.
To summarize, the right-hand column of Table 7.4 contains essentially the same information regarding
the whole farm as did Tables 7.1 and 7.2. In Table 7.4, however, costs/returns are 'split up' or
assigned as far as possible to the respective activities. When this cannot be done, they are assigned to
the farm as a whole. Obviously this conveys much more information than did Tables 7.1 and 7.2; but
whether this is sufficient to warrant the much greater clerical effort depends on analytical
circumstances. It is unavoidable if the intention is to proceed to the next stage, comparative analysis.
So far no information has been obtained regarding the productivity of the various resources - or more
specifically, the value of output which might be attributed respectively to the resources of land,
capital and labour.
In the whole-farm situation (Section 7.2.1), 2.8 ha of land were used in the production of Rs 23 510
of annual sustainable family income (measure I of Table 7.2). But this does not mean that the resource
land produced Rs 23 510/2.8 = Rs 8 396 of income per ha or that the productivity value of land is Rs
8 396 per ha. Similarly, in line L(iii) of Table 7.4 the net sustainable income G per day of family
labour used in each activity is respectively Rs 23 030/150 days = Rs 153 in paddy, Rs 29 in maize,
Rs 60 in bamboo and Rs 11 when used for cattle. Again these are only input-output coefficients: they
do not indicate, e.g., that the productivity value of family labour in maize is Rs 29 per day.
The reason that these coefficients are not measures of resource productivity is obvious: if, e.g., in the
case of the whole farm, all of the farm's net sustainable returns of Rs 23 510 were attributed to the 2.8
ha of land (at the rate of Rs 8 396 per ha), there would be nothing left with which to reward or 'pay
off the other factors of labour and capital - or, stated another way, this would imply that labour and
capital have played no role in farm production, or that the value of output due to them is zero.
Residual productivity values of resources
Resource values based on productivity can be estimated (with some qualification) by the residual-
value method. This is most conveniently applied to sets or categories of resources rather than to
individual resource items which comprise these sets. On Asian farms the most important sets are the
classical triad of land, labour and capital. The procedure is to select the category of interest, say
land; then 'pay off the other resource categories of labour and capital at their market rate (or other
appropriate rate, i.e., at their opportunity cost); then deduct these resource 'payments' for labour and
capital from total farm production value, and regard the residual as the productivity-based value of
land, the resource of interest. The procedure is then repeated to obtain the analogous values of the
other resources - taking each in turn, paying off the others and assigning any residual value to the
subject resource. Note that the 'payments' referred to in this procedure are not actual payments. They
are simply ascribed values used in the productivity calculation.
The procedure is illustrated in the worksheet of Table 7.6 in which the productivity-based values of
land, capital and labour are obtained for the whole-farm system. The upper part of the table records
the necessary data and their source. The selected measure of whole-farm performance on which the
analysis is based is item G of Table 7.2, i.e., farm net sustainable returns. In section M of Table 7.6
the returns to land, or that part of total income G which is attributable to land, is G (i.e., Rs 23 510)
less payments to capital and family labour for their respective services (i.e., Rs 5 660 and Rs 4 500)
giving a residual of Rs 13 350 attributable to land, or Rs 4 768 per ha. In this calculation the ascribed
payment to capital of Rs 5 660 is found by applying a market rate of 10 per cent to the total value of
capital (excluding land) of Rs 56 600 used in the system (Table 5.5). Similarly, the payment to labour
of Rs 4 500 is found by applying an opportunity cost of Rs 10 per day to the 450 days of family
labour used (Table 7.1). Note that this payment to labour does not include hired labour which has
already been accounted for by its inclusion as a direct cost (Table 7.1) in the calculation of G (Table
7.2).
The logic and steps in obtaining the next valuation measure, item N or returns which might be
attributed to use of the capital resource, are identical. Again starting with item G then 'paying' Rs 5
000 and Rs 4 500 for use of the land and labour resources respectively, the residual of Rs 14 010 can
be viewed as a measure of the productivity value of capital. On a per-unit-of-resource basis this is Rs
25 per Rs 100 of capital (except land) used by the system. And again using the same procedure, the
return that can be attributed to family labour in evaluation item 0 is Rs 29 per labour day.
It might be noted that these allocations or residual 'payments' to each of the sets of resources are not
to be regarded as holding true in any aggregative or financial accounting sense. If the total residual
returns to each of the resources are summed, i.e., 13 350 + 14 010 + 12 850 = Rs 40 210, the result is
obviously much greater than the amount of net sustainable returns (Rs 23 510, item G) actually
realized. Interpreted in this aggregated way, the results have no meaning because of the double
counting inherent in such aggregation. The correct interpretation of the results of the evaluations of
items M, N, O is:
If (a) all farm operating and fixed costs are deducted from the total value of farm production, and then (b) the residual from
operation (a) is allocated to the payment at normal prices of all production resource sets (land, labour, capital) except one, then
the final residual from operation (b) can be viewed as an amount that can be attributed to that one resource or factor for the
services it provides to the whole-farm system. This assumes that other resources do not generate returns which are greater than
(or less than) their normal price. Insofar as this assumption is incorrect, the productivity value of the resource will be over- (or
under-) estimated.
TABLE 7.6 - Worksheet for deriving per Unit Productivity Values of Farm Resources (Land,
Labour, Capital) on a Whole-farm (Annual) Basis
Basic Data
(a) farm land 2.8 ha
(aa) value of land 50 000 RS (from Table 7.5)
(aaa) 'payment' to land 5000 Rs (10% charge on (aa))
(b) farm capital 56 600 Rs (except land, from Table 7.5)
(bbb) 'payment' to capital 5 660 Rs (10% charge on (b))
(c) family labour used 450 days (from Table 7.1)
(cc) value of labour 4 500 Rs (Rs 10/day)
(ccc) 'payment' to labour 4 500 Rs
M. Net Farm Sustainable Returns (Income) to Land:

(i) Total: G less 'payments' to capital and labour: Rs (23 510 - 5 660 - 4 500) = Rs 13 350
(ii) Per unit (ha) of land: Rs (13 350/2.8) = Rs 4 768
N. Net Farm Sustainable Returns (Income) to Capital:
(i) Total: G less 'payments' to land and labour: Rs (23 510 - 5 000 - 4 500) = Rs 14010
(ii) Per unit (Rs 100) of capital: Rs 14 010/566 = Rs 25
O. Net Farm Sustainable Returns (Income) to Family Labour:
(i) Total: G less 'payments' to land and capital: Rs (23 510 - 5 000 - 5 660) = Rs 12 850
(ii) Per unit (day) of family labour: Rs 12 850/450 = Rs 29
The information generated by the residual-value method has two main uses. First, comparisons are
often required between farms regarding the relative productivity of one particular resource, e.g., the
productivity of labour on Java paddy farms vs Sumatra paddy farms, or between farms in the same
village. Second, the information may be used as a guide for determining a reasonable economic price
at which to buy or sell or exchange resources. In the example of Table 7.6 the productivity value of
family labour is Rs 29 per day. Given this information, family members would have a better basis for
deciding whether to stay home and work on the farm or to look for off-farm work. Likewise, an
economic price to pay for purchasing or renting land can also be approximated. On the farm of Table
7.6 the productivity value of land is Rs 4 768 per ha. This offers a guide to the annual rent the farmer
could pay to acquire a little more land.
Low or negative resource productivities
When the residual method is applied to determine the productivity of the family labour resource it is
not unusual to find that, by the time all other resources have been 'paid off at their going rates, there is
no production value left to assign to labour. Residual labour value might even be negative (meaning
that total productivity of this particular farm system is so low that even the other resources of capital
and land cannot earn a normal rate of return). This simply reflects the real world. Sometimes the
system weaknesses which this indicates can be removed by restructuring the system, using the
methods discussed in Chapter 9. But, in the Asian context, more often than not the problems lie at
higher levels - in systems of Order Level 13, 14, 15 - and their solution is beyond the capacity of
individual farmers.
The management resource
So far, no attention has been paid to system management as a separate factor or resource. Indeed, on
small Asian farms of Types 1 and 2 in particular but also often of Types 3 and 4, management is so
closely inter-linked with family labour that it might not be practicable to try to distinguish between
them. This contrasts with the situation on farms of Types 5 and 6 and on many Western farms where
the farmer expends relatively less physical effort in the fields but relatively more mental effort in
planning the use and assignment of his or her greater stock of capital. (Indeed, on Type 6 farms
(estates) there will usually be specialist hired management.) In such situations it is often desirable to
evaluate the contribution of management as an important separate resource.
This can be done by extending the residual-value method to encompass the four resource categories of
land, capital, labour and management rather than just the first three. The value of management is then
estimated as the residual of total farm output value after land, capital and labour have been 'paid off,
i.e., a value of Rs 8350 for the farm of Table 7.4. Management as a resource, however, exists in a
qualitative dimension. It can really only be measured quantitatively in terms of what it achieves after
the event, i.e., ex post, as discussed in Section 7.2.5.
The resource productivities obtained in the previous section refer to resource use on the farm as a
whole. In some situations it may also be necessary to derive such productivity data for each resource
used in each activity. As previously, these estimates are obtained using the residual method. Again,
based on the farm of Table 7.4, an example is shown in Table 7.7 where the productivity-based
values of land, capital and labour when used in producing maize are derived as Rs 1 280, Rs 2 and
Rs 5, respectively, per ha of land, per Rs 100 of capital and per day of family labour. Obviously the
value of capital employed in maize is very low. Just how low it is in comparison with capital
employed in the other activities could only be determined if similar analyses were conducted for
paddy, bamboo and cattle.
TABLE 7.7 - Worksheet for deriving per Unit Productivity Values of Farm Resources (Land,
Labour, Capital) on an Activity Basis for the Maize Activity of Tables 7.4 and 7.5 (Annual
Basis)
Data on Resources used for Maize
(a) land used 1.0 ha
(aa) value of land (Rs 50 000/2.8) 17 857 Rs
(aaa) 'payment' to land (Rs 17 857 x 0.10) 1 786 Rsa
(b) capital used (except land) 6 200 Rs
(bbb) 'payment' to capital (Rs 6 200 x 0.10) 620 Rs
(c) family labour used 100 days
(cc) value of labour (Rs 10 x 100) 1 000 Rs
(ccc) 'payment' to labour 1 000 Rs
Net Returns to Land used for Maize
(i) Activity total: G for maize less payments to capital and Rs (2 900 - 620 - 1 000) = Rs 1
labour: 280
(ii) Per ha of land: Rs 1 280/1 = Rs 1 280

Net Returns to Capital used for Maize
(i) Activity total: G for maize less payments to land and Rs (2 900 - 1 786 - 1 000) = Rs
labour: 114
(ii) Per Rs 100 of capital: Rs 114/62 = Rs 2
Net Returns to Labour used for Maize
(i) Activity total: G for maize less payments to land and Rs (2 900 - 1 786 - 620) = Rs 494
capital:
(ii) Per day of family labour: Rs 494/100 = Rs 5

a Note that this is an overestimate because there is some double cropping of maize and paddy.

Individual productivity values for land, labour, capital and management derived by the residual-value
method either on a whole-farm basis (Section 7.2.3) or on an activity basis (Section 7.2.4) can
provide a useful guide in farm diagnosis. Such individual productivity values, however, must be used
with caution. They suffer from two deficiencies. First, the residual-value method, by its nature,
attributes all the 'excess benefits' to the resource being appraised; other resources are only 'paid off at
their opportunity or market cost. Second, productivity values on an individual resource basis conflict
with a systems view of the farm. From a systems perspective, the essence of farm production is that
management is applied to a conglomerate set of factors (i.e., land, labour and capital resources as a
totality) which interrelate and interact with one another and with management to produce outputs.
From such a perspective, evaluation of the farm system implies that the resources used should be
assessed as a totality or unified whole, not as a set which can be broken down into separate
categories of land, labour, capital and management. Such a systems view leads to the concept of total
factor (or resource) productivity whereby productivity is assessed on a whole-farm or (with some
qualification) activity basis (Dillon and Hardaker 1993, pp. 79-91).
Total factor productivity
The essence of productivity is output per unit of input over some specified time period. Total factor
productivity is thus total output divided by total input. As in the case of individual resource
productivities, the problem of outputs and of inputs each being of diverse physical forms is met by
aggregating each to their respective total on the basis of the common unit or numéraire of money value
based on market price for outputs and market or opportunity cost for inputs. Using again the example
farm of Tables 7.1, 7.4 and 7.5, this is done on a whole-farm basis in the worksheet of Table 7.8.
Apart from the aggregation of land, labour and capital into a single money value, the analysis differs
from that of Table 7.4 in that the costs of all inputs, whether or not they actually involve cash
payments, are included. The appraisal is in economic rather than merely financial terms. Thus family
labour is now included as an input cost at its market value of Rs 10 per day and an assumed
opportunity cost of 10 per cent interest is now charged on all capital including land. Likewise,
depreciation is included because it also is a cost of production. Total costs are thus Rs 35 650 as
compared with the cost of B + C + D = Rs (12 500 + 3 510 + 4 480) = Rs 20 490 used in deriving net
sustainable returns (item G) in Table 7.4. Gross total factor productivity is thus:
Total gross returns/Total costs = 44 000/35 650 = 1.23,
i.e., Rs 1.23 of output is obtained per Rs of the conglomerate input of land, labour and capital
resources. Analogously, net total factor productivity is:
Total net returns/Total costs = 8 350/35 650 = 0.23,
i.e., Rs 0.23 net is obtained per Rs of input or, in percentage terms, the net return per Rs of input is 23
per cent.
Return to capital and to equity
Table 7.8 also presents two other whole-farm evaluation measures of an economic productivity
nature. These are return on capital (including land) and return on equity used in the farm. (Equity is
total farm capital minus borrowed capital used in the farm - it is the farmer's share of the total farm
capital.) Capital constitutes a relevant basis for evaluation because capital (including land)
constitutes both the physical resource base and the investment base on which farm production occurs;
as well, the capital value of the farm is also its market value if it is to be bought or sold. Thus return
on capital indicates how efficiently the farm asset base is used while return on equity can be
compared with the rate of return that may be available from alternative investments.
Return on capital is calculated as:
(Total net returns/Total capital)100 = (8 350/106 600)100 = 7.83%.
Now suppose, as specified in Table 7.8, that the farmer has a debt of Rs 10 000 on the farm at an
interest cost of 10 per cent per annum, i.e., he or she only has an equity of Rs 96 600 (or 91 per cent)
in the total farm capital of Rs 106 600.
TABLE 7.8 - Worksheet for deriving Total Factor Productivity, Return on Capital and Return
on Equity on a Whole-farm (Annual) Basisa
Whole-farm Value
Measure on a Whole-farm Basis
(Rs) (Rs)
Total Gross Returns 44 000
Costs
Direct costs
Labourb 6 900
Other 10 100
Fixed costs
General charges 910
Other 2 600
Depreciation 4 480
Interest on capitalc
Land 5 000
Other 5 660
Total Costs 35 650
Total Net Returns 8 350
Gross Total Factor Productivity 1.23
Net Total Factor Productivity 0.23
Total Capital 106 600
Return on Capital 7.83%
Equity Capitald 96 600
Borrowed Capitald 10 000
Cost of Borrowed Capitald 1 000
Return on Equity 7.61%
a Data, as required, from Tables 7.1, 7.4, and 7.5.
b Including family labour at Rs 10 per day.

c Assuming an opportunity cost of 10 per cent per annum.
d In distinction to
Table 7.4, it is assumed here that the farm has a debt of Rs 10 000 at an interest cost
of 10 per cent per annum.
Return on equity is calculated as:
[(Total net returns - Cost of borrowed capital)/Equity capital]100 = [(8 350 - 1 000)/96 600] 100 = 7.61%.
Reference was made in Section 7.2.3 to the possibility of estimating a return to the farm's management
resource based on the residual-value method. A preferable evaluation of management, however, is
provided by the measures of total factor productivity, return on capital and return on equity. Other
things being equal, the larger these indices, the better the performance of the farm's management.
Obviously, these indices are also useful in inter-farm comparisons. It should be recognized, however,
that these total productivity measures are not fully under the control of the farm's management. They
can be heavily influenced by, in particular, the vagaries of Nature (affecting product yields) and the
market (affecting product prices), as well as by other sources of risk (as outlined in Chapter 11).
Activity-specific total productivity measures
Analogously to the case for the whole farm, total factor productivity, return on capital and return on
equity can be estimated on an activity-specific basis. This is exemplified by the worksheet of Table
7.9 for the paddy activity of Table 7.4. Such estimates are again useful for inter-farm comparisons of
like activities but they are rather qualified. First, they take no account of those farm resources which,
while crucial to the operation of the farm, cannot be allocated to any specific activity. For the
example farm these are listed in the 'farm-as-a-whole' column of Table 7.4. Note, however, that in
contrast to Table 7.4, the analysis of Table 7.9 allows for an interest cost on the capital value (Rs 44
642) of the 2.5 ha of land used for paddy. Second, estimation on an activity basis may be distorted by
double counting. Thus, while the example farm has a total area of 2.8 ha, through double cropping of
at least 0.7 ha with paddy (2.5 ha) and maize (1 ha) it appears to use 3.5 ha of land. This double
cropping is not allowed for in the worksheet calculations of Tables 7.8 and 7.9.

7.3.6 Summary
As noted above, comparative analysis of a subject farm with other relevant farms is intended to
identify weaknesses which might exist within a farm-household system and to diagnose their causes,
i.e., identify causal factors. These might arise at any level within the farm system (Figure 1.2). A full
and formal comparative analysis would follow the sequence of steps I to V listed in Table 7.10. This
breaks the problem of diagnosis down into successively smaller but more numerous 'bits' for
successively more detailed investigation, in much the same way as a mechanic trouble-shoots a
recalcitrant engine. These bits are the subsystems of the farm-household system. In practice, some of
these levels of analysis can usually be by-passed since the analyst will often have an idea of where
the problem lies and will proceed directly to that level, e.g., directly to an examination of the kind of
technologies used on the farm at level V, rather than systematically working through analysis at levels
I, II, III and IV.
Analysis at this level is concerned with developing measures of performance of the whole-farm
system. Some of these measures, those relating to financial performance, were developed as factors
E, F,... J in Section 7.2.1. But to the extent that the farm will also have other operating objectives,
some of the other performance criteria of Section 6.2 will apply. The next step in level I analysis is to
compare the subject farm with other relevant farms, or with some standard farm (as discussed in
Section 7.4.2), in terms of these selected performance criteria, as in Table 7.11 (which uses a
different subject farm to that of Section 7.2). Sources of data for such whole-farm comparisons are
discussed in Section 7.4.
TABLE 7.9 - Worksheet for deriving Total Factor Productivity, Return on Capital and Return
on Equity on an Activity-specific (Annual) Basisa
Paddy Activity
Measure on an Activity Basis
(Rs) (Rs)
Gross Return 31 000
Costs
Directb 8 000
Fixed 970
Depreciation 500
Interest on capitalc
Land 4 464
Other 1 330
Total Costs 15 264
Net Return 15 736
Total Factor Productivity
Gross 2.03
Net 1.03
Capital 57 943
Return on Capital 27.2%
Equity Capital 52 508
Borrowed Capitald 5 435
Cost of Borrowed Capitale 543
Return on Equity 28.9%
a Data, as required, from Tables 7.1, 7.4 and 7.5.
b Including family labour at Rs 10 per day.

c Assuming an opportunity cost of 10 per cent per annum.
d Prorata share of loan of Rs 10 000 based on paddy activity's share of total farm capital, i.e., Rs (10
000) (57 943/106 600) = Rs 5 435.
e At an interest charge of 10 per cent per annum.
From Table 7.11, relative to the comparative farm, the subject farm is performing badly in terms of
productivity and profitability, but well in terms of the stability, diversity and time dispersion of
income, and of sustainability. In following discussion, it will be assumed that profitability, i.e., net
farm income, is the operating objective of interest, thus the subject farm's low comparative
profitability levels revealed in Table 7.11 warrant closer examination.
TABLE 7.10 - Broad Sequence of Steps in Comparative Analysis
Stage or Level in Subsystema Action

Comparative Analysis Focusb
I Whole-farm (Older Level Develop measures of whole-farm

10) performance, as in Section 7.2.
II Household (Order Level 11) Look for causal factors (of I) in the
household.
III Service Matrix (Order Look for causal factors (of I) in the
Level 9) amounts/kinds of fixed farm capital used.
IV Enterprises and Activities Look for causal factors (of I) arising in the
(Order Levels 3 to 7) mix of activities.
V Processes (Order Levels 1 Look for causal factors (of IV) arising:
and 2) (i) in the technology used in each process.
(ii) in the level of resources used in each
process.
a Relative to the farm-household system of Order Level 12.
b Order Levels as specified in Section 1.3 and Figure 1.2.

Possible causal factors to look for within the household can only be suggested: they will vary with
farm type and cultural environment. They fall into two general groups: factors which have a direct
effect on farm performance and those that might have only an indirect effect (broadly, sociological
non-economic factors). Some of these can be quantified and this will allow comparison with other
standard farms. Others have only an implicit effect on farm performance and do not permit
quantitative comparisons. A few are listed in Table 7.12. These and similar household factors would
be arranged for inter-farm comparisons as in Table 7.11. Only two of the factors of Table 7.12 - (7)
health and (8) debt - might require elaboration.
The importance of family health as a factor underlying farm performance is often overlooked,
especially by analysts removed from the village who take the absence of malaria, cholera, intestinal
diseases, trachoma, etc. for granted. But in many areas - the limestone hills of Java, southern Sri
Lanka, South India, the terai of Nepal, the African lakes - comparative analysis aimed at uncovering
reasons why farms and indeed whole villages are 'inefficient' need often proceed no further than the
factor of poor family health.
Household debt as a causal factor requires more careful interpretation. If incurred for food, debt may
be taken as an indicator of poverty and of the inability of the subject farm to provide family
subsistence. But where it occurs in comparatively rich areas such as the Kandy hills of Sri Lanka, it is
as likely as not incurred for frivolous purposes. Equally, household and social debt might often be
incurred to acquire status-giving possessions or to achieve a higher social standing. Whatever its
origin, the pressures for debt repayment often offer good insights into why a quick-growing but soil-
erosive and less profitable crop has been grown in preference to a more profitable and less erosive
crop; and (as often occurs with clove crops in Sri Lanka and paddy crops in some Central Java
villages) why a crop has been sold in its standing state at a fraction of the price that the household
could have received if they had waited a little longer and harvested the crop themselves. In short, the
occurrence of debt in the household component of a system, or more precisely the pressures for its
liquidation, can sometimes lead to apparently bizarre decisions within the farm component.
TABLE 7.11 - Level I Analysis: Comparative Analysis with Some Whole-farm Performance
Criteriaa
Performance Criteria Unit Subject Farm Comparative Farm
(1) Productivity (e.g., of staple food grain):
total farm t 7 10
per ha t 2.8 5
per family labour day kg 25 40
(2) Profitability:
total farm gross margin Rs 25 000 45 000
per ha Rs 10 000 22 500
per Rs 1 000 of fixed capital Rs 500 750
per family labour day Rs 89 180
(3) Stability of annual income
Coefficient of variation (CV) % 12 36
(4) Income diversity:
number of activities 6 4
number of separate products 14 8
(5) Flexibilityb: fair fair
(6) Time dispersion of income:
relative time-dispersion (RTD) 0.9 0.4
(7) System sustainabilityb: high low
(8) Environmental compatibilityb: fair fair

a Criteria are as per Section 6.2.
b Assessed subjectively in terms of poor/fair/good for criteria (5) and (8) or low/average/high from
an overall (i.e., private and social, economic and resource) perspective for criterion (7).
The factors of Table 7.12 might afford diagnosis of a problem, but it is difficult at this level for the
farm management analyst, as analyst, to prescribe solutions. Only a few of the factors, e.g., attitudes
to and management of debt, might be amenable to direct adjustment. Most of the factors will either be
given (household and labour-force structure and age) or they will exist as a consequence of higher-
order (social) systems in such fields as literacy, health, education. However, if the analyst happens to
be an extension agent or rural development officer (as he or she is likely to be) then he or she can take
action: e.g., ensure that health services are delivered to the village, plan a local literacy campaign,
develop a relevant farm-and-home extension package.
TABLE 7.12 - Level II Analysis: Indicative Aspects of the Household to be considered in
Diagnosis and Comparative Analysis
(1) Number of an indication of total needs for food and/or cash, thus of demands made on the
household system, thus of the kinds of farm activities which will be selected
members
(2) Income per a direct quantitative measure of well-being

household
member
(3) Labour force a direct quantitative measure of available labour resources
(4) Labour force per a better measure than (3) and an indication of the potential for further farm
ha adjustment, intensification
(5) Age structure of a qualitative measure of the labour resource

workforce
(6) Literacy of all another qualitative measure of labour resource and management potential, as
farm adults well as an indicator of the potential of the household to absorb new farming
ideas
(7) Health status another qualitative measure of the labour resource, but more importantly an
indicator of general household morale
(8) Household debt for
(a) essential food, subsistence
(b) household items
(c) social purposes
(d) farm operation or development
(9) Indicators of (a) school attendance

progress,
development (b) participation in non-agricultural activities
(c) contact with extension service
(d) possession of safe water supply, sewing machine, bicycle, radio, etc.
Assuming that the specific causal factors cannot be diagnosed or removed at this second level, the
analysis moves on to level III.
Analysis at level III is concerned with the farm service matrix - fixed-capital investment in land,
buildings, irrigation channels, institutional arrangements etc. that serves the whole farm. Again a large
number of possible causal factors of poor farm performance might be found within this capital stock;
the art is to concentrate on those capital items that are intuitively most relevant. In practice, it will be
necessary to consider the most appropriate measure of each item: for some the best measure is their
size/quantity/capacity; for others their capital value; others might be measured as 'poor/fair/good'.
An example of comparative analysis applied to the farm service matrix is presented in Table 7.13. It
relates to the same farms as in Table 7.11. Might any of the factors listed in Table 7.13 account for
the subject farm's poor performance with a total gross margin of only Rs 25 000 vs Rs 45 000 on the
comparative standard farm? Looking first at item (1) of Table 7.13, both farms have a roughly similar
capital structure. However, looking at item (2), the subject farm has 50 per cent more good land than
the comparative farm; but in spite of this it produced much less income, so clearly its poor
performance is not due to insufficient or poor quality land. In item (3), irrigation investments are
approximately similar, but the irrigation system on the subject farm covers only 40 per cent of the
farm area while that on the comparative farm covers 90 per cent. Inadequate irrigation could be the
first causal factor in low income.
TABLE 7.13 - Level III Analysis: Aspects of Fixed Farm Capital to be considered in Diagnosis
and Comparative Analysis
Capital Item Unit Subject Farm Comparative Farm
(1) Fixed capital value (except land):
total Rs 50 000 60 000
per ha Rs 20 000 30 000
per labour unit (worker) Rs (3) 16 700 (4) 15 000
(2) Land:
total ha 2.5 2
good paddy land ha 1.5 1
poor crop land ha 1 1
(3) Irrigation system:
total value Rs 15 000 13 000
value per ha (of farm) Rs 6 000 6 500
coverage ha 1 1.8
condition (poor/fair/good) good fair
water supply mths 7 6
(4) All field equipment:

total value Rs 25 000 20 000
value per ha Rs 10 000 10 000
(5) Work oxen:
number no. 3 2
number per ha no. 1.2 1
(6) Grain-drying facilities: Rs 0 4 000
(7) Grain storage: Rs 1 000 5 000

Continuing, levels of investment in field equipment and oxen are apparently not causal factors. But the
next item could be - there is no grain drying facility on the subject farm. How then does the farmer dry
his or her paddy and corn? This possible system weakness is partly confirmed by the next item - only
Rs 1 000 are invested in crop storage compared with Rs 5 000 on the comparative farm. These
factors provide sufficient reason for proceeding to analysis at level IV relating to the economics and
the technical structures of individual activities.
As per Table 7.10, level IV analysis concerns the individual enterprises/activities of Order Levels 3
to 7. (In doing so it also implicitly considers the farm resource pool of Order Level 8.) At this stage,
as exemplified in Table 7.14, two types of comparison are required: comparisons (a) between
activities on the subject farm, and (b) between pairs of similar activities on the subject and
comparative farm. Both types of comparison have two objectives: first, to evaluate the relative
performance of each activity in the system with a view to possibly changing the activity mix, i.e.,
replacing inefficient activities with efficient ones; and second, to identify specific agro-technical-
economic factors within each activity which warrant closer investigation at the next stage, level V. (In
actual practice, analysis at levels IV and V is often combined.)
The first step is construction of activity unit budgets (although, as noted previously, these are better
termed operating statements because they refer to past events rather than to future projections).
Obviously the more detailed these budgets/statements are, the more effective they will be in
identifying causal factors. Detailed budget analysis may be undertaken, as in Table 7.14, or only
summary unit budgets may be developed as shown at the top of the table. These summaries are
sufficient to meet the first objective of intra-farm comparison noted above. Considering the budget
results for GM per ha, the presence of the relatively unprofitable roselle activity offers one reason
why total income is so low on the subject farm. The indicated action would be to eliminate roselle
from the plan and increase the size of the more profitable paddy or maize activities. (Even though
paddy on the subject farm is poor relative to the standard farm, it is still twice as profitable as
roselle.) However, from Section 7.3.3. irrigation shortage would probably prevent expansion of these
high water-using paddy and maize crops. The cropping system followed on the more profitable
comparative farm offers another possibility: replace roselle with oilseed and thus increase farm TGM
by Rs (3 500 - 2 000) or Rs 1 500 per ha. The analyst would, of course, also consider other low
water-using crops as possible substitutes. If confident that the problem of low farm income could be
solved by substituting better activities, he or she would at this point begin restructuring the system
(Chapter 9). If not, the analyst would proceed to the second objective of level IV analysis, i.e., inter-
farm comparison, and derive further comparative factors from the activity budgets which might permit
identification of specific weaknesses within the three existing activities.
Some of the possible derived factors are listed in Table 7.14. From the budgets there are apparently
no problems with the maize activity, and the possibility of removing roselle entirely has been noted;
attention would therefore be concentrated on paddy, specifically in relation to the lower price
received (Rs 1 500 vs Rs 1 700), the lower net recovered yield (four vs five tonnes) and the higher
direct costs, all of which result in a much lower GM for paddy than on the comparative farm. Also, in
the prior level III analysis there were indications that poor post-harvest and irrigation practices might
be present, thus the first factors to consider at level IV are those which might verify or remove these
suspicions.
From Table 7.14, inadequate irrigation is obviously an important factor in low paddy yield. But
assuming that water supply is limited, there is little the farmer can do about it - except reduce his or
her paddy area and substitute more dryland crops.
The next item, grain-drying method, explains why there was no capital investment in this item in
Table 7.13: the farmer spreads the paddy out to dry on the village road (where inevitably there would
be a 15 to 20 per cent loss to the village chickens and ducks). This partly explains the low net sales
of only four tonnes per ha. The grain also becomes contaminated to some degree - which probably
explains the relatively low sale price received in comparison with the standard farm.
TABLE 7.14 - Level IV Analysis: Example of Comparative Analysis of Individual Activities
Subject Farm Comparative Farm
Paddy Maize Roselle Paddy Maize Oilseed
Unit Budget
Net sales (t/ha) 4 3 2 5 3.3 2
Sale price (Rs/t) 1 500 1 500 2 000 1 700 1 500 2 500
Yield value (Rs/ha) 6 000 4 500 4 000 8 500 4 950 5 000
Direct costs (Rs/ha) 2 000 1 500 2 000 1 500 2 000 1 500
GM (Rs/ha) 4 000 3 000 2 000 7 000 2 950 3 500
Factors Derived from Budget
Irrigation water (units per ha) 6 naa na 10 na na
Grain-drying method road road fence slab slab slab
Grain storage shed shed shed silo silo silo

Family labour (days/ha) 50 30 40 50 35 30
Hired labour (days/ha) 40 0 10 10 0 0
Humus applied (t/ha) 6 0 0 6 0 0
N fertilizer (kg/ha) 150 50 40 150 40 30
Agricides (Rs/ha) 50 0 50 0 0 0
Oxen (days/ha) 5 3 4 10 4 4
a na: not applicable to these rainfed crops.
Likewise, the next item, storage, explains the low capital investment reported for this item in Table
7.13. Paddy is stored in an open mud-floor thatch-roofed shed whereas on the comparative farm it is
stored in a rodent-proof silo. These two practices, roadway drying and insecure storage, could easily
result in a loss of 30 to 45 per cent of harvested grain and thus go a long way to explaining the
relatively low farm TGM of Rs 25 000 vs Rs 45 000 on the standard farm (Table 7.11).
The next item, crop labour requirements, also looks a promising lead. Noting that direct paddy costs
of the two farms are Rs 2 000 vs Rs 1 500, this extra cost on the subject farm may be explained
largely in terms of hired labour (30 days x Rs 10). But why was it necessary to employ so much more
labour than on the comparative farm?
The last item listed, ox days used per ha, could also be significant. The subject farm used only half as
much oxpower inputs for paddy as the comparative farm. Why the large difference? What actual
operations were performed with these oxen?
To find the answers to these questions, the analysis now moves to the next level and looks at specific
agro-technical processes which underlie these paddy activities.
'Planting rice is never fun;
Bent from mom till set of sun;
Cannot stand and cannot sit;
Cannot rest for a little bit.'
From an old Filipino song.
Analysis to this level has broken the search for causal factors down into successively smaller 'bits' or
subsystems of the farm-household system. It now becomes a search within the lowest subsystems, the
agro-technical processes of Order Levels 1 and 2. Causal factors here might be found in either of the
components of a process, the intensity levels at which they are used or the ways - technologies - by
which they are applied (Section 5.1).
From the level IV analysis, the problems within the paddy activity may be summarized as:
(i) (implied) low field yield.
(ii) (implied) high wastage in drying and storage.
(iii) low net (recovered/sold) production.
(iv) low sale price (probably due to contamination).
(v) high direct costs (mainly due to high labour inputs).
Accordingly, the causal factors considered in level V analysis will concentrate on these five areas.
The analyst would list each step or operation in the activity from land preparation to sale of the crop
and, within each operation, list processes which might be relevant in relation to the problematic
performance factors (i) to (v) above, as shown in Table 7.15 on a per ha basis.
Factor (1) of Table 7.15 now explains why ox-day inputs on the subject farm were so low in Table
7.14: the farmer performs only the minimum tillage operations in comparison with the more intensive
practice followed on the standard farm. This factor has two effects: a direct effect on tilth and thus on
yield, and an indirect effect through weed reduction both on yield and on subsequent hand-weeding
requirements (and thus on labour and direct costs). Factors (1), (4) and (6) are closely related. These
factors imply a process. Since the technology, i.e., use of oxen, cannot be changed, attention would
concentrate on the intensity component: ox-day inputs or number of times the field is initially
cultivated vs yield and subsequent labour costs for weeding. The kind of response and cost
relationships which are probably present are indicated in Figure 7.2. For optimization of this
response, the separate relationships shown - number of cultivations vs yield, cost of weeding labour
and cost of oxpower - could be consolidated into a single 'response curve' showing net return in
relation to number of cultivations. (Ox days would bear no cost unless they were hired.)
The point in introducing simple response analysis into the inter-farm comparison is that, while
recommendations for action on the subject farm might be based on what the comparative farm is
doing, it is possible that both farms are suboptimal in regard to factors (1) to (6) of Table 7.15. Both
may benefit from response analysis of at least the main factors, although probably the subject farm
more so than the standard farm.
TABLE 7.15 - Level V Analysis: Example of Comparative Analysis of Processes
Process Factors relating to Paddy Subject Farm Comparative Farm
(1) Land preparation:
ox days/ha 5 10
plow, times 2 2
harrow, times 1 3
(2) Seed used:
type HYV HYV
quantity, units/ha 5 5
days in nursery 30 23
(3) Transplanting:
spacing pattern random square

distance apart, cm 20-30 25
method 'eye' pegboard
(4) Weeding:
hand weed, times 3 1
hired labour, days/ha 28 0
(5) Inter-row cultivation:
method none sorok
number 0 2
(6) Grain:
harvested, t/ha 4.5 5
lost, t/ha 0.5 0
sold, t/ha 4.0 5

Moving to factor (2) in Table 7.15, both farms are similar in regard to variety and amount of seed
used, but dissimilar in the age at which seedlings are transplanted to the field. This latter factor again
suggests the possibility of process optimization through response analysis of rice yield vs seedling
age or days in nursery, as suggested in Figure 7.3. Assuming the response of yield to seedling age is
as depicted in Figure 7.3, and given that the cost per day of holding seedlings in the nursery is
virtually zero, the optimal age of seedlings for transplanting is about 25 days, i.e., where rice yield is
a maximum. This is five days less than the seedling age of 30 days used on the subject farm which
clearly leads to a suboptimal yield.
The transplanting factors listed in Table 7.15 also indicate a significant difference between the farms
in terms of the transplant method used. This technology component is probably more important than
the planting-intensity component (which could if necessary be measured in terms of seedling spacing
or population). The technologies can be directly compared by partial budgeting (Section 4.5) of the
labour and cost of random planting vs the labour and cost of purposeful planting based on guides
provided by a pegboard or knotted rope. This transplanting factor is also closely associated with
others, particularly (4) and (5). Although random spacing on the subject farm might possibly be
cheaper than peg planting, it leads, in factor (5), to an inability to control weeds quickly and with
minimum labour by use of a 'sorok' (a light weeding and cultivating wheel which is pushed along by
hand between the paddy rows). In turn this inability leads in factor (4) to the necessity to hand-weed
and thus hire 28 days of labour, increasing direct costs on the subject farm by Rs 280. As noted
above, the need for more weeding on the subject farm, by whatever means, is probably due to
inadequate initial land preparation.
FIGURE 7.2 - Yield, Hand-weeding Cost and Oxpower Cost in relation to Number of
Cultivations
Factor (6) provides more detail regarding post-harvest operations. Losses here are serious on the
subject farm: 0.5 tonne is lost in the roadway drying and storage operations. There are several ways
by which these operations could be improved through response analysis; e.g., in drying, quantity of
grain lost vs the number of children employed to keep the ducks away; and in storage, alternative
levels of investment to give alternative levels of security. But obviously, with factors and processes
of this type, the alternative technologies are more important than their intensity levels. This requires
only comparative partial budgeting: in drying, the cost of a slab drying-floor and low grain losses vs
free drying on the road with high losses; in storage, the cost of one or two alternative types of new
storage and no losses vs continued free use of the open shed with high grain losses.
FIGURE 7.3 - Response of Rice Yield to Seedling Age at Transplanting
7.3.6 Summary
This concludes the illustrative comparative analysis of the farm-household system of the subject farm
considered in this section. The analysis has diagnosed problems at five system Order Levels as per
the schema of Table 7.10 and offered prescriptions for their solution:
(i) From level I (whole-farm system), there is a need to increase the farm's profitability.
(ii) From level II (the household system), the family should be encouraged to participate more
actively in extension activities, particularly those relating to the main crop (paddy). But equally this
recommendation might also be directed at the extension service: it should strengthen its efforts in
providing an accessible and relevant extension package for under-performing farmers. (The content of
this package will be clear from this summary.)
(iii) From level III (farm capital structure), the farm should increase investment in grain drying and
storage.
(iv) From level IV (farm activities), weak activities such as roselle should be replaced by more
profitable activities and it might also pay to concentrate the limited available water on a smaller area
of paddy.
(v) From level V (processes), the indicated action in restructuring the paddy activity is to instigate
better land preparation, earlier transplanting of seedlings, purposeful rather than random planting and
weed control by sorok rather than by hand.
Expansion to Order Level 13 systems
The above discussion of comparative analysis has been concerned with adjustments and
improvements on individual farms. In Asian (and Pacific and African) villages, farm systems are
markedly homogeneous with respect to the types of crops grown, types and levels of resources used,
technologies employed, and technical and economic problems encountered. If production is efficient
(or inefficient) on one farm, it is likely to be efficient (or inefficient) on most other farms in the
village. Further, and in spite of the previous discussion at levels IV and V regarding restructuring the
activity mix and adjusting the underlying processes on the subject farm, it is not easy to overturn
village conventions and traditions. If farm development on individual farms is to occur, it must often
be based on village consensus. Thus it will often be more productive to apply comparative analysis
to the 'farm-village' system, working with representative or mean or model farm data, rather than with
specific individual farms. With a few obvious modifications, the same step-by-step procedures
outlined above can be applied to removing many of the weaknesses which might occur in Order Level
13 village systems, which would of course achieve the same goals as analysis aimed at individual
farms.
7.4 DATA SOURCES

Data required for limited evaluation (Section 7.2) of the farm component of a farm-house could
system and for comparative analysis of the whole system will vary in degree of necessary detail.
Since limited evaluation concerns the farm as a whole, but not the household component (Figure 7.1),
the data for this may be of an aggregative nature. The degree of required detail increases if the
examination then proceeds in prescriptive mode to a comparative analysis of successively lower
Order Level systems.
There are four main sources of data relating to a subject farm:
· written records maintained by the household.
· intensive interviews with members of the household.

· records maintained for the farm component by a club, cooperative or other farm association.
· records maintained for the farm and/or its household components by an extension program or other
agency specially designated for this purpose.
Very few if any farms of Types 1, 2 or 3 keep any sort of useful farm or household records. Type 4
farms might maintain records relating to their specialist activities, but seldom relating to the
household component. Large commercial farms and estates of Types 5 and 6 usually keep good
records but often not of the degree of detail required for lower Order Level comparative analysis.
Thus even if farm/estate records are available they usually must be supplemented by personal
interviews. These will usually yield satisfactory results when applied on large farms and estates
(where incidentally the household component will be of less importance or entirely absent) but on
small family farms they have three limitations.
First, since the data will be based on memory recall, they will contain inaccuracies. These increase
in rough proportion to the period of time which has elapsed since the occurrence of events and the
degree of complexity or number of activities in the system. Second, successful interviews require a
considerable degree of technical and economic understanding, interpersonal skills and sympathy on
the part of the interviewer - a combination of attributes often difficult to find within a single
individual. Third and most serious, it takes time and patience to gain the confidence of the household
regarding its farm, and even more time, observation and patience to gain an insight into those
household factors which dominate the farm. One can forget about Oxford PhDs: the best interviewers
are usually women with a village background who speak the local language.
Records relating to some aspects of the farm component might be maintained by a farmer club,
cooperative or association. But these will often relate only to the most important farm activity (a club
keeping records of inputs supplied to and produce received from its member maize growers; similar
records kept by an export company on behalf of sugar or banana growers). These will in any case
relate only to the farm and not to the household component.
Best of all are data obtained from a continuing small-scale but intensive survey (usually by a
government extension or farm management service) of both the farm and household components over
time. The Pakistan and Bangladesh panel surveys of the 1960s still stand as models. They attached
one enumerator to each cluster of 20 to 25 farm households, with one cluster in each of the main
farming regions. Each household was visited at least once every 10 days when farming and relevant
household events were recorded (for often illiterate families) more or less as they occurred. Equally
important, these panel surveys covered household processing of farm outputs and their marketing or
other final disposal as well as non-farm economic activities and household consumption and
expenditure. They were successful largely because the enumerators were given time - about a year -
to first win the trust of initially suspicious villagers. As ever, however, the cost of such data
collection must be assessed against its benefits. These benefits can arise not only through comparative
analysis but also through use of the data in policy analysis and formulation.
As outlined by Dillon and Hardaker (1993, Ch. 2), Collinson (1983, Part II) and Upton (1973, Chs 10
and 11; 1987, Ch. 11), standards against which the performance of a subject farm can be measured
might be obtained from such sources as:
· an actual 'high-performing' farm.
· an actual farm subjectively or objectively chosen as being representative of some relevant group of
farms.
· the mean or median performance data of a group of similar actual farms.
· the performance of a model or demonstration farm which has been established (by government) for
this purpose.
· a synthetic 'farm' which has been fabricated by the analyst from estimates by research workers
regarding how an actual farm should function if the best mix of activities were to be operated and -
within practical limits - the best production technologies employed.
· census-type surveys.
Data for comparisons from a 'high-performing' farm require little discussion, except perhaps to note
the problem of selecting a farm which is good but not too good. The same problem exists in selecting
a representative farm.
The use of model actual farms has declined. They were fairly widely used in the 1950s and 1960s,
primarily for testing research results, as an extension vehicle and as a captive data source. They are
generally not effective as an extension vehicle because farmers see little relationship between what
governments can achieve with unlimited resources on a model farm and what resource-poor farmers
can hope for at village level. The same weakness exists in their data-generating role: data generated
within a 9-to-5 civil-service structure is often not applicable to real farms. In any case, model-farm
data cannot be generated for the household.
Use of synthetic data offers an alternative. These can be selected to refer to both the farm and
household - which latter will require information from other than agricultural departments.
Concerning agricultural performance standards, the problem of defining appropriate realistic
standards and recommendations again arises. It is more than a technical one. Farm development
strategy is not infrequently set and standards adopted, perhaps unconsciously, which serve the special
interests of the local bureaucracy or research industry. They might also reflect the ambitions of
purveyors of (elsewhere unwanted) agricides and of agronomic miracles, the passions of
mechanization experts or the innocence of sectoral economists. In most of South Asia the farm
performance standards which these things imply are only dreams; they are not aspirations to which the
people may reasonably aspire.
Census-type surveys as a means of obtaining farm data are unfortunately common. In some small
countries they can run to the attempted coverage of up to 2 500 farms. Apart from excessive cost, they
often lack clarity in purpose and precision in design. They might be based on once-only interviews
(which fail for reasons discussed above), or purport to be continuing surveys in that the effort is
planned to be repeated over several years. They are the most expensive but least effective of all
methods of obtaining detailed data for farm and household systems analysis. So voluminous is their
output that seldom is more than 10 per cent of it ever used - but, given their typical lack of reliability,
this must usually be counted a good thing!
Sources of data for the evaluation and adjustment of agro-technical processes by response analysis
are discussed in Chapter 8.

Three difficulties may arise in the evaluation and comparative analysis of a farm system. First,
relevant and reliable data need to be obtained. This difficulty was discussed in Section 7.4 above. It
may be compounded, however, by the second difficulty. This is the degree to which a farm is
subsistence rather than market oriented. To the extent that a farm system is subsistence oriented (as
may be expected particularly of Type 1 farms but also for farms of Types 2 and 3), evaluation may
have to be conducted mainly in terms of physical inputs and outputs, supplemented where possible by
financial analysis based on the use of opportunity or market costs and prices (Dillon and Hardaker
1993, pp. 80-82; Makeham and Malcolm 1986, p.65). Such financial analysis will obviously be less
reliable the greater the subsistence orientation of the farm system and the greater its isolation from the
market. In the extreme, for a purely subsistence system, the focus of analysis may need to be in terms
of family labour effort and staple food production in relation to household nutrition.
The third difficulty relates to a fallacy in comparative analysis of resource productivities. It is a
difficulty of which the analyst must be aware and aim to avoid.
The potential fallacy in comparative analysis
Comparative analysis has proved itself in the field as a very useful guide to the improvement of farm
systems (Upton 1973, pp. 234-243). It must be used with caution, however, as a guide to the mix of
resources that might best be used. This is because, as shown by Candler and Sargent (1962),
deductions about improved resource use made on the basis of comparative analysis may be quite
erroneous. The reason for this is illustrated in Figure 7.4.
Suppose that for two variable inputs X1 and X2 used in producing a product Y, the locus of all
combinations of X1 and X2 which will produce the fixed level of output Y* is shown by the ellipse of
Figure 7.4. This locus is known as the isoquant for Y = Y*. In terms of the symbolism introduced in
Section 5.1.1, the isoquant is represented algebraically by the function X1 = h(X2/Y*) and is derived
from the production function Y = f(X1, X2) by fixing Y at the level Y*.
From Figure 7.4 it can be seen that use of X1 or X2 at levels beyond OC or OF, respectively, is
physically inefficient; Y* units of output can always be produced with a lesser quantity of X1 or X2
lying on the isoquant segment AGB. Just where on this segment will be the economically most
efficient (i.e., profit-maximizing) combination of X1 and X2 to produce Y* will depend on their
relative prices. It will be at the point where the (clearly negative) slope of the segment AB (given by
the ratio D X1/D X2 where D Xi denotes a small change in Xi) is equal to the negative of the inputs'
inverse price ratio p2/p1, i.e., where
D X1/D X2 = -p2/p1
so that
-p1(D X1) = p2(D X2)
which implies that X1 and X2 are in economic balance in the sense that, at this point, the positive
saving in cost of -p1(D X1) from a small decrease (i.e., D X1 <0) in the use of X1 is exactly offset by
the increase in cost of p2(D X2) from the small increase (i.e., D X2 >0) in the use of X2 needed to
maintain output at the fixed level Y'. This profit-maximizing combination of X1 and X2 could be, say,
at the point G on AGB.
Now suppose that in comparative analysis the subject farm is at point A and the comparative farm at
point B of Figure 7.4 in terms of their use of X1 and X2. At A, X1 has relatively low productivity of
Y*/OC and X2 has relatively high productivity of Y*/OE while at B these productivities are reversed.
Based on comparative analysis of the productivity of X1, an analyst might suggest that the subject farm
should reduce its use of X1 from OC to OD and increase its use of X2 from OE to OF so as to have a
resource allocation pattern like the comparative farm. This, however, would be a mistake. For profit
maximization, both the subject farm and the comparative farm should operate at point G, not at A or B
respectively. Thus the productivity of individual resources considered one at a time may lead to
fallacious conclusions from comparative analysis.
FIGURE 7.4 - Isoquant showing Combinations of Two Variable Inputs X1 and X2 to Produce a
Fixed Level of Output Y*
7.6 REFERENCES
Candler, W. and D. Sargent (1962). 'Farm Standards and the Theory of Production Economies',
Journal of Agricultural Economics 15(2): 283-290.
Collinson, M. (1983). Farm Management in Peasant Agriculture, Westview Press, Boulder.
Upton, M. (1973). Farm Management in Africa, Oxford University Press, London.
Upton, M. (1987). African Farm Management, Cambridge University Press.
8. OPTIMIZATION OF RESOURCE USE

LEVELS: RESPONSE ANALYSIS
8.6 REFERENCES
'But for the law of diminishing returns, it would be possible to grow all of the world's grain needs in a bucket.'
Earl O. Heady (1916 - 1987)
From Chapter 7, the two main weaknesses that may be uncovered by farm management analysis in
diagnostic Mode are, first, those arising from the use of suboptimal levels of resources within
individual activities and, second, inefficient combinations of activities. Analogously, in analysis in
prescriptive Mode (system planning), the two main tasks of the farm management analyst are to define
the best levels of resources to use in potential future activities and the best mix of such activities.
Optimization of the activity mix is discussed in Chapter 9. This chapter is concerned with the first
problem, i.e., determining the best levels of resources within activities.
For either restructuring weak activities or formulating new ones, the problem reduces to constructing
an input-output activity budget. However, unlike a budget which simply describes the actual resource
composition of a past or existing activity, or what a farmer has done, this planning or adjustment
budget now specifies what the farmer should do in terms of allocating resources to the activity.
Specifying the 'best' levels of resources to use in respective activities will usually present few
problems. Farmers will base their estimates of the best seeding rates, irrigation frequencies, fertilizer
rates etc. on their own experience or village practice or local lore or, in a few cases, on reference to
farm records. For new inputs with which they are not familiar, the estimates might be based on the
(sometimes dubious) advice of fertilizer or agricide agents or on the more objective
recommendations of extension workers.
These sources of data will usually be a sufficient basis for making at least reasonable estimates of the
best levels of resources to use in individual activities. However, in a few cases it will be necessary
to take a more formal approach to determine the best levels of resource use in an activity, viz.:
(i) When the cost of an input and/or the value of its effect on output is economically significant, i.e., sufficiently great as to
warrant the cost of obtaining more precise information.
(ii) When the likely effect of an input is quite unknown and the farmer (or more likely an analyst or
extension worker on behalf of a group of farmers) sees the need for a small-scale field trial before
deciding on the use of the input.
(iii) If sharp or significant changes occur in what have been stable input unit costs and/or output
prices, resulting in the likelihood that past levels of resource use should now be changed.
(iv) When entirely new activities or technologies are introduced to an area so that new activity
budgets must be constructed.
(v) On mono-crop farms and estates where more intensive effort will routinely be devoted to input-
output relationships which, for these same activities on small mixed farms, might be of only marginal
interest and where rough approximations are good enough.
In these situations the more formal methods of response analysis might be required to determine
optimal input levels. An introduction to such methods is given here. A more extensive introduction is
provided by Amir and Knipscheer (1989, Ch. 3) and Dillon and Hardaker (1993, Ch. 7) while a
relatively comprehensive formal presentation is to be found in, e.g., Beattie and Taylor (1985),
Debertin (1986), Dillon and Anderson (1990), Doll and Orazem (1984, Chs 2, 3, 4 and 5) and Rae
(1977, Chs 2, 3 and 4). A variety of applications of response analysis is to be found in Heady and
Dillon (1961), Heady and Bhide (1984) and Hexem and Heady (1978).

Agricultural input-output relationships are, in themselves, of a physical nature. Their manipulation
and (so far as possible) control by the farmer so as to achieve his or her goals, however, implies that
these physical relationships have to be evaluated in value terms as determined by the market price or
opportunity cost of outputs and inputs.
Physical relationships in response analysis
An illustration of the physical response curve or production function showing total physical product
(TPP) for a single input/single output situation is given in Figure 8.1. Note that, by convention and for
convenience, the symbols Y and X respectively denote both a particular type of output or input and the
quantity or level of that output or input per technical unit of production such as per ha of land or
other fixed resource (e.g., per animal unit or per tree). Two further physical relationships can be
derived from TPP. These are average physical product (APP) and marginal physical product (MPP),
both of these being measured on the same technical-unit basis as TPP, These three physical measures
are defined as follows:
TPP: Total physical product is the yield or response of output Y in physical units to inputs of X, also in physical units. Thus TPP = Y =
f(X) where f(X) denotes the relevant production or response function as outlined in Section 5.1.1.
APP: Average physical product is the amount of output achieved per unit of input. Thus APP =
TPP/X = Y/X.
MPP: Marginal physical product is the change in TPP in response to a marginal unit change in input.
Thus MPP = D TPP/D X = dTPP/dX = dY/dX where the symbol D (.) denotes a small increment in (.)
and dY/dX is the first derivative of Y = f(X).
While the TPP and MPP curves are illustrated in Figure 8.1, the APP curve is not shown. However,
note that APP for any point on the TPP curve is given by the slope of a radius from the origin to the
point on the TPP curve since such a radius has slope of Y/X = APP.
As drawn in Figure 8.1, the TPP curve reflects the law of diminishing returns. This is because, as
additional variable input X is applied in conjunction with the amount of fixed resource being used.
MPP declines. Initially, while TPP continues to increase, MPP is positive. Eventually, however,
TPP reaches a maximum and then declines absolutely at which stage MPP becomes increasingly
negative.
FIGURE 8.1 - Total Physical Product (TPP), Marginal Physical Product (MPP) and Total Value
Product (TVP) per Technical Unit for a Single-variable Production Function Y = f(X)
Value relationships in response analysis

Since optimization is to be achieved in economic (here financial) terms, it is necessary to convert the
above physical relationships to a value basis per technical unit. This is done simply by multiplying
TPP, APP and MPP by the unit price py of the output to obtain:
Total value product: TVP = (TPP)py
Average value product: AVP = (APP)py
Marginal value product: MVP = (MPP)py
Of these value relationships, only TVP is shown in Figure 8.1. In fact, for the economic analysis of
response, only the TVP and MVP relationships are needed; the average relationships APP and AVP
are not further discussed except to note that, on the basis of purely physical efficiency, the amount of
input used should never be less than the level at which its MPP equals its APP so that its APP is
decreasing, not increasing. Likewise, and again on purely physical grounds, the amount of input
should never exceed the level at which its MPP becomes negative. Thus the use of an input should
always lie within the range where its MPP satisfies the condition APP ³ MPP ³ 0. Within this range,
the optimal level of the input will be determined by the relative price ratio px/py as outlined in
Section 8.2 below.
It should also be noted that in making the conversion from physical to value terms it is necessary to
assume that output price py remains constant over all levels of output, i.e., that the unit price which an
individual farmer receives for his or her product does not change as he or she produces less or more
of it. Constant unit prices usually apply to individual small farmers; they most likely will not apply to
farmers in aggregate and to very large producers who produce a sufficient amount of the commodity
to influence its market.
On the input side the two relevant parallel relationships are:
Total cost (TC): the cost of inputs required to achieve any specified level of output Y; it is the sum of any relevant fixed cost (FC)
(see below) and the variable cost (VC) associated with the variable inputs. This VC is calculated as the number of input units times
their unit price, i.e., Xpx for a single input X. (With multiple variable inputs X1, X2...Xn, VC = S Xi pi for i = 1 to n.)
Marginal cost (MC): is the change in total cost as one more (i.e., marginal) unit of variable input is
used. Thus, with px constant, it is simply the unit cost of the variable input.1 Hence, since D FC = 0,
for D X = 1:
MC = D TC/AX = D TC = D (FC + VC) = D FC + D VC = D VC = px
1 In production economics theory (as here), marginal cost is defined as cost per unit of input; in cost
theory it is defined as cost per additional unit of output.
Here again a parallel assumption is that for any individual small farmer the unit cost of the input is
constant regardless of how much of it is purchased. Also, as previously noted (Section 5.3), in
addition to fixed costs associated with non-variable or fixed inputs, the cost of such variable inputs
as seed, water etc. can also have a 'fixed' component such as the cost of owning/maintaining the
machine or facility by which the variable input (seed, water etc.) is applied. However, inclusion of
such a fixed component - should it exist - will increase only TC; being a fixed cost and therefore
constant, it will not affect MC.

The objective of resource-use optimization is to maximize the level of net benefit generated by
applying a resource to produce an output. Net benefit is simply the difference between total output
value and total cost, however measured (Dillon and Anderson 1990, pp. 29-32). In practice, net
benefit is usually measured or assessed as money profit. As discussed in Sections 2.1 and 6.1, money
profit provides a direct measure of net benefit in the case of market-oriented farms and often a
convenient surrogate measure for subsistence-oriented farms. Optimization of simple lower Order
Level systems (single- and two-input processes and activities) can be achieved using partial budgets
(Chapter 4) or graphs or the mathematical equations on which these latter are based; more complex
activity and whole-farm systems usually require a mathematical approach.
The physical input-output data presented in the first two columns of Table 8.1 are from an on-farm
trial concerning some animal or crop production process. Note that the data are standardized on the
basis of a technical (production) unit of one hectare. The unit cost of the input is Rs 5 and the unit
price of output is Rs 0.40. A fixed cost of Rs 6 per ha is associated with the fixed land input and the
application of the variable input. First, the base-data input (X) and output (Y) columns are expanded to
obtain a partial budget for each level of input showing total value of production (TVP), total cost (TC)
and profit (TVP - TC) in the following columns. The optimal level of input use is then found by
inspection as that at which profit is a maximum, here Rs 96 resulting from the use of six units of input.
This can readily be checked: in the table the use of one less or one more input unit would reduce
profit below this maximum level.
TABLE 8.1 - Response Optimization by Partial Budgeting of Profit
Trial Data TVP Costs Profit TVP-TC
X Y FC VC TC
(kg/ha) (kg/ha) (Rs/ha)
2 224 89.6 6 10 16 73.6
3 264 105.6 6 15 21 84.6
4 292 116.8 6 20 26 90.8
5 316 126.4 6 25 31 95.4
6 330 132.0 6 30 36 96.0
7 340 136.0 6 35 41 95.0

8 344 137.6 6 40 46 91.6
9 340 136.0 6 45 51 85.0
Equivalently, partial budgeting can also be applied using the marginal value relationships, MVP and
MC. Here the optimizing rule is to extend input use to that level at which MVP = MC, or alternatively
stated, where MVP - MC = 0 (or stated another way, where the use of one less or one more unit of
input would result in a decline in profit). These marginal-based partial budgets are shown in Table
8.2 and, as with using the budgets in total terms of Table 8.1, the MVP - MC = 0 condition is
achieved with an input of six units approximately - from graphical analysis, the exact optimal level is
6.4 units.
Sometimes the desired physical input-output relationship is available in graphical form. More often,
in simple single-input situations, a freehand curve can be sketched through an array of available
input-output data such as those of Table 8.2. On the other hand, the relationship might be of such
practical economic importance that greater accuracy is warranted in preparing the graph; if so, it must
then be derived from an underlying or statistically estimated mathematical response equation or
production function as outlined by Dillon and Hardaker (1993, Ch. 7). By whatever means the graph
is obtained, its use in resource optimization is usually straightforward. The only practical condition is
that the relative prices of the input and the output must be capable of expression as a price ratio,
px/py.
TABLE 8.2 - Response Optimization by Partial Budgeting of Marginal Value Relationships
Trial Data TVP

MVP MC MVP - MC
X Y
(kg/ha) (kg/ha) (Rs/ha)
2 224 89.6 - - -
3 264 105.6 16.0 5 11.0
4 292 116.8 11.2 5 6.2
5 316 126.4 9.6 5 4.6
6 330 132.0 5.6 5 0.6
7 340 136.0 4.0 5 -1.0
8 344 137.6 1.6 5 -3.4
9 340 136.0 -1.6 5 -6.6

The graphical approach also makes use of the marginality condition for resource-use optimization,
MVP = MC. Noting that MVP is defined as MPP times the unit value or price of the product Y, i.e.,
MVP = (MPP)py, the optimality condition can be restated as (MPP)py = MC. Noting also that MC is
simply the (constant) unit cost of the variable input X, the condition can also be restated as (MPP)py
= px or, transposing terms, as MPP = px/py. Since MPP is the rate of change of TPP as X changes,
this final relationship states that optimality is achieved by continuing to apply variable input to the
process until the resultant rate of change in output (MPP) is equal to the ratio of input unit price px to
output unit price py i.e., px/py. The input-output process is then in economic balance in the sense that
use of either more or less input would reduce profit. Analogously to the discussion of trade-off
between input factors in Section 7.5, this is shown by the fact that MPP = px/py, implies D Y/D X =
px/py, or, rearranging, (D Y)py = (D X)px so that the gain (D Y)py, from producing D Y is exactly
balanced by the cost (D X)px, of its production. In other words, for maximum profit, the last unit of
input must just pay for itself. Because of diminishing returns, prior units of input will have more than
paid for themselves; further units of input will not cover their cost.
The operating steps in graphical analysis of response are as follows:
(i) Sketch the physical response curve (TPP of Figure 8.1).
(ii) Obtain the price ratio px/py

(iii) Draw a straight-line segment which has a slope ('the rise over the run') equal to the price ratio
px/py, and move this towards the TPP curve until it is tangent to (i.e., just touches) this curve.
(iv) At this point of tangency, the TPP curve has the same slope as the px/py price-ratio line. Thus, at
the tangency point, the rate of change of Y as X changes, dY/dX, is equal to the price ratio px/py. But
from Section 8.1, dY/dX (or dTPP/dX) is the definition of MPP. Therefore, at the tangency point,
MPP = px/py, or transposing, (MPP)py = px but (MPP)py is defined as MVP and px is the cost of a
marginal unit of input, MC. Thus the tangency point locates the input level at which the optimality
condition MVP = MC is met.
(v) Read off from the graph that level of input directly below the tangency point. This is the input
level at which MVP = MC, at which profit (TVP - TC) is at a maximum, and where the use of one
(marginal) unit less or more of input in the process would result in a profit decrease.
An example of the above procedure is shown in Figure 8.2 for the simple single-input process of
applying nitrogen to a grain crop on a per ha basis. The unit price of nitrogen, denoted by pn is
assumed to be Rs 0.16 and the unit price of grain, denoted by pg, is Rs 0.90. The pn/pg, ratio is thus
0.16/0.90 = 0.18. The input level per ha at which the price-ratio line having this slope is tangent to
the TPP curve is 103 units of nitrogen per ha. At this optimal level of N, the optimal level of output is
read off as 121 units of grain per ha.
The use of a response equation to determine the optimal level of a variable input is appropriate if the
physical input-output data are available in equation form and/or if a high level of precision is
required.
As previously, the optimality condition MVP = MC is used and now applied to the explicit form of a
response relationship G = f(N) where G denotes grain output and N denotes nitrogen input. MVP is
equal to (MPP)pg, and MPP (i.e., the rate of change of grain output as nitrogen input is marginally
changed) is obtained as the first derivative of G with respect to N, i.e., dG/dN. The optimality
condition MPP = pn/pg is rewritten as dG/dN = pn/pg.
As an example, suppose the physical response relationship on a per ha basis is
G = 70 + 0.8N - 0.003/N2
and the respective unit prices of nitrogen and grain are Rs 0.16 and Rs 0.90 as previously. The price
ratio, as before, is thus 0.18. The optimality condition is then obtained as dG/dN = 0.18, i.e., 0.8 -
0.006N = 0.18. Solving this for N, the optimal level of nitrogen is 103 units per ha. Substituting this
value for N in the response equation results in a corresponding output of grain of 70 + 0.8(103) -
0.003(1032) or 121 units per ha.
Whichever of the above response-optimization methods - partial budgeting, graphical analysis or
response equations - is used, proof that optimality in resource use level is defined by the condition
MVP = MC is offered by the graph of Figure 8.3. This diagram shows the graphed values of MVP =
(dG/dN)pg = Rs(0.8 - 0.006N)0.90. On the other hand, the horizontal line is the graphed value of MC
which is the constant unit cost of N, Rs 0.16. These graphed values are equal, i.e., MVP = MC, at the
point of curve intersection. This occurs at an input level of N = 103. 'Proof consists of inspection of
the intersection point and observing that with the use of one unit less of input, MVP would be greater
than MC and it would pay to add an extra unit of input; with the use of one unit more of input, the
(marginal) cost of this last unit of input would be greater than the (marginal) value of product derived
from its use. Thus 103 units per ha is the optimal input level.
FIGURE 8.2 - Example of Graphical Analysis of Response
It is often more convenient to approach optimization of resource use in terms of the basic profit
equation. Profit was defined previously as total gross return less total cost, i.e., denoting profit by pi,
the profit from a response process is
p = TVP-TC
where TC = VC + FC. Total gross return equals TVP which is output times its unit price. FC is here
assumed to be Rs 5 per ha. VC is variable input quantity times its unit price. Again using the nitrogen-
grain example with pg = Rs 0.90 and pn = Rs 0.16, the equation for profit in Rs per ha is given by:
p = TVP - VC - FC = pgG - pnN - 5 = 0.90(70 + 0.8N - 0.003N2) - 0.16N - 5
which reduces to
p = 58 + 0.567N - 0.0027N2
which is at a maximum when the level of N is such that using one (marginal) unit more or less would
cause profit to decline, i.e., where the marginal rate of change of p with respect to N, i.e., dp /dN, is
zero. This condition implies
dp /dN = 0.56 - 0.0054N = 0.
FIGURE 8.3 - Graph of Optimality Condition MVP = MC

Solving this equation for N, the profit-maximizing level of N is found to be 103 units per ha, the same
as obtained previously using the equivalent criteria MVP = MC or MPP = pn/pg. With this optimal
level of N, as before, the corresponding levels of G and p are, respectively, 121 units per ha and Rs
87 per ha.
Distinction needs to be recognized between maximum output and that level of output which maximizes
profit (i.e., the economically optimal output). Again using the above nitrogen-grain example,
maximum output is defined as that resulting from an input of N which if altered marginally in either
direction would result in output decreasing or, stated another way, where the response curve peaks
and has a slope of zero implying that the first derivative of grain output with respect to nitrogen input
(i.e., dG/dN or MPP) is zero. Thus for the nitrogen-grain response equation G = 70 + 0.8N - 0.003N2,
the output-maximizing level of N is found directly from dG/dN = 0.8 - 0.006N = 0 as 133 units per ha
which compares with the profit-maximizing level of 103 units. The maximum level of output is thus G
= 70 + 0.8(133) - 0.003(1332) = 123 units per ha with an associated profit of Rs 86 per ha which
compare with the respective profit-maximizing levels of 121 units and Rs 87 per ha.
In a large-scale commercial situation with the input to be applied over a large area, the difference
between these two levels could be seen as significant so that exactly 103 units of N would be applied
per ha. But on small farms, especially those of a near-subsistence nature, the farmer would be likely
just to throw on a few extra units of fertilizer in the knowledge that the amount which will give the
'best' crop is about the same as that which will give the 'best' net return.
Obviously, the achievement of optimal resource use is possible only if the resource is available in
sufficient quantity or - what amounts to the same thing - only if the farmer has sufficient budget
available to buy it in sufficient quantity.
Again using the nitrogen-grain example, it is easy to see if such a budget constraint would be present.
The optimal input of N is 103 units per ha which would cost Rs 16.48. Suppose there is only a budget
amount of B per ha available for purchase of N. The implied budget constraint is thus B³ 0.16N. It
will be immediately apparent whether the necessary input cost of Rs 16.48 per hectare is greater or
less than whatever budget level B might be operative; and if greater, the actual amount of N which can
in fact be afforded. Thus a budget B of Rs 10 per ha would permit the budget-constrained use of only
B/pn = 10/0.16 = 63 units of N per ha regardless of what the optimal level might be. This simple type
of single-input situation hardly warrants a formal statement of the budget constraint in equation form.
However, in multi-input and multi-process situations, formal statements of any constraints to
optimization are required - see Dillon and Anderson (1990, pp. 40-53).
When the use of variable input factors is limited by a budget constraint, the law of diminishing returns
implies that the limited budget should be spread across all the variable factors and all the technical
production units (i.e., units of land area cropped or animals fed). The limited budget should not be
concentrated on a limited number of variable factors or technical units. Figure 8.3 with its graph of
declining MVP = 0.90(0.8 - 0.006N) for the response process TPP = 70 + 0.8N - 0.003N2 provides
an illustration. (Note that this declining MVP is a reflection of diminishing returns to N in the
response equation.) Faced with this response situation, suppose the farmer could only afford 80 units
of N for his or her two hectares of crop. It would be best for 40 units of N to be allocated to each
hectare rather than all 80 units to one hectare. The former strategy would generate a total profit
(assuming the same prices as before) of Rs 152. In contrast, the latter strategy would only generate Rs
58 on the unfertilized hectare and Rs 85 on the fertilized hectare, giving a total profit of only Rs 143.

Optimization of two- and three-variable input processes can also be achieved by partial budgeting
(using total relationships as in Section 8.2.1 above) but this method would be limited by the work
involved in preparing the large number of necessary partial-budget tables. In practice, optimization of
resource use in multi-variable input processes is best approached by an extension of the response
equation method discussed in Section 8.2.3 above.
As outlined in Section 5.1, single-variable input production relationships of the type G = f(N) are
usually a greatly oversimplified view of reality. More realistic would be the statement G = f(N, W/P,
K, L, CK) in which grain yield is recognized as depending explicitly on the level of nitrogen and
water while also acknowledging that soil phosphorus and potash status, amount of labour applied,
cultivating frequency etc. are also important factors. More realistic still would be analysis which
explicitly evaluates all of these factors as variable inputs. But usually a balance must be struck
between achieving greater realism on the one hand and analytical capacity on the other. This is
generally based on selection for analysis of only those input factors which (a) have significant
economic relevance and (b) can be manipulated or controlled by management. (Many factors are not
easily amenable to farmer manipulation: e.g., the necessity sometimes of simply accepting whatever
volume of water comes down the irrigation supply channel. Other factors such as, e.g., rainfall and
solar radiation, are totally beyond the farmer's control; from his or her perspective they are not
decision factors except in so far as supplementary strategies - such as irrigation to overcome rainfall
deficits - may be used to ameliorate his or her lack of control.)
Thus, in practice, analysis of most multi-variable input processes of Order Level 2 is limited to
examining the simultaneous effects of two or three variable-input factors which are under the farmer's
control. In contrast, analysis of the resources used in production function analysis of whole-farm
systems of Order Level 10 might consider four, five or six categories of inputs as outlined, e.g., by
Dillon and Hardaker (1993, pp. 221-224) and Heady and Dillon (1961, pp. 218-228). (The analysis
of whole-farm systems using production functions is largely confined to work in Field D relating to
policy guidance (Section 2.1.7) which is outside the scope of this discussion.)
Resource use in multi-variable input processes is optimized using an extension of the response
equation method outlined in Section 8.2.3 above, regardless of whether two- or three- or n-variable
inputs are involved. Here, for purposes of illustration, two variable inputs are assumed but the
method extends directly to three or more variable input factors. Again the relevant relationships are
the physical response function, the MVP of each variable input used and the unit price of each input in
relation to the unit price of the output. The only operational change is that the partial derivative of
each variable input is now used and the MVP of each variable input is now obtained as this partial
derivative times the unit price of the input.
The optimizing rule is to use each input at the level at which the ratio of its marginal physical product
to its unit price is equal to the inverse of the unit product price. Thus the response process Y = f(X1,
X2) is optimized when
MPP1/p1 = 1/py = MPP2/p2
where MPPi denotes the MPP of Xi. To illustrate the procedure it is assumed that a response function
relating grain output (denoted by G) to input of nitrogen (N) and water (W) could be obtained by
fitting some appropriate equation to input-output data from a field trial to obtain the response function
G = f(N, W). The unit price of grain is given as p and the unit costs of the two inputs are given as pn
and pw respectively.
Given the response relationship G = f(N, W), the marginal value product of N, denoted by MVPn, is
defined as pg(MPPn) = pg(¶ G/¶ N) where ¶ G/¶ N denotes the partial derivative of G with respect to
N. Likewise, MVPw = pg(¶ G/¶ W).
Setting these respective marginal value products equal to their respective input prices (i.e., the
marginal cost of each input factor), profit maximization implies that:
(i) pg(¶ G/¶ N) = pn
(ii) pg(¶ G/¶ W) = pw
Dividing each of these equations by pg gives

(iii) ¶ G/¶ N = pn/pg
(iv) ¶ G/¶ W = pw/pg
Dividing (iii) and (iv) through, respectively, by the prices pn and pw,
(v) (¶ G/¶ N)/pn = 1/pg
(vi) (¶ G/¶ W)/pw = 1/pg
Since both equations (v) and (vi) are equal to 1/pg they are equal to each other and, noting that the
left-hand numerators are the respective marginal physical products of N and W, equations (v) and (vi)
can be rewritten as the optimality condition:
(vii) MPPn/pn = MPPw/pw = 1/pg
or, equivalently, after multiplying through by pg,

(viii) MPPn/pn = MPPw/pw = 1
Example
Given the response function
G = 500 + 100N + 4W - 5N2 - 2W2 + 2NW
for grain (G) in relation to nitrogen (N) and water (W) applications on a per ha technical unit basis
and output and input unit prices of pg = Rs 20, pn = Rs 2 and pw = Rs 3, and working in terms of the
basic optimization relationships MVPn = MCn and MVPw = MCw, (i.e., MVPn = 2 and MVPw = 3),
corresponding to equations (i) and (ii) above:
2000 - 200N + 40W - 2 = 0
80 + 40N - 80W - 3 = 0.
Solving these two equations simultaneously for N and W (e.g., by multiplying the first by two and
adding the result to the second to eliminate the W term), the optimal input levels are obtained as 11.31
units of N and 6.62 units of W per technical unit with a corresponding optimal yield of G = 1 080
units per ha. Substituting these derived values of N and W into equation (viii) shows that they satisfy
the optimality condition.
As noted previously, often a more convenient way of approaching optimization is to work directly
with the profit equation. From this viewpoint the multi-input optimization problem is that of using
each variable input factor up to a level at which the resultant difference between TVP and TC, i.e., p ,
is at a maximum. This will be achieved when the use of one more or one less (marginal) unit of the
input factor would lead to a reduction in profit. This occurs at input levels at which the rate of change
in profit, relative to each input, is zero - or, in mathematical terms, when the partial derivative of
profit relative to each variable input factor appearing in the profit function p = TVP - VC - FC is
zero.
As before, given the underlying physical response relationship, the first step is to convert this to value
or TVP terms. Using the same response equation and prices as in Section 8.3.1:
TVP = 20(500 + 100N + 4W - 5N2 - 2W2 + 2NW).
On the cost side of the profit equation, variable costs are simply the amount of N used in the process
times its unit price plus the amount of water used times its unit price, i.e., Npn + Wpw. Fixed costs are
assumed to be Rs 10 per ha. Thus
p = TVP - VC - FC = (10 000 + 2 000N + 80W - 100N2 - 40W2 + 40NW) - (2N + 3W) - 10 = 9 990 + 1 998N + 77W - 100N2 - 40W2 + 40NW.
Taking partial derivatives of pi with respect to N and W, and setting these equal to zero (because such
equality defines the condition for profit maximization), the following two equations are obtained:
¶ p /¶ N = 1 998 - 200N + 40W = 0
¶ p /¶ W = 77 + 40N - 80W = 0.
Solving these two equations simultaneously for the values of N and W at which profit is at a
maximum, these values are 11.31 units of N and 6.62 units of W per technical unit, the same as were
obtained in Section 8.3.1 using the less direct MVP = MC approach.
Note that fixed costs (FC) are irrelevant to determination of marginal rates of profit change with
respect to variable inputs. This is because, by their nature, FC are invariant relative to input levels
and profit. Thus, unlike VC, they play no part in the profit optimizing decision.

Possible sources of data for response analysis are briefly reviewed.

In most Asian-Pacific countries there exists a large volume of potentially useful data relating to the
major crops and their inputs/outputs in specific farming environments. In colonial times the emphasis
was on the export crops (tea, rubber, coconut, oilpalm, cacao, jute etc); the village or 'minor' (often
staple!) crops were relatively neglected. In post-colonial decades, emphasis has shifted to the major
food crops (paddy and the other cereals, food legumes and root crops).
Commodity-specific national research institutes such as the Rubber Research Institute of Malaysia,
the Coconut Research Institutes of the Philippines and Sri Lanka, the Tea Research Institute of Sri
Lanka etc. continue to generate much information from field experiments and this work is now
generally broadened to include other crops grown in association with the main export crops (e.g.,
food crops inter-planted in young rubber and similar under-cropping with coconut). As well, much
response information on staple crops and on livestock is available from such international
agricultural research institutes as AVRDC (vegetables), CGPRT Centre (coarse grains, pulses, roots,
tubers), CIAT (beans, cassava, tropical forages), CIMMYT (maize, wheat), CIP (potato, sweet
potato), ICLARM (aquaculture), ICRAF (agroforestry), ICRISAT (sorghum, finger millet, pearl
millet, chickpea, pigeonpea, groundnut), ILRI (cattle, buffalo, sheep, goats) and IRRI (rice).
Too, in spite of their general past neglect in comparison with the export crops, some of the village
crops were in fact intensively investigated decades ago: the libraries of Bogor, Kuala Lumpur, Los
Baños, Calcutta, Kandy, Islamabad ... as well as the archives of the ex-colonial administrations are
depositories of vast amounts of information covering a wide range of input/output relationships for
field and tree crops and livestock. This is now often neglected but a considerable part of it is still
relevant. Of particular value (and fascination) are the old monographs and Agriculture Department
journals of the 1920s and 1930s. This was indeed a golden age in terms of the range of crops
investigated and practical guidelines which were developed for their management: kitul palm,
palmyrah, areca, ginger, breadfruit, citronella, sago, cassava, nipah palm ... are only some of the
'people's crops' that were studied - and only now are some of these crops being officially
rediscovered.
Useful data from this earlier period usually concern yield relationships relative to such husbandry
variables as plant spacing, pruning method, manure application, crop age at harvest (Figure 8.4),
harvest frequency and water application. These relationships are seldom presented as formal
response functions but as tables of input/output data from which such functions may be constructed.
As an example, Figure 8.4 shows the relationship between crop age and yield for two varieties of
cassava as determined from field trials conducted in the 1920s. Assuming that the varieties are still
current and the environments are similar, the response curves shown might well be of direct present
use. Or, if the varieties have been superseded, the relationships shown might at least serve as a guide
in structuring an on-farm trial to up-date such information.2
2 Continuing validity of data from past crop trials will obviously depend on whether or not the old varieties are still current. In
general, the major short-term food-crop varieties are frequently superseded by new cultivars, as are some of the major (export)
tree crops such as rubber and oilpalm. But most tree crops and most minor crops are replaced by improved varieties only once in
several decades, if at all.
When working with farmers in Modes 3 and 4, i.e., respectively, diagnosis of system weaknesses and
prescription of solutions, most farm analysts (consultants, extension officers, village development
agents) will be able to organize simple field trials in cooperation with farmers for such short-term
crops as paddy, maize, sorghum, wheat, clover etc. These will typically involve maintaining the trial
over some short period of three to six months. They will also be able to execute on-farm trials
involving those aspects of livestock production which involve a rapid short-term response, e.g.,
feeding at different input levels in relation to yield of milk or fish, but not, e.g., long-term breeding
trials aimed at the genetic improvement of livestock.
FIGURE 8.4 Sketch of Field-trial Results for Two Varieties of Cassava
As a consequence of the recognition of the usefulness of a systems approach to farm research and of
the benefits of farmer participation in such research, much has been written on the need for and
conduct of on-farm trials - see, e.g., Amir and Knipscheer (1989), Dillon and Anderson (1990. Ch.
8), Farming Systems Support Project (1987a, b, c), Hildebrand and Poey (1985), Mutsaers and
Walker (1991), Mutsaers, Weber, Walker and Fischer (1997), Nagy and Sanders (1990), Norman,
Worman, Siebert and Modiakgotia (1995) and Tripp and Woolley (1989). The following points are
offered as a general guide.
Selecting the farm: In setting up a farm experiment on behalf of a group of farmers or village the trial
should be located on a farm which is as representative as possible of all those other farms to which
the results may be relevant.
Selecting the farmer: Selection of the cooperating farmer on whose land the experiment is to be
located is usually more difficult. The farmer should be representative of his or her peer group, neither
outstandingly advantaged or disadvantaged; but in addition he or she should be an opinion-leader
though not an official. However, the most cooperative farm families are often those of the village
chief or other officials (local-level officials are often also farmers); and these people often enjoy
significant farming advantages in owning the best lands, getting priority in water supply, having
access to credit etc. These factors, which may be favourable to the successful technical conduct of an
on-farm trial might later become disadvantages when they are cited as reasons why 'ordinary'
villagers don't or can't apply trial results.
Identifying the decision-maker: It should not be assumed that the male head of the household is the
'farmer' or 'decision-maker'. In many societies the farmers, as managers and operators if not as
resource owners, are women. Obviously any on-farm trial must be carefully explained to and if
possible planned with all relevant members of the cooperating household, not just its nominal head.
Ensuring maximum participation: It is essential that on-farm experiments involve the maximum
number of potential beneficiaries at all stages of planning, execution and evaluation. While most
farmers will not understand the concepts of resource optimization as these have been presented in this
chapter, they will understand when 'optimality' is presented in down-to-earth terms and supported by
visual demonstration.
Defining the subject for investigation: The analyst will probably approach the setting up of a trial
with a fairly firm idea of what is to be investigated. But unless he or she sits down and discusses this
with the household (or even better, the village), he or she might not learn, e.g., (a) that the object of
the trial is judged by them to be not important; or (b) that, although important, there are even more
important relationships which should be investigated first; or (c) that there are better ways of
measuring the proposed relationships than those the analyst had intended to use; or (d) that even
though the trial might perhaps yield positive results, the village would have absolutely no intention of
applying them because of local economic/institutional/social/political factors of which the analyst, as
an outsider, is ignorant.
Defining the trial boundaries: Regarding the trial as a system, its boundaries should be carefully
defined to include secondary outputs or side effects even when these cannot be accurately measured.
An example is offered by the practice of many Sri Lankan farmers of using much more water on paddy
than is needed for optimal rice output. The extra water is used to reduce weed growth. Being natural
economists, the farmers have figured out the optimal balance between grain lost due to excessive
water and weeding-labour saved. A water input/grain output trial which excluded consideration of
the labour-saving output of water would be of little practical relevance to these villagers.
If working with large farms/estates growing perennial tree crops, analysts and development officers
will be able to conduct both short- and long-term trials. Short-term trials will involve factors having
a rapid short-term response, e.g., yield in relation to fertilizer application, or yield in relation to
harvest frequency. Long-term trials will aim at evaluation of the effect on yield of such factors as tree
variety and spacing, or pruning patterns, or the mutual effects of different species in a tree-crop mix.
Because of the long periods of time required to obtain results and difficulties of trial control, long-
term trials with perennial crops will usually not be practicable on small farms; only short-term
(usually seasonal) trials of some aspects of their production will be possible.
Figure 8.5 shows the results of a typical long-term tree-crop trial relating the yield of cashew nuts in
response to a range of tree spacings (and their corresponding input levels of trees per acre).
Obviously, because of the long waiting period involved before the trees begin to bear and the cost of
monitoring this type of trial over many years, it would not be practicable to conduct it in a small-farm
environment. The yield curves of Figure 8.5 refer to only the first few years of this experiment; the
effect of 'crowding' at the higher tree-population levels (closer spacing levels), which will later
become operative through competition for moisture and nutrients, is not yet apparent. It would be
necessary to continue the trial over several more years in order to quantify the full effects of spacing
on yield. Such a trial might have to be monitored over 20 years or so (over which time the response
schedules would probably take on a more conventional curvature indicating diminishing returns as
tree density increases with closer spacing). Note also that this example is useful in illustrating that
there might well be more than one basis for measuring output in terms of the technical unit of
production used. Here, yield per unit of land area would be appropriate for a farm where land is in
short supply, while yield per tree would be appropriate if the farmer's cashew is scattered over a
relatively large area (and perhaps intercropped).
Obtaining long-term data from cross-sectional surveys
One way that long-term tree-crop 'trials' can be 'conducted' rapidly is to use cross-sectional data from
a survey of tree-crop stands on a sample of farms as a surrogate. This approach is illustrated in
Figure 8.6 which relates yield of cacao to tree age and population density, and thus depicts the two-
input process Y = f(age, density). The data were obtained by a survey of a sample of farms, all
growing cacao of the same variety on similar soils and inter-planted with a similar mix of other tree
crops of different ages and densities. Although the data from this type of cross-sectional survey come
from many farms, it can usually be used as a guide for an individual subject farm as long as this is
similar in essential respects to the surveyed farms. The main point is that data which would otherwise
take 25, 30... years to obtain can often be obtained in two or three weeks by a field survey.

The data on which response analysis is based and the information which it gives can never be perfect
(Dillon and Anderson, Chs 7 and 8). First, there will always be uncertainty about the effect of such
uncontrolled factors as climate, pests and disease. Second, relative to its proposed use, the data on
which the response relationship is based may be imperfect, e.g., it may relate to a different location
(the problem of location specificity) or to a superseded variety or a soil of different fertility status.
Third, the assumed response relationship can only be interpreted as an average relationship across
some set of, preferably, representative observations. The better the spread of these observations
across time and space, the more reliable will response analysis be. Data from trials of a multi-year
multi-location nature are to be preferred to data from a trial conducted only once in a single location.
Fourth, prices - particularly of output - may not be known with certainty. Fifth, every farm household,
farmer and farm is unique. Resource qualities vary between farms. Farmers vary in their managerial
ability, their knowledge, their opportunity costs, their assessment of uncertainty and their reactions to
it, and in their preferences about the possibilities they see as open to them.
FIGURE 8.5 - Results of a Cashew Spacing Trial
Source; North-wood and Tsaliris (1967).
For all the above reasons, information based on response analysis should be used with caution and
judgement. While it can provide useful guidelines (as illustrated relative to Figures 7.2 and 7.3),
response analysis is no panacea to the problem of deciding on the best allocation of resources to
production processes and activities. Generally, from the perspective of a farm management analyst
working with farmers, a simple graphical approach to the appraisal of response processes is likely to
be more fruitful than a more complex full-blown multi-variable analysis based on response equations.
This is especially so relative to small farms involving a subsistence component.
FIGURE 8.6 - Sketch of Cacao Yield relative to Tree Density and Age from a Farm Survey in
Sri Lanka
Source: Based on a 1974 survey by P.C. Kuhonta and others.
8.6 REFERENCES
Amir, P. and H.C. Knipscheer (1989). Conducting On-farm Animal Research: Procedures and
Economic Analysis, Winrock International Institute for Agricultural Development, Morrilton.
Beattie, B.R. and C.R. Taylor (1985). The Economics of Production, Wiley, New York.
Debertin, D.L. (1986). Agricultural Production Economics, Macmillan, New York.
Wiley, New York.
Farming Systems Support Project (1987a). Diagnosis in Farming Systems Research and
Extension, FSR/E Training Units: Vol. I, International Programs, Institute of Food and Agricultural
Sciences, University of Florida, Gainesville.
Farming Systems Support Project (1987b). Design Techniques for On-farm Experimentation,
FSR/E Training Units: Vol. II, International Programs, Institute of Food and Agricultural Sciences,
University of Florida, Gainesville.
Farming Systems Support Project (1987c). Analysis and Interpretation of On-farm
Experimentation, FRS/E Training Units: Vol. III, International Programs, Institute of Food and
Agricultural Sciences, University of Florida, Gainesville.
Heady, E.O. and S. Bhide (1984). Livestock Response Functions, Iowa State University Press,
Ames.
Heady, E.O. and J.L. Dillon (1961). Agricultural Production Functions, Iowa State University
Press, Ames.
Hexem, R.W. and E.O. Heady (1978). Water Production Functions for Irrigated Agriculture,
Iowa State University Press, Ames.
Hildebrand, P.E. and F. Poey (1985). On-farm Agronomic Trials in Farming Systems Research
and Extension, Lynne Rienner Publishers Inc., Boulder.
Mutsaers, H.J.W. and P. Walker (eds) (1991). On-farm Research in Theory and Practice.
International Institute of Tropical Agriculture, Ibadan.
Mutsaers. H.J.W., G.K. Weber, P. Walker and N.M. Fischer (1997). A Field Guide for On-farm
Experimentation, International Institute of Tropical Agriculture, Ibadan.
Nagy, J.G. and J.H. Sanders (1991). 'Agricultural Technology Development and Dissemination
within a Farming Systems Perspective', Agricultural Systems 32: 305-320.
Norman, D.W., F.D. Worman, J.D. Siebert and E. Modiakgotia (1995). The Farming Systems
Approach to Development and Appropriate Technology Generation, FAO Farm Systems
Northwood, P.J. and A. Tsaliris (1967). 'Cashew Nut Production in Southern Tanzania: III - Early
Yields from a Cashew Spacing Experiment', East African Agricultural and Forestry Journal 33: 81-
82.
Rae, A.N. (1977). Crop Management Economics, Crosby Lockwood Staples, London.
Tripp, R. and J. Woolley (1989). The Planning Stage of On-farm Research: Identifying Factors
for Experimentation, CIMMYT, Mexico, D.F. and CIAT, Cali.
van der Zijl, C.E. (1930). Improvement of Cassava Cultivation by Field Experiments, Agricultural
University, Wageningen.
9. PLANNING WHOLE-FARM SYSTEMS:
ALLOCATION BUDGETING, SIMPLIFIED
PROGRAMMING AND LINEAR
PROGRAMMING
9.7 REFERENCES
'The news are not good. We must start from fresh.'
Pun Seet Leen, Johor girl
Methods of formulating plans for the construction or reconstruction of whole-farm systems of Order
Level 10 are presented in this chapter. It consolidates the discussion of previous chapters which were
concerned with lower Order Level systems and such relevant elements as planning objectives,
performance evaluation criteria, enterprises, activities, resources and the optimization of resource
use within activities. The planning problem is to integrate these elements into a whole-farm system
which, relative to the objectives of the farm household, optimizes farm resource use.
There are two main types of whole-farm planning: agro-economic planning, which is discussed here,
and physical farm planning, which is not discussed. Agro-economic planning amounts to the
preparation of blueprints for farm production. Physical farm planning is aimed at providing a
physical base for the production activities and thus is concerned with structuring or restructuring the
farm service matrix, a system of Order Level 9 (Figure 1.2) - e.g., providing irrigation and drainage
facilities, soil conservation, crop storage etc. Physical farm planning with its focus on engineering
and construction of farm infrastructure is outside the scope of this book. Note, however, that the kind
of physical capital which will initially be needed on a farm will be determined by the requirements of
its several activities and might change as these change. Necessary adjustments to this capital might
also be uncovered from time to time by the comparative analysis of whole-system performance
(Chapter 7). Special purpose planning might also be needed in areas such as livestock breeding but,
except in the case of mono-activity farms, this relates only to farm subsystems.
Appropriate agro-economic planning methods are determined first by the type of whole-farm system
under consideration (Section 2.2). For single-activity systems the task is to evaluate alternatives and
select the best single activity. This is done using comparative budgeting (usually of gross margins,
Section 4.3). Typical of such problems is selection of the best (profit-maximizing) crop activity for a
mono-crop farm. However, if this single crop produces several outputs which can be used as the
basis for subsequent on-farm processing/marketing activities, the question might arise as to which
combination of these second-stage activities offers the greatest potential. The problem then becomes a
multi-activity one (as illustrated by the coconut example of Figure 5.5).
In constructing a multi-activity system, the planning requirement is to select from all feasible
alternatives the optimal combination of activities relative to the objectives of the farmer and the
resources and opportunities available. Methods suitable for the planning of multi-activity farm
systems are discussed in the following sections of this chapter. Focus is on the planning of small
farms of Type 1 (subsistence) and Type 2 (semi-subsistence) in the context of a short-term planning
horizon of up to one year. No consideration is given to planning for the longer term or under
conditions of uncertainty which are, respectively, considered in Chapters 10 and 11.

Four types of economic planning may be required within a farm household's total 40-, 50-... year time
horizon, viz.:
(1) short-term planning only, within each seasonal or annual phase. Here the problem is to select the best mix of short-term crop
and livestock production activities in each phase. The necessary resources are either initially present or internally generated
within each phase or held over from some previous phase or obtained externally by barter or purchase. Planning on the great bulk
of small Asian farms is of this nature.
(2) long-term planning only. This is required in selecting those production activities which have a
lifespan extending over several phases, or possibly even over the household's entire time horizon, as
in the case of selecting the best single or best mix of perennial tree crops having lives of 30, 40... 100
years. These are best selected by comparative activity budgeting (Chapters 10 and 11).
(3) both short- and long-term planning. This is required, e.g., when the problem is to determine the
best mix of corn, citronella, cassava etc. to grow under a permanent stand of coconut palm. It is
essentially the same problem as (1) except that growing/agronomic conditions (particularly shade
effects and moisture requirements) for the short-term crops might change from year-to-year with
changing development of the long-term crop.
(4) either short- or long-term production planning combined with long-term farm development
planning. In new intensive settlement areas the planning emphasis is often on farm development
rather than current production. A common objective is to produce some minimum level of cash
income or food in each short-term planning phase while devoting most effort to land clearing, water
supply etc. so as to create a new farm service matrix of Order Level 9, as distinct from maximizing
the current output from the existing more extensive whole-farm system of Order Level 10.
These various situations provide a time dimension for whole-farm planning. They largely determine
the applicability and relevance of the various planning methods discussed in this and the following
two chapters.

The general objective of whole-farm system planning is to select from all feasible activities that
combination which is expected to best achieve the goals of system beneficiaries (here the farm family
represented by the farmer) within constraints imposed by available resources, the lack of perfect
information and other limiting factors. Household goals provide specific operating objectives in
formulating the plan. Planning consists essentially of a systematic search through all of the possible
activity combinations for that combination which best meets the planning objective. This constitutes
the optimal plan for the whole-farm system. As long as any selected plan can still be improved (by
some other combination of activities), it remains sub-optimal.
As always, the first general step in this type of planning is to specify the problem. This is done by
preparing a base table as exemplified by Table 9.1. This base table has the following four
components:
(i) An explicit or acknowledged planning objective (e.g., maximization of net cash income or food production).
(ii) A listing of all the feasible activities by which the objective might be achieved. This listing is in
the form of unit activity budgets which show the inputs required per unit of each activity, i.e., the
input-output coefficient for each activity standardized on a per unit of activity basis. Thus, in the
simple example of Table 9.1, a unit of paddy activity is specified as requiring one unit of land, three
units of water and zero units of feed. The various types of activities that may need to be considered
are discussed in Section 9.3.1 below.
(iii) As part of these activity budgets, a quantification of net activity outputs in a form by which the
possible contribution of each activity to achievement of the planning objective can be measured. Thus
if the objective is to maximize money income, the net outputs also must be measured in money units.
(iv) A list of available resources and other factors which might constrain or limit the levels to which
each of the activities can be brought into the plan. There are four main types of constraints:
Objectively based material resource constraints which typically consist of the limited amounts of available land, water, labour etc.
and which, to at least a certain degree, the analyst must accept as being fixed in the short term.
Objectively based non-material resource constraints such as limited access to markets, production
quotas etc.
Subjectively based structural constraints which the analyst might wish to impose in order to partly
pre-determine the kind of plan that will be formulated. These are discussed in Section 9.3.3 below.
Subjectively or objectively based consistency constraints which might be needed to ensure that the
plan is consistent with the family's other broad goals beyond that priority goal which has been
specified in terms of the immediate operating objective of (i) above. These also are discussed in
Section 9.3.3 below.
TABLE 9.1 - Example showing Components of a Base Table for Whole-farm Planning
(i) Planning objective (e.g., maximization of net cash income)
(ii) Activity unit budgets
(iv) Resources, constraints Soybeans Paddy Cows
Land 8 units 1 1 0
Water 6 units 0 3 0
Feed 10 units 0 0 5
(iii) Activity output ($GM) 5 10 12

Three relevant planning methods are available. In increasing order of their relative power,
effectiveness and sophistication, but also - if done by hand - in order of their increasing clerical and
tedious calculation requirements, these are allocation budgeting (AB), simplified programming (SP)
and linear programming (LP). All these methods are essentially formalized budget procedures
(Rickards and McConnell 1967). Unlike LP, however, AB and SP do not consider the (infinite) array
of all possible budgets of activity combinations and thus may not lead to the optimal farm plan. In
contrast, within the context of its assumptions and the data used, LP does lead to the optimal plan
(Dillon and Hardaker 1993, Section 4.5). This is because of LP's underlying algebraic basis as
shown by, e.g., Heady and Candler (1958, Ch. 3) or Doll and Orazem (1984, Ch. 9). Indeed, LP
(considered here in only its simple or standard form) has been further extended in many ways and,
like non-linear programming, is but one part of the general field of analysis known as mathematical
programming (Hazell and Norton 1986; Winston 1991).
Before looking at AB, SP and LP, the following general points might be noted regarding their relative
applicability in different planning situations.
The applicability of AB, SP and LP varies significantly depending on the type of farm to be planned
and the types of activities it involves. In general, LP is much more widely applicable than AB or SP.
Types of farms
Each of the three planning methods is applicable to mixed-activity family-farm systems, i.e., to small
farms of Type 1 (subsistence) or Type 2 (semi-subsistence) and to large farms of Type 5
(commercial). They are not generally applicable to the specialist family farms of Type 3
(independent) or Type 4 (dependent) or to the basically mono-crop tea, rubber, oilpalm etc. estates of
Type 6 (Section 2.2.1). On these latter estates, the (presumably best) production activity has already
been chosen; or if not, it is selected by comparative budgeting. However, if long-term activities are
involved, such as tree crops, these will generate their returns and costs as streams extending over
many years and it is necessary to first bring these (by their future nature, uncertain) income and cost
streams to some common basis in time by applying the methods of Chapter 10.
One situation in which the three methods (especially LP) can be applied to mono-crop farms is when
these produce or could produce a range of second-stage outputs and the problem then is to determine
the best combination of such outputs. Another situation is when the problem is to select the best
combination of technologies/inputs with which to produce the given mono-crop or farm-processed
product: e.g., sheet or crepe rubber on a rubber estate; copra, charcoal, oil etc. on a coconut estate.
Generally, however, the main planning problem on mono-crop farms is to determine the optimal
levels of individual resources to apply to the selected single activity and this is best done using the
methods of response analysis (Chapter 8).
Note, however, that even highly mixed farms do not automatically call for application of these
planning methods. If the economic environment (prices, costs, yields) has been stable for some time
and is expected to remain so, all that might be required are occasional marginal changes to the
production plan rather than its complete reformulation. This is best done using partial budgeting
(Section 4.5).
Types of activities and activity budgets
Activities were discussed in a general way in Chapter 4, particularly in Section 4.4.1. Here they are
reviewed from the viewpoint of their use with each of the three planning methods; in doing so, some
additional types of activities needed in whole-farm planning are noted.
(i) Final product-generating activities are those which produce in the current planning phase only a direct output which has value
in terms of the planning objective. They generate only some final return or GM. They can be handled by AB or SP or LP.
(ii) Final product + resource-generating activities produce primarily the former, but incidentally
also produce some amount of resource which is to be or could be used either within the current
planning phase or stored for later use. Sometimes a value can be attached to this resource (in which
case this is included in the activity GM), sometimes it cannot. These activities also can be handled by
AB or SP or LP.
(iii) Resource-generating activities produce only a resource for use by other activities. Typically
these activities generate only a cost or negative GM and can be handled satisfactorily only by LP.
(iv) Resource + final product-generating activities are the same as (iii) except that they also
produce some final product. The cost of producing the resource will be partly offset by the value of
the final product and this is to be reflected in the GM of the activity. Or it might be exactly offset and
result in a GM of zero. If the cost is more than offset it becomes a type (ii) activity. Usually these
activities require to be handled by LP.
(v) Development activities are aimed at producing a resource for use at some time in the future
beyond the current planning period. They carry only a cost, i.e., they return a negative GM in the
current planning period. If such an activity is undertaken fully or partly within the current planning
phase and if it is economically related to other activities in the current plan, at least part of its costs
should be charged proportionately (as an activity negative GM) to the current plan phase, e.g.,
construction of a dam entirely or partly in the current planning phase, say Year 1, if it uses resources
which could otherwise have been available to the other (production) activities in that year, should be
partly charged as a negative GM in Year 1. At the same time, insofar as its resource will become
available to future production activities over the next 30, 40... years, the costs of the dam should be
spread over these 30, 40... years rather than charged as a (large) one-off negative GM in the current
year. This requires application of the discounted cash-flow methods of Chapter 10 in order to obtain
an appropriate current-year GM for the dam construction activity. These and the following activities
require LP.
(vi) Development + production activities are the same as (v) except that here the activity also
produces some final product. This latter can occur entirely within the current planning period or as a
flow of output over some future time period. In South East Asia when jungle is cleared and land is
prepared for planting rubber, the costs of this development activity, which can occur over several
years, are often partly or entirely offset by the income-generating activities of selling, in the current
year, logs, ratan and other forest items. Or some logs might be stockpiled for producing sawn timber
or charcoal to generate activity income streams over several years. Again, as in (v), it is first
necessary to bring the costs and returns streams of this type of activity to the same value basis or
equivalence in time as that of the other activities.
(vii) Resource-purchase activities were discussed previously (Section 4.4.1) and are another way of
obtaining resources. Oxpower, labour etc. may simply be purchased. These activities also require a
budget, the GM of which is negative and equivalent to the purchase price per unit of activity. As well
as a negative GM, these budgets will show activity 'requirements' (Tables 4.4 and 4.5). Usually the
resource will be paid for with cash or credit; if this is in limited supply, the purchase-activity's
requirements (for the farmer's limited cash/credit supply) must be shown as internal or structural
coefficients in the activity's budget. If no such limits apply, there is no need to show their
requirements; the activity's budget will then consist only of its negative GM or price and the amount
of resource which the purchase activity obtains.
(viii) Resource-exchange activities are another way of generating resources by a farmer exchanging,
e.g., some (surplus) oxpower for a neighbour's (surplus) labour (Table 4.4). Again a budget is
required. If an 'even swap' has been arranged, the activity will carry a GM of zero. If the potential for
exchange is limited, e.g., either by the amount of oxpower the farmer can give up or by the amount of
labour the neighbour can give up, these quantities become 'requirements' in the activity budget.
(ix) Transfer activities were also discussed in Section 4.4.1. Typically they do not generate a
positive GM. They usually carry a GM of zero - it costs nothing to use hay to feed cows rather than
sheep. However, they might in fact incur a cost, as when a cow barn (animal storage) must be rat-
proofed before it can be transferred as grain storage space for use by a wheat activity. Transfer
activities then carry a negative GM.
In summary, as the above listing of activity types indicates, there is a variety of ways by which
resources can be provided to final product-generating activities. In general, these rather special
activities require the use of LP rather than AB or SP analysis. Their use is illustrated in the examples
of LP analysis given in Sections 9.6.2 and 3.
(x) Resource-disposal activities are primarily of a technical nature and are used only in LP. They
permit the non-use of resources. In a practical sense, since it usually costs nothing not to use all or
some part of a resource such as farm land, water, ox days etc. as long as doing so means that an
improved plan can be formulated, disposal activities usually carry a GM of zero. But there are
important practical exceptions where these activities do incur a cost, thus a negative GM. Disposal
activities are discussed further in Section 9.6.1.
Activity structural independence and dependence
All three planning methods can readily handle final product-generating activities and activities which
are structurally independent of other activities. But only LP is really capable of handling activities
which are primarily designed to provide resources internally or which are structurally dependent on
other activities. As discussed later, the reason for this is that, operationally, activities are selected for
inclusion in the plan according to some specified criterion such as their relative GM value. Since
resource-producing activities generate directly only a cost, i.e., a negative GM, they would be
excluded from the plan using AB or SP methods of analysis. (The same is true of most long-term
development activities.)
The matter of structural dependence/independence between activities is illustrated schematically in
Figure 9.1. Situation A shows a set of independent activity subsystems arranged for planning (as in
Section A of Table 9.2 below). Since each activity carries a positive GM and is independent of the
others, application of either AB or SP or LP methods is possible. In planning situation B of Figure
9.1, some activity dependence is depicted: here the cassava is both sold and used as live supports for
the beans, which are thus dependent on the cassava; and the legume crop is now not sold but used to
provide fertility for the corn crop which is in turn used to provide a grain resource for the dairy
cows. In contrast to situation A, in situation B, according to the GM activity selection criterion using
AB or SP methods, neither legumes nor corn would be considered for inclusion in the plan; and beans
also would be unlikely to come into the plan because this activity, though itself carrying a high GM,
first requires production of some amount of low-GM cassava. In summary, situation B is readily
handled by LP, but not by AB or SP unless the beans + cassava and the cows + corn + legume
activities are redefined as new composite activities. Essentially this would 'tie' a negative or low-
value GM activity to a high GM one, thus permitting the former to be considered for plan inclusion.
This is possible up to a point but is limited by the amount of clerical work involved and the fact that it
partly predetermines the planning outcome.
Given their assumptions about the structure of the farm system and its operation, AB, SP and LP each
generates an optimal plan relative to the planning criterion used and the limitations of the planning
method. Such plans, however, may in reality be far from optimal due to deficiencies either in the
assumptions made about farm structure or in the data used for planning.
Assumed optimality of resource use within activities
As implied by the above discussion, each planning method involves the manipulation of previously
defined activities. Each method requires or assumes that the level of each resource used in each
activity is at its 'best' level - e.g., that in an individual activity budget for paddy, the alleged input
'requirements' of paddy for water, fertilizer, labour etc. are economically the best amounts to provide
to this activity. If this assumption cannot be made, it might be necessary first to determine the optimal
input levels for activities by response analysis (Chapter 8). If such within-activity near-optimality is
not ensured, the subsequent application of AB, SP or LP methods might result only in selecting the
best combination of bad activities. Relatively greater gains will usually be made by devoting more
attention to structuring sound activities and then applying simple planning methods than by combining
'sick' activities into a plan using sophisticated methods.
Assumed certainty in activity budgets
Another assumption in using AB or SP or standard LP is that the results (yield, production, gross
margin) which are projected for the individual activities will in fact be achieved with reasonable
certainty (Dillon and Hardaker 1993, Section 4.5). Even the most sophisticated of plans will be of
little value if there is not a reasonable chance of them being borne out by events. Again, in such
circumstances it is likely to be better to formulate a simple plan using simple budgeting methods, and
then to subject this to the kind of analysis discussed in Chapter 11.
FIGURE 9.1 - Schematic Example depicting Whole-farm Planning Situations with (A)
Independent and (B) Dependent Activities
Optimality in plan formulation
While AB, SP and LP are respectively tools of increasing power and sophistication, the quality of
their output is always determined by how competently the planning problem has been specified. A
large subjective element enters into this. Further, because of unforeseen variations in prices and
yields, very seldom will activity budgets be exactly borne out by events. Thus an 'optimal' plan is
something of a misnomer. If the planning assumptions are a valid, accurate, comprehensive reflection
of perceived reality, the best that can be hoped for is a plan which is proven by later events to have
been near-optimal. More often it will only tend towards an optimum.
Length of the planning period
Each of the methods is most commonly used to formulate short-term plans - i.e., for a season or
operating year. Only LP is readily extended to multi-stage or multi-year planning, e.g., to generate in
one analysis all of the two, three, four ... annual phases of a two-, three-, four-... year plan - but only
if the work is done on a computer. With the same proviso, only LP is suitable for multi-stage
development planning.
Outputs of the planning process
Each of the three methods generates two types of output. The first consists of some identified
combination of activities constituting the required plan. The second output is informal and consists of
insights afforded into how the formal plan might still be further improved after the AB or SP or LP
analysis has been concluded (Rickards and McConnell 1967). However, not infrequently analysts
decline to accept these insights, preferring to crunch numbers rather than reflect on what these
numbers mean in the context of practical farm management. Too, because of the effort involved in
applying AB, SP or LP, analysts may sometimes be prone to attach undue credence to the plans
generated.
Trial-and-error approximations
Application of any of the planning methods should proceed by trial-and-error as a series of closer or
improving approximations to the final optimal plan, rather than as an arithmetical problem to be
solved, as it were, with a single blow. Inevitably, as the analyst proceeds with formulation of the
plan, he or she will think of possible additional or improved activities, or of the need to include
additional constraints, or of possible modifications in the activity resource requirements. Even on
farms with an apparently simple structure there might in fact be dozens of ways in which resources
can be used or commodities produced; they will not all be apparent when the planning problem is
initially specified.
Computerization
Computer programs are widely available for standard LP and other more sophisticated types of
mathematical programming - see, e.g., Brooke, Kendrick and Meeraus (1992) and Schrage (1991) or
Lindo Systems Inc. (1994). Similar programs could also be written for AB and SP. However, there
would be no point in doing this because LP is so much the superior method in terms of scope and
applicability; if a computer is to be used at all, it should be used for LP. AB and SP methods require
and warrant only use of a hand calculator. Their advantage lies in their 'hands-on' nature and the
direct feel for the resource-allocation problem that they give the analyst. Their disadvantage is that
they can only be applied with any surety to relatively simple problems.
The advantages of carrying out LP by computer are that larger farm systems (i.e., ones with larger
numbers of possible activities and constraints) can be considered, together with the ease and
quickness of computation as well as the avoidance of calculation errors. The associated danger is that
these features, combined with an impressive printout of results, may lead the analyst or client to
attach greater credibility to the generated plans than they deserve.
A deficiency in AB and SP is that if all the activities to be considered compete for exactly the same
resources, the first activity to come into the plan - unless it is otherwise restricted - will exhaust the
supply of its limiting resource, thus leaving no possibility for other activities to subsequently enter the
plan because they too will require this (now exhausted) resource. The generated plan will then
consist of only one activity and significant amounts of the non-limiting resources may be left idle. In
such cases, consideration of other possible plans involving a mix of activities can, if AB or SP are
being used, only be considered on a trial-and-error basis. With LP, however, constraints relating to
the farm's structure and/or to family goals may be introduced to overcome this problem.
Subjectively based structural constraints
This type of constraint was only noted in passing in Section 9.2. Briefly, in addition to specifying as
constraints the material and non-material resources which are available, in most planning situations
the analyst must also start with a preconceived idea of the kind of plan he or she wants to formulate -
i.e., the kind of system he or she wants to construct. This amounts to predetermining some selected
aspects of the plan. If the problem is one of constructing a mixed-farm system, a minimum limit would
be imposed on the number of activities to enter the plan; if working with farmers who are good with
livestock but not with crops, the analyst would perhaps impose a maximum limit on the area to be
used for crops.
In this type of structural pre-specification, the analyst must achieve a balance. If the problem is under-
specified (not enough constraints imposed), the plan might well run directly to some logical but
nevertheless idiotic solution. If the problem is over-specified (too many constraints), the planning
process might have only one possible outcome, in which case there would obviously be no point in
executing the analysis.
Constraints to achieve consistency with broad family goals
Farm planning usually emphasizes two dimensions of performance. These are productivity and
profitability measured relative to direct output either, for subsistence farms, in terms of income in
kind or, for commercial farms, in terms of cash GM per unit of land, operating capital or family
labour. But, as discussed in Section 6.2, there are at least six other dimensions or generally desirable
properties of a whole-farm system, viz.: income stability, product diversity, flexibility, time-
dispersion of income, sustainability and environmental compatibility. Operationally, these can be
incorporated into the planning process as constraints. The kind of constraint which is implied in each
case is briefly noted as follows:
· Income stability. This could be achieved by imposing maximum or ceiling limits on the size of those crop/animal activities whose
yields/prices/costs are highly uncertain and fluctuate widely over time.
· Diversity of income/food sources. The appropriate way to achieve this system property would be
to set a minimum or floor limit on the total number of different activities required to be in the plan.
· Flexibility in product disposal/use. Some activities produce outputs which can be disposed of or
used in a variety of different ways. (Use-flexibility is one means of achieving income diversity.) One
or more of these methods of disposal or use could be forced into the plan by imposing them as
minimum constraints or mandatory elements.
· Time-dispersion of production or income. This also could be achieved by imposing minimum
constraints on those activities whose outputs are highly time-dispersed rather than time-concentrated.
If applicable, perhaps the best way would simply be to allocate at least some specific area of the
farm to the highly time-dispersed tree crops, i.e., treat them as mandatory activities.
· System sustainability. This could be facilitated or enhanced by applying maximum constraints on
the erosive or fertility-depleting crops, or more general maximum limits on all crops according to the
slope and/or erodability of different parcels of land. If sustainability is threatened by a rising
watertable and salt, it might be possible to counter this by imposing maximum limits on the area of
individual high water-using crops. Alternatively, minimum limits might be set on the areas of soil-
rehabilitating crops such as pastures, trees and legumes.
· Environmental compatibility. Finally, the general desirability of formulating a plan which is
compatible with the wider physical and social environment will suggest ways of achieving this by
imposing maximum limits on certain activities or on the amounts of hazardous inputs to be used.
Considering these and the other types of constraints, in general, maximum constraints can be handled
by any of the three planning methods; but planning situations involving more than one or two minimum
constraints require the use of LP methods.

Allocation budgeting (AB) is the simplest of the three whole-farm planning methods; it is also the
easiest and cheapest to apply. Its limitations have been noted in Section 9.3 above.
AB aims at selection of the optimal mix of activities on the basis of the GM per unit of what is judged
to be the most limiting or scarce resource. Most commonly, the criterion used is either:
(i) GM per unit of land;
(ii) GM per unit of operating capital; or
(iii) GM per unit of family labour.

This is the most widely used type of AB. It is the most relevant measure of resource productivity and
activity performance where maximization of short-term money income is the operating objective and
where land is the most limiting resource. But, as discussed in Chapter 6, some other objective such as
subsistence food production may be more important for farm households not well-linked to the
market. Too, not infrequently, land will not be the most limiting resource.
AB proceeds by six general steps as illustrated in Table 9.2. This relates to a Javanese mixed farm of
one ha of which 0.1 ha is under a permanent mix of tree crops, dominated by coconut and used to
supply some subsistence requirements, while the balance of 0.9 ha is available for short-term field
crops. The six steps in AB are as follows:
1. Determine the most appropriate planning objective given the type of farm and the goals of the individual farm household.
Ideally this should be done in participatory/consultative fashion with the farm family. In the example of Table 9.2, production is
assumed to be constrained primarily by land and the first objective is to meet basic sustenance requirements after which the
objective is to maximize money income. The planning criterion is therefore (modified) GM per land unit.
2. Determine the appropriate planning period: in the example of Table 9.2, this is the coming wet
season of six months.
3. Identify all of the possible feasible final product-generating and resource-generating activities
(remembering that AB has only limited capacity to handle the latter). Activities considered could be
those that have been operated on the subject farm in the past, or new or adjusted versions of these, or
completely new activities. Depending on the Mode and Field of analysis, they might include one or
more mandatory activities (e.g., mixed-farm plans formulated in support of a cassava research project
would have to include some cassava).
In the present example, the possible activities are listed in a base table comprising Section A of
Table 9.2. As shown there, it might sometimes be necessary to distinguish between similar activities
on the basis of their intended purpose: HYV paddy is intended for cash sale while low-yielding
traditional paddy is preferred for the family's own use. In the example a minimum area of this
traditional paddy is considered as a mandatory family-sustenance activity. In this situation it would
therefore be wrong to specify only one 'paddy' activity. The second mandatory activity is maintaining
the existing area of coconut mix. This is mandatory not only because its products are needed for
sustenance (food, fuel, building materials etc.) but also because these trees would be harvested and
generally maintained regardless of any incidental cash GM which they might produce.
4. Quantify each activity by a unit budget as in the respective columns of the base table of Section A
showing the unit size of the activity, its outputs and the value of its GM (in the present instance, from
cash sales per unit of land), and its resource requirements. The unit budgets are formulated on a per
unit of land basis. The land unit used here is the (Javanese) 'patok' equivalent to about 0.1 ha. Note
that for resource-allocation purposes, it is not necessary to know the cash GM of any mandatory
activity which is included in the plan for other than GM-generating reasons. Here, e.g., both the
mandatory activities of traditional paddy and coconut mix only incidentally generate a cash GM.
5. Define and list the planning constraints as shown in the lower-left side of the base table of
Section A in Table 9.2. In the present example there are nine of these constraints. They consist of
eight physical resource constraints plus a market-based constraint. The latter is a production limit
placed on roselle to reflect the fact that there is an assured market for only two patoks of this
commodity. Note that the land constraint refers to arable land and does not include the coconut-mix
area (which therefore has a zero input-output coefficient relative to the arable land constraint). (The
one patok of coconut-mix land could have been included as a further constraint. However, since
coconut mix is a mandatory activity and its land is not used by any other activity, this is not essential.)
Note also that the labour constraints relate to family labour. There is no hired labour.
6. Formulate the farm production plan, developing Sections B and C of the AB worksheet by the
following steps:
(i) First, bring into the plan one of the mandatory activities to the required level - in Table 9.2, the one patok of coconut mix is
brought in. List the requirements of coconut mix in the first inner column of Section C; then subtract these from the initial stock of
available resources (B-1) (as shown in the resource constraints list of Section A of Table 9.2) to obtain the new Resource Balance
(B-2) as shown in Section C.
(ii) Repeat this in turn for each of the remaining mandatory activities, if any (here, only one patok of
traditional paddy giving the new Resource Balance (B-3) in Section C).
(iii) Maintain a list of each selected activity level and its associated GM as shown in the lower part
of Section B of Table 9.2.
(iv) After all mandatory activities have been brought in, select next that non-mandatory activity
having the largest GM: here roselle with a GM of $60 per patok is the first to be selected.
(v) To find the maximum level to which roselle can enter the plan, divide each element of the current
Resource Balance (B-3) of Section C by the corresponding element of the roselle requirements
column of Section A and enter the results as shown in Section B of Table 9.2. The maximum possible
level (i.e., number of units or, in this case, patoks) of roselle is given by the smallest number
appearing in this generated column of Section B; this is '2' (it is placed in parentheses for ease of
reference), i.e., the most limiting constraint for roselle is its market constraint of two patoks.
(vi) Enter roselle resource requirements as shown in Section C of the worksheet. Because there are to
be two patoks of roselle in the plan, these resource requirements are obtained as two times the unit
resource requirements of roselle from Section A, i.e., (2) x 1 = 2 units of arable land, (2) x 20 = 40
units of cash etc. Subtract the total roselle activity requirements so obtained from the Resource
Balance (B-3) to obtain the new Resource Balance (B - 4) of Section C.
(vii) Repeat steps (iv), (v) and (vi) to bring other activities into the plan on the basis of the highest
GM (here only HYV paddy) to levels permitted by the (declining) available resources of Section C
and enter the TGM of each selected activity in the lower part of Section B. This concludes the formal
AB analysis. For the example farm of Table 9.2, as shown in the bottom part of Section B, the optimal
plan based on AB using the criterion of GM per unit of land consists of one patok of coconut mix, one
patok of traditional paddy, two patoks of roselle and 4.3 patoks of HYV paddy. The plan meets the
mandatory activity requirements and is expected to generate a farm TGM of $384.
TABLE 9.2 - Example of AB Worksheet for Whole-farm Planning based on GM per Unit of
Land
However, there still remains the important informal step of considering if and how the worksheet-
based plan might be further improved. Thus, e.g., inspection of the worksheet's final Resource
Balance (B-5) reveals that 1.7 patoks of land remain unused as well as 3.6 units of irrigation water.
Few farmers would accept this. The analyst would now reconsider the initial planning specifications.
Could other possible activities be considered? Could the resource requirements be relaxed? Could
additional resources be found? Thus, would perhaps a little more HYV paddy be possible?
Expansion of this activity is prevented by prior exhaustion of the farm family's planting-labour
resource. Could a little more planting labour be found? But even if this proved possible, HYV paddy
expansion would then be limited to only 0.5 additional patoks by irrigation water (i.e., 3.6/8 = 0.5
patoks). The same limit and maximum possible additional level of 0.5 patoks would apply if an
expansion of traditional paddy was attempted (this additional area, over and above family food
requirements which have already been met, could now be regarded as a cash crop). Perhaps maize is
a possibility. Dividing Resource Balance (B-5) by the requirements for maize, the maximum possible
area of maize is limited to zero by the lack of planting labour and next would be set by irrigation
supply at 3.6/2 = 1.8 patoks. Thus there would be more than enough water to permit using the
remaining 1.7 patoks of land for maize as long as a few additional days of planting labour (1.7 x 5 =
8.5 days) could somehow be obtained. At this point the initial problem specifications in Section A of
Table 9.2 regarding available resources, particularly planting labour, would be checked. The farmer
who has provided the information on his or her constraints might concede that, even though the supply
of planting labour is tight, he or she could nevertheless possibly obtain the necessary nine additional
days of labour by exchanging, say, 18 of his or her surplus growing-labour days with a neighbour. If
so, 1.7 patoks of maize would then be possible and the final plan would be as shown in Table 9.3.
This plan gives an expected gain in farm TGM of $43 over the initial plan of Table 9.2. However,
some unused resources still remain. The analyst might continue to look for ways - other activities - to
use them up. But since land and water are now exhausted, such other activities (if they can be found)
could not be field crops but only some intensive activity - such as stall-fed livestock or a fish pond -
which uses only cash and labour.
TABLE 9.3 - Final Whole-farm Plan based on GM per Unit of Land for the AB Worksheet
Example Farm of Table 9.2
Unused Resources
Production Activity Level (patoks) TGM ($)
Resource Amount
Coconut mixa 1 40 Cash ($) 28.0
Traditional paddya 1 30 Growing labour (days) 22.0b
Roselle 2 120 Harvest labour (days) 8.8
HYV paddy 4.3 194 May oxpower (days) 8.7
Maize 1.7 43 Irrigation water (units) 0.2
Farm TGM 427

a Mandatory activity.
b 51.9 days from Resource Balance B-4 of Table 9.2, less 11.9 days for the 1.7 patoks of maize now
brought into the plan, less 18 days exchanged with neighbours for nine days of planting labour for the
maize.
Calculating net farm income
In the above AB and the following SP and LP analyses, the activity budgets account for only their
respective direct costs; they do not account for those farm costs which are not activity-specific. The
expected cash return from the plan of Table 9.3 is measured as a farm TGM of $427. If the level of
farm net cash return is needed, it is obtained as farm TGM less any farm fixed costs: e.g., if whole-
farm fixed costs are $100, farm net cash return for the plan of Table 9.3 is $(427 - 100) = $327.
There is one possible hazard. The cost of some existing capital item, e.g., a maize shelter, might have
been attached to some specific activity, e.g., growing maize, at the budget preparation stage. If then at
the analysis stage maize does not come into the plan, the fixed costs of the sheller will not have been
accounted for and must now be included as a part of whole-farm fixed cost in calculating net farm
cash income.
This planning criterion is based on the assumption that operating capital (cash and access to credit) is
the most limiting resource. The assumption can be a valid one on farms in both the developed and
developing world. The large sheep, cattle and wheat farms of Australia have historically been more
short of capital than of land; and the same is true, e.g., of many farms on the geographical fringes of
development in Kalimantan, Bhutan and West Irian.
As exemplified by the AB worksheet of Table 9.4, the problem is again specified by construction of
the base table of Section A. However, while the activity unit budgets are again on a per unit of land
basis, the base table now includes the new selection criterion of GM per unit of operating capital.
Operating capital is defined as cash plus available credit or, if appropriate, as cash and available
credit plus the cash value of resources in store which have required a past cash outlay (e.g., such as
an on-farm store of fertilizer). In the example of Table 9.4, operating capital consists of only cash. It
is also assumed that traditional paddy is now only a marketable crop and not mandatory. The only
mandatory activity is the existing one patok of coconut mix. Having Section A of the worksheet, the
procedure to develop the farm plan is the same as in the previous example of Table 9.2 but with
priority for the entrance of activities being determined by their GM per unit of operating capital. Thus
the first activity to enter the plan after the mandatory coconut mix is maize. The full plan based on the
uncompleted worksheet of Table 9.4 is given in Table 9.5 along with a listing of its associated
unused resources. This plan has an expected farm TGM of $418. As with the plan of Table 9.2,
further partial budgeting may lead to a marginally better plan.
The criterion of GM per family labour day is probably the most relevant planning criterion for
farmers who have no shortage of land but also no real potential for usefully employing more operating
capital, even if it became available. The bulk of subsistence-oriented (i.e., Type 1) farms in
geographically isolated areas fall into this category. Often the only significant resource of these
families is their own labour supplemented with a hoe, a shovel and an axe. Indeed, as farms tend
further towards the subsistence end of the farm-type scale (Figure 6.1), it might even be necessary to
specify activity outputs in terms of their food/energy content. In extreme cases it might be appropriate
to reduce all activity outputs to whatever the local basis of exchange happens to be - maunds of
paddy, 'tins' of maize, bottles of butter oil etc.
Once the base table has been prepared, the steps in selecting activities to enter the plan using the
criterion of GM per unit of family labour are the same as outlined above for the criteria based on land
and operating capital. The procedure is illustrated by the uncompleted worksheet of Table 9.6. For
the present example, this leads to exactly the same plan as that based on GM per unit of operating
capital (Table 9.5). Because there is little difference between the TGM of each plan generated by AB
on the basis of GM per unit of land, operating capital and family labour, it can be concluded that these
three resources are, for all practical purposes, equally constraining to the example farm.

Simplified programming (SP) is used for the same purpose as AB. However, it proceeds by
successive reference to two criteria rather than only one as in AB and thus requires a good deal more
clerical work. In this it is held to be more sophisticated than AB; but it is subject to the same
limitations.
An example of SP is shown in the worksheet of Table 9.7 with its five sections A, B, C, D and E. As
with AB, the first step in SP is to define or specify the planning problem in terms of a base budget
table as given by Section A. These budgets are here expressed on a per unit of land basis. The second
step is to develop two intermediate working tables, Sections B and C. The analysis then proceeds
through the operations shown in Sections D and E.
TABLE 9.4 - Example of AB Worksheet for Whole-farm Planning based on GM per Unit of
Operating Capital
TABLE 9.5 - Whole-farm Plan based on GM per Unit of Operating Capital for the AB
Worksheet Example Farm of Table 9.4
Unused Resources
Production activity Level (patoks) TGM ($)
Resource Amount
Coconut mixa 1 40 Arable land (patoks) 0.6
Maize 3 105 Operating capital ($) 52
Roselle 2 120 Growing labour (days) 44
HYV paddy 3.4 153 Harvest labour (days) 21
May oxpower (days) 12
Farm TGM 418 Irrigation water (units) 13

a Mandatory activity.
The elements of Section B, maximum possible activity levels, are obtained by dividing each element
of the initial resource constraints column (B-1) of Section A by the corresponding coefficient in the
respective activity requirements column of the unit budgets' table of Section A. Thus the paddy
activity column for Section B of Table 9.7 is obtained as good land 6/1 = 6 units, rough land (suitable
for fodder and sesame but not paddy) 3/0 = ¥ (i.e., non-constraining and therefore ignored), cash
200/20 = 10 etc. The lowest value obtained in each of these derived B columns is then bracketed.
This bracketed number indicates the maximum possible level to which the respective activity could
enter the plan with the given set of constraints, e.g., in the case of paddy, the supply of water would
limit paddy to a maximum of 2.9 units before any other constraint became operative.
The bottom row of Section B of Table 9.7 is then developed by multiplying the bracketed (activity
limiting) values of the B columns by the per unit cash GM values of the same respective activities as
recorded at the foot of the base table of Section A. The derived activity TGM values in Section B
now show the total possible maximum contributions which each activity could make to the whole-
farm plan. This row of maximum possible activity TGM values in Section B is later used as the first
basis for activity selection; these TGM values provide criterion (i) as detailed below.
Section C is next developed to obtain the activity GM per unit of each resource if it were to be used
in the production of each activity. The coefficients of the columns of Section C are found by dividing
(a) each (per unit) activity GM of Section A by (b) the per unit resource requirements (coefficients)
of the same activity, also found in Section A. For example the paddy column of Section C is found as
good land $50/1 = $50, rough land $50/0 = ¥ (ignored), cash $50/20 = 2.5, planting labour $50/10 =
5 etc. These derived coefficients of Section C are later used as the second criterion in activity
selection, criterion (ii).
TABLE 9.6 - Example of AB Worksheet for Whole-farm Planning based on GM per Family
Labour Day
TABLE 9.7 - Example of SP Worksheet for Whole-farm Planning
Having the base table of Section A of Table 9.7 and Sections B and C derived from it, the farm plan
is now formulated by applying successively the two SP criteria. The operating steps are as follows.
(1) Apply criterion (i): Select in Section B that activity which would give the highest activity TGM. It is maize with a TGM of $240.
Now identify the constraint in the maize column of Section B which would limit maize. It is 'good land'. Maize is tentatively
selected to enter the plan.
(2) Now apply criterion (ii): Go to Section C, inspect the maize column and compare the value of the
limiting constraint from Section B for good land with the value of good land if it was used for any
other activity. If it is equal to or greater than any other value in the good-land row of Section C, select
maize to finally come into the plan. If it is less than any other value/coefficient in this good-land row
of Section C, drop maize, return to Section B and select for consideration that activity in Section B
which has the next highest TGM after maize. In the example of Table 9.7, the value ($40) of good land
for maize in Section C is exceeded by the values of good land for both roselle and paddy ($75 and
$50, respectively); therefore drop maize from the plan at this stage.
(3) Return to Section B. By criterion (i), the next highest possible activity TGM would be $180 from
sesame. It would be limited by rough land to three units.
(4) Again apply criterion (ii). Go to Section C; compare the value of the rough-land coefficient for
sesame, $60, with the value of this same resource if it were to be used for any other activity. It is
higher than any other coefficient in this rough-land row (only fodder with a rough-land coefficient
value of $30). Therefore select sesame to enter the plan.
(5) As previously with AB, obtain now the sesame column of Section D by dividing the per-unit-of-
activity requirements of sesame (of Section A) into the corresponding elements of the current
Resource Balance (B-1) of Section A. Then bracket the smallest coefficient found in this new sesame
column of Section D. This shows the resource/constraint which now limits the level to which sesame
can come into the plan (i.e., rough land) and the maximum possible level of sesame permitted by this
limiting resource, three units.
(6) Again as previously with AB, begin to develop Section E by listing the resource requirements for
three units of sesame in the innermost column of Section E. Subtract these from the previous Resource
Balance (B-1) to obtain the new Resource Balance (B-2).
(7) Using criterion (i) select in Section B the next activity to (tentatively) enter the plan. Again it is
maize with an activity TGM of $240. Again it is limited in Section B by good land. Again apply
criterion (ii) by moving to Section C and comparing the GM per unit of good land for maize with the
other values in the good-land row. It is still less than the other values of $75 and $50, so again drop
maize.
(8) Return to Section B. Using criterion (i), select the next best activity. This is roselle with a
maximum possible activity TGM of $150. The limiting constraint is the market limit set on roselle of
two units. Apply criterion (ii) by considering this constraint in Section C. In no other activity is the
value of this constraint in Section C greater than its value for roselle, $75, so bring roselle into the
plan.
(9) Form a roselle column in Section D; form the column for total roselle requirements (for two units
of roselle) in Section E and subtract from Resource Balance (B-2) to obtain the new Resource
Balance (B-3) available after the introduction of roselle.
(10) Return to Section B and identify the activity with the next highest TGM. It is still maize. When
the second criterion is again applied in Section C, maize is still prevented from entering the plan by
the higher good-land coefficient for paddy. (It is also higher for roselle, but since roselle is already in
the plan, its coefficients in Section C are now inoperative.)
(11) Return to Section B. Again applying criterion (i), select paddy, noting that the limiting resource
is water. Go to Section C and apply criterion (ii). The value of water if used for paddy, $7.1, is
greatest. So bring paddy into the plan at the feasible level of two units generating a paddy TGM of
$100. Complete the paddy columns of Sections D and E.
(12) Repeat the procedure. Maize now enters the final plan at a level of two units generating a maize
TGM of $80.
(13) Obtain the new Resource Balance (B-5), i.e., $110 of cash and 16 units of planting labour. There
remains only one activity which has not been brought into the plan and which might possibly use these
resources. This is the fodder and goats activity. However, comparison of the (B-5) resource-supply
column and the fodder + goats resource-requirements column of Section A indicates that this activity
would not be feasible because of the prior exhaustion of rough (fodder) land.
This concludes the formal part of the SP analysis. As previously with AB, a running total of plan or
farm TGM is maintained at the bottom of Section D of the worksheet of Table 9.7. The final plan is as
shown in Table 9.8.
TABLE 9.8 - Final Whole-farm Plan for the SP Worksheet Example of Table 9.7
Unused Resources
Production activity Level (patoks) TGM ($)
Resource Amount
Sesame 3 180 Cash ($) 110
Roselle 2 150 Planting labour (days) 16
Paddy 2 100
Maize 2 80
Farm TGM 510
Extensions to SP by partial budgeting

As with AB, it will usually be possible to further improve the formal SP worksheet plan by relaxing
one or more of the constraints (especially if these were subjectively based), or by adjusting the initial
assumptions regarding activity requirements, or by thinking of additional possible or improved
activities. In the example of Table 9.7, the analyst might now look for activities which could
productively use the idle cash balance. There are no land or irrigation resources left so any new
production possibility obviously could not be a crop activity. On the other hand, an activity such as
stall-fed goats which uses a fair amount of cash and very little else might offer possibilities in
contrast to the budgeted goat activity which is based on fodder cropping and thus needs land. Could
perhaps a second type of goat activity be considered? One which consists of housed goats fed on crop
residues from the maize, paddy and sesame? If so, this might result in worthwhile improvement to
farm TGM. One would explore the possibility with a little side budgeting. Likewise, a fish pond for
carp or tilapia or silver barb might be a possibility.

In setting up a base table for AB or SP, the analyst will also have taken the most important and
difficult step in linear programming (LP), keeping in mind that more complex activities - especially
resource-generating activities - can now be included in the analysis. On the other hand, the amount of
tedious arithmetic calculation required in LP is greater than in AB or SP. For this reason, if the
problem is a relatively simple one involving no more than five or six activities and constraints, and
when resource- or cost-generating activities are not a significant part of the problem, it will probably
be best handled using one of the simpler methods (especially AB). As the number of activities or
constraints becomes large or activities become complex (mutually dependent), or resource- or cost-
generating activities become important, LP offers the only feasible method of analysis - but only if it
is done on a computer using such software programs as, e.g., GAMS (Brooke, Kendrick and Meeraus
1992) or What'sBest!ä (Lindo Systems Inc. 1994; Schrage 1991).
The objective in this section is limited to consideration of (1) how the base table in which a problem
is specified needs to be extended to make it suitable for standard LP application; (2) the mechanical
steps in standard LP analysis; and (3) how the main types of problems which arise in planning of
Type 1 and Type 2 whole-farm systems can be handled by standard LP. More comprehensive
treatment of the application of LP to agricultural problems is to be found in, e.g., Hazell and Norton
(1986), Dent, Harrison and Woodford (1986), Rae (1977, Chs 7 and 8) and Rae (1994, Chs 5 and 6).
As well as other types of mathematical programming, Hazell and Norton (1986) and Winston (1991,
Chs 3, 4 and 5) outline many of the extensions to LP beyond the standard form considered here.
Based on Heady and Candler (1958, Ch. 3), an example of LP analysis is shown in the worksheet of
Table 9.9. This worksheet uses what is known as the simplex method of LP analysis (Mao 1969,
Ch.3). The example involves only type (i) activities (i.e., activities generating only a final product)
(Section 9.3.1); it does not include any resource-generating activities. As previously, a base table is
prepared in which the problem is specified. As shown in iteration or part I of Table 9.9, the base
table of the LP worksheet consists first of a set of unit activity budgets including the GM of each
which, following LP convention, is now termed the activity 'price' and denoted by p; each has exactly
the same meaning as a GM and they are listed as a row of p values at the top of the base table. The
possible farm activities to be considered are referred to as real activities. The base table also
contains a list of constraints (here, for simplicity, only the supply of resources available - days of
labour, $ of working capital and units of irrigation water - and no other types of constraints) which
might limit the real activities. The base table then begins to differ from the worksheet format for AB
and SP, viz.:
(i) In addition to real activities, unit disposal activities are specified, one for each resource/constraint. Each permits the respective
resource to be disposed of, i.e., not used, by the real activities. Since in most cases (as here) the non-use of a resource costs
nothing and earns nothing, the price (p or GM) attached to each of these disposal activities is $0. Each disposal-activity column
consists of a coefficient of unity in the resource row to which it refers and zeros in the other cells. Note, however, unused
resources can sometimes incur an operating cost. If some land is left idle it might rapidly run to weeds, subsequently increasing
cultivation costs by some amount per ha the next time the land is used. If some water which is supplied to a farm is left unused, it
might lie as a public health hazard and the farmer might be charged for this at $5, $10... per unit of such unused water.
(ii) The 'resource' column B now has a more inclusive function than formerly. In the initial
specification of the problem, part I of the worksheet, it shows the amount of each resource which is
available to the activities; but in those subsequent parts or iterations II, III ... which are developed as
the programming analysis proceeds, this B column also includes those activities which have
previously been brought into and currently remain in the plan. In effect, insofar as already selected
activities can be taken out again, if by doing so the plan can be further improved, each of these
activities is a quasi-resource. (In fact something like this happens in AB and SP when a worksheet-
derived plan is marginally adjusted by side budgeting to permit the partial or full substitution of one
selected activity by another if it is thought that such a change will increase farm TGM.) In short, the B
column at any stage (iteration I, II, III ...) of the LP planning process shows resources not yet used
(and therefore still available for formulation of the next plan or iteration) as well as activities which
are in the plan at that current stage.
(iii) A work column K is needed at the right-hand side of the table in which is indicated the limit to
which any incoming activity can in fact enter the plan and the identity of the limiting resource or
constraint in the B column. (This is obtained in the same way and serves the same purpose as did
Section B of the AB worksheet of Table 9.2.)
(iv) A column p is needed at the left of the LP table in which the unit price (GM) of each item
currently in the B column (resource or other constraint or activity) is entered. As relevant, these
prices are taken as needed from the activity price row (p values) at the top of the base table.
(v) An addition to the base table as its last two rows respectively are a set of Z and Z - p values. The
specification of these is detailed below.
Iteration and plan I
Programming consists of moving from an initial iteration and plan I to successively better plans,
through iterations II, III, IV ..., according to a specified operating criterion (detailed below). In the
example LP of Table 9.9, the first three rows of the B column of iteration I constitute the first 'plan'.
The B column at any stage identifies those activities/resources which are in the plan at that stage and
their levels. Under B in iteration I, those activities in the plan are the disposal or non-use activities
for the initial resources of labour, capital and water; they are 'in' the plan to the levels shown under
B: 100 units of labour, 100 units of capital and 80 units of water (i.e., all of these resources are
disposed of, not yet used).
The TGM of this or any later plan is found as the sum of the products of each of the activity/resource
levels currently in the plan (shown in the B column) times its respective price p, given in the left-side
p column of Table 9.9. TGM under plan I is therefore (100)($0) + (100)($0) + (80)($0) = $0. This is
shown in the Z row of the B column of iteration I.
TABLE 9.9 - Example of LP Worksheet for Whole-farm Planning with only Final Product-
generating Activitiesa
a Based on heady and Candler (1958, Ch. 3).
The Z coefficients or values for the other (activity) columns in iteration 1 are now obtained in the
same way by summing the results of multiplying each coefficient of each activity column by the
corresponding price shown in the left-hand p column. Thus the Z coefficient for roselle is (1)($0) +
(0)($0) + (1)($0) = $0; the Z coefficient for cows is (1)($0) + (1)($0) + (0)($0) = $0 etc. These Z
coefficients for the various activities show the opportunity cost of bringing one unit of a real or
disposal activity into the plan. This is in terms of the income which would be lost by any necessary
forcing of other activities out of the plan. Thus, in iteration I, paddy has a Z value of $0 because, if it
were brought into the plan at this stage, it would not be at the expense of any other activity and thus
would have no opportunity cost. The main purpose in obtaining these Z coefficients for the activity
columns is to now use them in forming the Z - p operating-criterion row.
The p part of the Z - p criterion coefficient is obtained from the p row at the top of the table. Thus, in
iteration I, the Z - p coefficient for roselle is $0 - $30 = -$30, for cows 0 - 10 = -10, for paddy 0 - 40
= -40, for sesame 0 - 12 = -12; and since the prices p of the disposal activities are all zero, the Z - p
value for disposing of labour (denoted by DL) is 0 - 0 = 0, for disposal of capital (DC) is 0 - 0 = 0
and for disposal of water (DW) is 0 - 0 = 0.
This concludes iteration I. The LP now proceeds by asking and answering three questions:
(i) Is it possible to improve on the existing plan?
(ii) If so, which activities should be brought in, and in what order?
(iii) To what level can such activities be brought in?
The just-obtained Z - p coefficients are used to answer the first two of these questions. This is
because these Z - p values indicate whether the opportunity cost (Z) of including a unit of the activity
in the plan is greater or less than the net return (p) obtainable by having one more unit of the activity.
Thus if Z > p, i.e., Z - p is positive, the activity should not be included because its net return or GM
measured by p would not cover the opportunity cost of its inclusion. If Z < p, i.e., Z - p is negative, it
will pay to include the activity in the plan because its p value is larger than the opportunity cost of its
inclusion. Moreover, the best activity to enter the plan is that with the largest negative Z - p value.
Thus, in answer to questions (i) and (ii) above:
(i) Yes, it is possible to improve the existing plan (of iteration I) since there are negative Z - p coefficients below the real-activity
columns in iteration I of Table 9.9; and
(ii) That activity having the highest negative Z - p coefficient is the one to bring into the next iteration.
In the Z - p row of iteration I this is paddy which has a Z - p coefficient of minus $40.
Iteration and plan II
In iteration II, paddy is brought in but, as in the case of SP, this inclusion is only tentative: it might
later be driven out by some superior activity.
The third question is now addressed: To what maximum level can the incoming activity, i.e., paddy,
enter the plan? The procedure for determining this maximum is the same as before: divide the
resource-requirements column of the incoming activity into the current-resource column B; record the
result in column K; bracket the lowest of these recorded values. This simultaneously identifies both
the maximum possible level of the incoming activity and the outgoing resource (or activity) from the
B column. (For every such incoming activity there will be at least one outgoing resource or
previously selected activity.) In the example, the paddy activity will come in to a level of 40 units
and, as indicated by the left-side arrow from B in iteration I, the outgoing resource will be the 80
units of water.
The first step in iteration II for obtaining the second plan is to enter there the row of relevant
coefficients for the incoming activity, paddy (marked now by an incoming arrow on the left-side of
B). Construction of this and of the other new rows of coefficients in iteration II requires a substantial
amount of arithmetic; clerical error is common and for large problems with many activities the use of
a computer is almost mandatory. First we derive the new row of coefficients for paddy.
Coefficients in the incoming-activity (paddy) row of iteration II are found by dividing (a) the
corresponding coefficients in the outgoing row of iteration I (i.e., the water row) by (b) the
requirement of the incoming activity (paddy) for the outgoing (water) resource, i.e., by the coefficient
in iteration I at the intersection of the outgoing (water) row and the incoming (paddy) activity. This
pivotal value at the row-column intersection for water and paddy of iteration I has, in the present
example, a value of two; it is bracketed for easy reference. These operations to find the coefficients
for the incoming row (paddy) in iteration II are also performed on the B column. The new coefficients
for the incoming (paddy) row in iteration II are thus found as follows:
B R C P S DL DC DW
(a) 80 1 0 2 0 0 0 1
(b) 2 2 222 2 2 2
(a)+(b) = 40 0.5 0 1 0 0 0 0.5

While it is convenient to bring the incoming row (paddy) in to occupy the same row position as was
occupied by the outgoing row in iteration I (i.e., to occupy the third row of iteration II), this is not
essential.
After the coefficients for the incoming (paddy) row in iteration II are found, the coefficients for the
other rows in iteration II are obtained. Starting with the top or labour row, as illustrated below, each
of the coefficients for this in iteration II is found as follows:
(a) record the column coefficient of the same row (labour) in iteration I, then
(b) calculate the product of the coefficient of the same column in the incoming row (paddy) in
iteration II and the labour input-output coefficient for the incoming activity (paddy) for this
resource/row (labour) in iteration I. (This input-output coefficient is the paddy resource-requirement
coefficient for labour from the unit budget for paddy.)
(c) subtract the value obtained in (b) from the value in (a) to obtain the corresponding coefficient in
the labour row of iteration II.
For the example worksheet of Table 9.9, these calculations are:
(a) the row coefficients for labour in iteration I were:
B R C P S DL DC DW
100 1 1 1 1 1 0 0
(b) the coefficients of the incoming paddy row in iteration II are:
B R C P S DL DC DW
40 0.5 0 1 0 0 0 0.5
The input-output coefficient of paddy for labour in iteration I was one. Thus the set of values to be
subtracted in step (c) from the elements in (a) are:
B R C P S DL DC DW
40(1) 0.5(1) 0(1) 1(1) 0(1) 0(1) 0(1) 0.5(1)

(c) Subtracting the final values in (b) from those in (a), the coefficients for the labour row in iteration
II are thus:
B R C P S DL DC DW
60 0.5 1 0 1 1 0 -0.5
The new coefficients for the remaining rows in iteration II (only capital) are found in the same way.
As in iteration I, the next step is to form the row of Z coefficients for iteration II, including the Z value
for this current plan. The Z coefficient for B is now (60)($0) + (100)($0) + (40)($40) = $1 600. The
other Z coefficients for the activities are found in the same way, i.e., as the sum of the products of the
elements in each activity column of iteration II by their corresponding p value in the left-side p
column of iteration II. Thus the Z value for roselle in iteration II is (0.5)($0) + (0)($0) + (0.5)($40) =
$20. The final step in the iteration is to obtain the Z - p coefficients by the procedure outlined
previously for iteration I, i.e., by subtracting each activity's p value given at the top of the table from
its Z value. Thus the Z - p value for roselle is ($20 - $30) = -$10. Inspecting these activity Z - p
values as the criterion for possibly bringing further activities into the plan, the largest negative
coefficient is now -$12 for sesame. Sesame is therefore tentatively brought into the plan in the next
iteration.
Subsequent iterations and plans
In iteration III, as determined in the K column of iteration II, sesame comes into the plan at a level of
60 units as constrained by the available supply of labour. Further, iteration III results in roselle
having the highest (and only) negative Z - p value. Roselle should therefore be brought into the plan of
iteration IV. The logic of this is that according to the Z coefficients of iteration III, the marginal cost
of a unit of roselle (in terms of income reduction by driving other crops from the plan) is $26, or $4
less than the marginal income of $30 from roselle. Therefore it will pay to bring roselle in even
though it will force some other activity out. This activity substitution occurs in iteration IV where
roselle replaces paddy because the paddy row of iteration III has the lowest value in the K column. In
iteration III no other activity except roselle could come in without either leading to an income
decrease (of $2 per unit in the case of cows) or leading to no improvement in income in the case of
(more) paddy or (more) sesame.
At the end of iteration IV there are no negative coefficients in the Z - p row; thus no activity is present
which could increase plan income still further if it were brought in or substituted for any activity
already in the plan. The positive Z - p coefficients of iteration IV warn that if the unutilized cows
activity were to be forced into the plan, it would actually reduce total plan income at the rate of $2
per unit of cows brought in. Similarly if paddy was to be forced back into the plan, having left the
plan at the end of iteration III, this would reduce plan income by $8 per unit of paddy. As shown in
the B column of iteration IV (the final iteration in the example of Table 9.9), the final (and thus
optimal) plan specified by LP has an expected income in farm TGM terms of $2 640 based on 20
units of sesame and 80 units of roselle; $90 of capital is left unused. It will be recalled that this plan
TGM of $2 640 has not accounted for farm fixed costs. If required, these would now be deducted
from the plan TGM to obtain net farm income.
Value of resources
In addition to formulation of an optimal plan, LP analysis also provides useful management
information relating to the marginal value or shadow price that can be placed on the resources used
by the plan. This value is based on their actual contribution at the margin to the achievement of plan
income. In the last iteration of Table 9.9, the labour, capital and water disposal activities are shown
to have a Z - p value of $12, $0 and $18 respectively. The interpretation is that if, e.g., one unit of
labour were to be forced into non-use (or the amount of labour which is actually employed in
executing plan IV is reduced by one unit), this would reduce plan income by $12. Thus, to this
particular farmer executing this particular plan under the agro-economic conditions specified in the
base table, the value to him or her of one unit of labour less or more is $12. For similar reasons the
marginal value or shadow price of water is $18 per unit. On the other hand, because - unlike labour
and water - working capital is not fully used, its marginal value to the farmer is zero. Of course, other
farmers executing different plans with different activities and/or different input-output coefficients
would have different shadow prices placed on their resources if an LP analysis were to be done for
their farms. The usefulness of this information will be obvious. From iteration IV, the subject farmer
could afford to pay up to $12 per unit but not more for a little extra labour, and up to $18 per unit but
not more for a little extra water.
Such shadow prices are marginal productivity-based resource values. A method of finding average
productivity-based resource values was discussed in Section 7.2. Clearly these LP-based marginal
productivity resource values are more helpful to farmer decision making than are average
productivity values. This is because of their marginal orientation combined with the fact of their
holistic derivation in that they take into account the system effects of all the resources being used.
The structural characteristics which distinguish farm systems of Types 1 and 2 from other types
(Section 2.2) are that they generate most of their own resources and have a generally subsistence
rather than commercial orientation. However, the difference between Type 1 and Type 2 farms is one
of degree of emphasis placed on these respective characteristics and thus on the type of activities by
which they are structured.
On Type 2 farms the internally-generated resources (with possibly substantial externally-sourced
resources) are intended primarily to support cash-generating activities, as well as some direct
subsistence activities. On Type 1 farms the internally-generated resources (with very few external
resources) are intended primarily to support direct subsistence activities - but since complete self-
sufficiency is now rare, these Type 1 farms must also generate a little cash. There is also a difference
in planning focus. On Type 2 farms the focus is somewhat more on optimizing output of the farm
component; on Type 1 farms it is on meeting the subsistence requirements of the farm household.
These distinctions cannot be pushed too far: both types of farms merge into each other (McConnell
1992; Prabowo and McConnell 1993). But they are of sufficient magnitude to require that a somewhat
different technical approach be taken in formulating production plans for each farm type. This section
relates specifically to small farms of Type 2. The more complex more heavily subsistence-oriented
small farms of Type 1 are discussed below in Section 9.6.3.
Resource generation and activity interdependence
As noted, Type 2 farms generate most of their own resources; there is a high degree of integration of
activities and thus of structural dependence. This was implied in the definition of various activity
subsystems in Section 9.3.1. When reduced to LP format for the purpose of determining the profit-
maximizing combination of activities, their unit budgets might appear as shown in Table 9.10.
Structurally, the difference between this base table and the base table of iteration I of the LP example
of Table 9.9 is that one of the activities, growing a legume crop, is now specifically intended to
generate resources (and thus carries a negative GM or p value equivalent to its direct costs), while
the other activities of feeding cows and growing maize, each primarily a final product or cash-
generating activity, also incidentally generate resources.
As previously, in the unit budgets of Table 9.10 the requirement of an activity for a resource is
indicated by a positive input-output coefficient shown against the particular resource; e.g., one unit of
maize crop requires one unit of land. If, on the other hand, an activity generates a resource, the
coefficient for such resource is preceded by a minus sign. Thus while cows need cow feed, they also
provide fertility via their manure; while the legume activity needs land and water, it also provides
cow feed; and while the maize activity needs land, water and fertility, it also provides cow feed. This
differs from the system of Table 9.9 in that some of the resources might not be present at the initial
planning moment: the system must generate them. (It might also be possible to obtain them by
purchase or barter.)
TABLE 9.10 - Example of Base Table with Resource Generation and Activity Interdependence
Activity: Cows Legume Maize
Activity unit: per head per ha per ha
Activity GM/unit ($): +30 -20 +60
Resources B Input-output coefficients
Land 10 0 1 1
Water 10 0 1 2
Cow feed 0 1 -1 -1
Fertility 0 -1 0 1
The planning objective remains the same, i.e., to maximize farm cash TGM, but here it is subject to
the technical dependence conditions that: (a) high-profit maize needs soil fertility (but there is not yet
a 'supply' of this in the resource pool B); (b) such fertility could be provided by cows, but cows need
feed (and there is initially no feed in B); (c) such feed could be obtained from legumes and/or maize
but, from above, maize is not yet possible. Therefore, the only way to achieve the TGM objective is
to grow some legume (at a negative GM) for the cows; then obtain fertility from the cows for the
maize which, incidentally, would provide yet more feed for the cows to produce more fertility for the
maize etc. until some critical resource (land, water, labour) is exhausted.
A more realistic planning situation for a Type 2 farm is shown in the LP base table of Table 9.11.
Features pertinent to or exemplified by the worksheet example of this table are outlined below. The
example used is a small Javanese farm which well displays the resource-generating activities
characteristic of Type 2 farms.
Planning period
The assumed planning horizon for the LP analysis of the example farm of Table 9.11 is one year,
corresponding to an operating year which consists of two six-month phases: a wet-season followed
by a dry-season. Some of the Phase I (wet-season) activities generate cash outputs which become
available as cash inputs to (dry-season) activities in the following Phase II. Of course, if the system
was being planned for a longer period of 2, 3, 4... years, then outputs of any Phase II would be
available as inputs to Phase I of the next annual cycle.
Agronomic conditions as the basis for system phases and activities
In areas such as the Solo Valley of Java, high rainfall supplemented by seasonal irrigation permits
production of a large range of short-term crops in the wet season (Phase I) of the annual cycle; these
crops are planted in February to be harvested in June (Prabowo and McConnell 1993). In the (Phase
II) dry season of July-December, lack of water restricts cropping to crops which can be grown on the
dwindling supply of residual soil moisture from Phase I crops, supplemented by the little remaining
water in the public irrigation system. Further, because soil moisture is rapidly decreasing, the time
which would be required for normal land preparation for Phase II crops is in fact not available, so
that possible Phase II crops are only those which can be planted quickly by 'gejik', i.e., by placing
seeds in small shallow holes made with a pointed stick. This is done immediately after the Phase I
crop has been removed and without further land preparation. In practice, since maximum residual soil
moisture is built up under a previous wet-season paddy crop, dry-season Phase II crops consist
mainly of soybeans and peanuts which are 'gejik'-planted on ex-paddy land.
Time dimension of activities and resources
In contrast to the situation depicted in Table 9.9, some of the activities of Table 9.11 now have a time
dimension, or are time-specific in terms of some of their resource requirements. As shown in the
table, the farm's activities fall into four groups: five wet-season activities which need Phase I
resources; two dry-season activities which need Phase II resources; a third group of six activities
which continue through most of the annual cycle and require both Phase I and II resources; and a
fourth group of three barter or transfer activities whose time relationships vary.
The activities
All the activities of Table 9.11 are optional. These activities are now briefly noted as per the listing
of activity types (i) to (x) given in Section 9.3.1. There is only one type (i) activity - growing chillies
for sale. There are four type (ii) activities: beans, paddy, soybeans, peanuts. These are primarily
intended to produce a final output but also incidentally produce resources - all produce roughage by-
products for cow feeding and (except for paddy) also contribute to the degree shown to maintaining
base field fertility (for use by the fertility-extractive activities of maize, paddy, cassava and chilli).
Thus one unit (patok) of beans, as well as contributing $40 to farm cash TGM from sales, generates
0.4 tonnes of cowfeed roughage and 0.2 units of fertility to the field.
TABLE 9.11 - Example of LP Base Table for Whole-farm Planning of Small (Type 2) Farms
with Resource-generating Activities
Growing maize and keeping cows are structured as type (iii) activities, producing only intermediate
outputs/resources. The GM of maize of -$4 is the direct cost of producing one unit (patok) of maize.
As Table 9.11 shows, the cows require no external inputs/costs; they need only household labour (20
per cent of 'February' and 'other' household labour) and internally-generated feed (maize and
roughage from maize, beans, paddy, soybeans, peanuts and cassava) and thus have a GM of $0. But it
is possible to specify these activities in other ways: e.g., if half the maize output is intended for direct
sale while the remainder is to be used as resource for subsequent maize-using activities such as chips
or cow feed, those direct sales could well offset or more than offset the $4 maize production cost
(and so give this maize-growing activity a unit GM or p value of say -$3 or $0 or $10).
Kitchen processing of maize chips for sale in the village market and processing/selling milk, butter,
cheese are second-stage type (i) activities. In general, when there is a possibility of using a farm-
produced output in several ways, it is useful to view this as forming a resource 'pool'. Thus, in the
example of Table 9.11, maize grain flows into the 'maize pool' (at the rate of 0.3 tonnes of grain per
patok of maize) as a result of the maize-growing activity; it would then be withdrawn from the pool
by the activities of chip processing or feeding cows (at the rate of 1 tonne per tonne of chips and 0.7
tonnes per cow, respectively). These withdrawals will occur in the LP analysis according to the
relative total economic benefits to the whole system of using the resource for either purpose of chip
processing or feeding cows.
Regarding cows, this activity could be specified as 'keeping a cow and selling her milk' but this
would predetermine that this is the best use for the milk. It is better to set up a 'milk pool' in the B
column and let the analysis decide if and to what extent the most profitable use of the intermediate
resource is to sell it or to make butter, ghee, cheese or feed skim milk to animals etc., each of these
activities having a different GM and other resource requirements. Because only LP can consider and
evaluate all of these possibilities, only this planning method is really suitable in formulating 'optimal'
plans of the required degree of complexity for Type 2 farms.
There are no activities of type (iv) (resource + final product-generating) or type (v) (development) or
type (vi) (development + production) in the example of Table 9.11. There is only one type (vii)
activity; this consists of the simple purchase of green manure (the leaves of leguminous trees, litter
scraped up off the ground in the public forests). (These traditional methods of maintaining base soil
structure and fertility are still common in parts of Java and elsewhere in South Asia.)
Resources may also be obtained by barter - a type (viii) activity. In the example of Table 9.11, milk
can be bartered for (probably) scarce February field labour at the rate of 200 labour days per tonne
of milk - or, in more realistic terms, five litres of milk per day of labour. Therefore, one indirect limit
on February labour supply (and thus on the levels of crops using this labour, mainly maize and paddy)
would be the relative economics of those other activities of the system which also compete for milk,
i.e., production of cheese and butter.
Using household labour (additional to the available field labour) for field operations in the busy
February planting month is a typical transfer activity of type (ix). From Section 3.3.4, a farm's family
labour force is typically differentiated on the basis of the type of work each person is able or
expected or willing to do. Normally the housewife, older children and aged people look after the
livestock and kitchen processing/marketing activities (in addition to purely domestic chores) but they
might also be needed in the field at planting/harvesting times. Thus this transfer activity simply states
that if one such member of the household of four persons - which, in this example, excludes adult
males who are routinely occupied with only crops - is needed for field work, this will require the
reduction of household labour supply in February by 25 per cent and yield 28 additional days of field
labour in that month.
Disposal activities
As previously, one disposal activity is required for each resource/constraint row. These consist of a
diagonal line of '1' coefficients and carry a price of $0; they are not shown in Table 9.11 or discussed
further.
Double counting of the output of activities
The last activity on the right side of Table 9.11 is of a technical nature. As shown in the row labelled
'Cash income available for use in Phase II', the cash income generated by Phase I activities, and half
the income generated by the year-round dairy activities, is assumed to become available for meeting
the establishment and operating costs of Phase II activities. This technical activity specified in the
final column of the base table is a mechanism for ensuring that the income from other activities is not
counted twice, once as direct activity income (GM) and again if/when this income, as savings,
enables a second GM to be generated by some other activity.
The resources
The resources/constraints of Table 9.11 are largely self-explanatory. They are grouped into four
categories according to the period to which they refer. Some are initially present (at the start of
Phases I and II); some such as maize grain and milk must be generated; some such as field fertility
must be maintained (system sustainability, Section 6.2.7). Cash income and store-type resources can
be transferred forward through time from Phase I to Phase II. Flow-type resources (e.g., labour) must
in general be used as they become available, or go into disposal or non-use (but some of these can
sometimes also be transferred forward by structuring activities to allow sale and later repurchase of
their resource outputs, or by barter of a resource now for a similar resource in the future). Also, it
might be necessary to differentiate in resource use on the basis of quality, custom or tradition. Thus,
in the example of Table 9.11, household labour is specified as normally being available and required
only for keeping cows and dairy processing activities but, as need be, it could be specified as
transferable to some other uses (as is done in Table 9.11 for February household labour which, like
'other household labour', has its input-output coefficients specified here in terms of the percentage of
the total (fixed) amount available).
As shown by the respective coefficients of '1' and '-1' in the cash-transfer activity column of Table
9.11, cash for establishing and operating Phase II activities is generated by Phase I activities (being
careful to avoid double counting). Finally, the land resource must be precisely defined in terms of
time, quality and purpose. In the example of Table 9.11, 'land' means successively wet-season (Phase
I) land, dry-season (Phase II) land, year-round land and, specific to Phase II, land immediately after a
wet-season paddy crop has been removed from it.
Analysis and solution
Though the base table of Table 9.11 is much more elaborate than that of Table 9.9. it would be
processed in exactly the same way - first adding a p column, disposal activities and a K column, then
forming Z and Z - p rows, then selecting the most negative Z - p value etc. Too, in a real situation the
system would probably be given a longer assumed life and the program run over a planning horizon of
several annual cycles to determine the optimal plan after the system had stabilized; as noted above,
this would require rows/columns allowing the flow-on of resources to successive phases and cycles.
Solving such problems by hand is extremely tedious. However, they are readily solved by computer
using such software packages as GAMS (GAMS Development Corporation 1997) and What'sBest!ä
(Lindo Systems Inc. 1994). The more difficult and important task is to ensure that the farm is
modelled adequately, in particular that its activities are structured appropriately and that its
constraints are specified correctly.
Summary
Table 9.11 is probably a reasonable representation of production opportunities and the variety of
activities involved on highly mixed small farms. But it still does not reflect the second structural
characteristic of these farms - their subsistence orientation. This is discussed in the following section.
All the activities of Table 9.11 were optional: they might or might not enter the final plan, depending
on their relative contribution to farm cash TGM. However, the characteristic purpose of Type 1 farms
is to seek subsistence self-sufficiency directly by the production of household food and
household/farm materials as commodities rather than through the medium of cash-generating
activities. Thus the focus now shifts to the household; the planning objective is to meet its food and
material needs. In consequence, the system is now dominated by mandatory subsistence food- and
material-producing activities.
The activities
The production activities on a Type 1 farm fall into seven groups. Five of these are household
oriented and two are farm oriented.
Household-oriented activities consist of producing food, including beverages and condiments, and
non-food materials for direct use by the household component of the system as distinct from its farm
component (although it is often difficult to distinguish between materials intended for household use
and those intended as farm resources). Food-producing activities are often of a mandatory nature and
fall into four groups according to the types of commodities they produce (which of course differ
according to the physical environment, family preferences and custom):
(1) bulk staples/carbohydrates: e.g., grains, root crops, breadfruit, jackfruit.
(2) vegetable and animal proteins: e.g., beans, peas, meat, fish.
(3) vegetable and animal oils and fats: e.g., sesame, mustard, coconut, peanut, soybean, maize, palm,
butter, ghee.
(4) vitamins/minerals/taste: e.g., chilli, pepper, vegetables, paan.
As noted in Section 2.2.1 relative to the extraordinary product diversity found on many Type 1 and
Type 2 farms, a variety of non-food materials are produced for household and farm use as
subsistence items. Common examples are:
(5) bamboo (for construction timber, water pipes, furniture, flooring); coconut, kitul, aren and areca palms (for fibre, twine,
containers, tools); and stems of cassava, cotton, roselle (for fuel and trellising).
Farm-oriented activities are of the same various types as those discussed previously in Section 9.6.2
relative to Type 2 farms. Thus:
(6) final-product activities are intended to generate extra-subsistence income by sale or barter.
(7) resource-generating activities are intended to support all or some of the above activities (1) to
(6).
As indicated by the example of Table 9.11, most field and livestock activities will in fact generate
both final products and resources. Usually both these categories (6) and (7) will consist of sets of
optional activities.
General vs specific requirements
There are two ways of handling mandatory subsistence-oriented activities. If appropriate, the
requirements of such an activity might simply be subtracted from the initial stock of resources, then
the planning analysis is directed at finding the best mix of optional activities for the remaining
resources. This common-sense approach is adequate if the mandatory activity is predetermined - if,
e.g., the family must have 0.25 ha of paddy; exactly or at least 0.1 ha of beans; at least one patok of
sesame etc.
But from (1) to (4) above, there are many possible crops/activities/ways of meeting the household's
general requirements for each of the four food groups. Although meeting each of these general
requirements (carbohydrates, protein etc.) is mandatory, the specific activities for achieving this are
themselves optional. In theory there are so many ways of meeting, e.g., a household's carbohydrate
requirements that determination of the least-cost combination of crops/activities for achieving this
would itself require a large LP analysis. The same is true for the other food groups. However, in
practice - because of household preferences, tastes, custom as well as actual resource constraints -
the household will predetermine only some of the requirements in terms of specific crops, while
leaving its options open as to how the other general requirements can best be met.
As an example the household might specify the following conditions as the basis for formulating its
production plan:
· at least 0.5 ha of country paddy.
· at least 0.2 ha of pulses (but these can be from any crop/species/variety).

· at least 200 litres of vegetable oil (but this can be pressed from any oil source).
· at least 0.2 ha of chillies interplanted with leafy vegetables, but no more than 0.5 ha.
· maintenance of the existing grove of bamboo (for fuel, building timber, twine, fences, water pipes).
· any crops at all which will generate maximum net cash income (to meet household and farm needs
not met by the above items).
· any activities at all to generate the resources needed by the above items.
Example
This type of subsistence-oriented farm planning problem is set up for LP as was the problem of Table
9.9, but with certain modifications. An example is shown in Table 9.12. The steps are summarized as
follows:
(i) Distinguish clearly between maximum and minimum planning constraints or requirements. (Table 9.9 dealt only with maximum
requirements.)
(ii) For the (real) potential production activities:

· in section B of the worksheet, define the available resources and other constraints which would limit production or expansion of
the several real activities;
· construct a unit budget/requirements column for each of these real activities and enter the GM of
each in the top p row; and
· again as previously, construct a disposal column/activity for each of the resources/constraints of B
and attach prices to them which will be $0 unless the non-use of a resource actually bears a cost (as
is the case with water in the present example of Table 9.12).
Now, as illustrated in Table 9.12, the modifications begin:
(iii) For each minimum requirement of the plan:
· include this requirement as a planning constraint Q in the B column. Enter one Q row for each minimum requirement of the plan,
Q1, Q2 ... ;
· regard this minimum requirement also as an artificial activity and construct an activity column for it.
Designate this activity column Q1, Q2... to correspond with the minimum requirements already
designated as Q1, Q2... in the previous step;
· these artificial Q1, Q2... activity columns will consist only of a '1' coefficient at the intersection of
each Q row and its corresponding Q activity column;
· in the p row, attach a very high negative p value to each artificial Q activity; denote this notional
value by -M where this is unquantified in terms of money value but is understood to be higher than the
money price of any real (i.e., non-artificial) activity. Having done this, any Q activity carrying such a
very high and negative price -M is always the least profitable activity to have in the plan; it will
therefore be driven out in the subsequent LP iterations where the operational objective is to replace
low-value activities with relatively better ones (Section 9.6.1);
· as with the real production activities of (ii), construct a disposal activity column for each minimum
plan requirement Q and attach to this a price of $0 in the p row. Specify the unit coefficient of each of
these disposal columns at its intersection with the corresponding Q row as '-1' rather than as '1'
(which latter was the practice with the disposal activities/columns for real non-Q resources).
TABLE 9.12 - Example of LP Base Table for Whole-farm Planning of Small (Type 1) Farms
with Subsistence-oriented Mandatory Activities
The activities
On the example farm of Table 9.12 there are six possible production activities - cotton, paddy, chilli,
sesame, mustard and bamboo - to first meet household requirements and, if there are any resources
left over, to generate the maximum amount of net cash income. Typical requirements were discussed
above. In this example they are specified as:
· at least 0.5 ha of grain, but only paddy is acceptable and any paddy in excess of 0.5 ha can be sold.
· at least 200 litres of oil which might be obtained from 0.2 ha of either sesame or mustard; but while
any amount of sesame oil in excess of household requirements can be sold, there is in this village no
market for mustard oil; its sale price is $0.
· at least 0.2 ha of chillies for household consumption, but because the market for chillies is uncertain
the farmer will not take the risk of producing more than 0.5 ha; chillies thus have both a minimum and
a maximum limit.
· cotton to any amount as purely a cash crop.
· the present area of bamboo must be maintained.
As previously noted, the best way of handling any of these mandatory activities which involve no
alternative supply activities is simply to deduct their resource requirements from the initial stock of
resources on the farm. Thus, applying this to bamboo, total net available resources for entry into the B
section of Table 9.12 are calculated as follows:
Land Cash Water
Initial resources: 3.3 ha $250 250 units
Bamboo requirements: 0.3 50 0
Net resources (B values): 3.0 200 250

Table 9.12 is largely self-explanatory; only two points might be noted. First, on this example farm the
disposal or non-use of irrigation water carries a cost of $10. This is based on the fact that, if the
farmer does not use her or his full quota of water in any season, it will be reduced by this amount in
the next season and she or he must then buy it back or pay a fine of $10 for each unused unit. The other
disposal activities bear no such cost. Second, for each minimum requirement specified in the B
section, there will be a disposal activity which carries a zero price, and an artificial activity which
carries a very high negative price of -M. Finally, given this base table specified so as to
accommodate mandatory activities, the operating steps to determine the LP-based optimal farm plan
are the same as in previous examples: the Z and Z - p rows are obtained and the analysis proceeds as
in Table 9.9. While such analysis is clearly tedious (and prone to error) if done by hand, it is easily
carried out on a personal computer using available LP software such as, e.g., GAMS (Brooke,
Kendrick and Meeraus 1992) or What'sBest!ä (Lindo Systems Inc. 1994). The skill and artistry
involved reside in formulating the base table so as to adequately reflect or model the reality and
objectives of the farm being programmed.
As Sections 9.6.2 and 3 evidence, the application of LP to Type 1 and Type 2 farms can be quite
complicated - not in terms of calculation (which is easily done using available computer software)
but in terms of specifying the base table with its various types of activities and their unit budgets.
Such specification requires skill if not artistry and can be quite demanding of the analyst's time.
Obviously, the question must be asked as to whether the benefit gained is worth the cost involved. In
general, the answer to this relative to one-off analyses of small farms will be negative. However,
there are two situations for which the benefit may exceed the cost involved. The first is when the
analyst is concerned (either for farm advisory, development or government policy purposes) with a
recommendation domain consisting of a significant number of relatively homogeneous small farms.
LP analysis of a representative farm may then provide broad guidelines or insights relevant to all the
farms of the recommendation domain. The second situation is when the analyst may wish to have a
better professional understanding of the operation of various small-farm systems of concern.
Modelling of these systems by developing at least the LP base table (if not full LP analysis) of a
number of distinctive case-study farms may then greatly enhance his or her understanding of such farm
systems.
While there is no doubt that LP can enhance an analyst's understanding of a farm system, he or she
should beware of taking the LP-based optimal farm plan at full face value. In particular, any
significant difference between what a farmer is actually doing and what LP analysis suggests he or
she should be doing should not necessarily be attributed to farmer irrationality, ignorance or
inefficiency. Rather, such differences should be seen as a reason to review and possibly respecify the
LP analysis. Thus, in general, LP analysis should be seen as a starting point or guideline input to the
choice of a final plan; it should not be seen as the final arbiter to choice of a plan. Obvious reasons
for such caution are (i) the possibility that the LP base table does not properly specify the objectives,
activities and constraints pertinent to the farm system, (ii) the possibility that the input-output data
used in the LP analysis are not error-free and (iii) the reality that the real world of the farm system
involves both non-linear input-output relationships (particularly diminishing returns) and risk arising
from uncertainty about, in particular, yields and prices. The first of these deficiencies -
misspecification of the base table - may be lightened by dialogue with the farmer in developing the
LP base table; the second - errors in the input-output data - by the use of sensitivity analysis via
parametric LP (Hazell and Norton 1986, pp. 125-131; Rae 1994, pp. 109-114); and the third - non-
linear relations and risk - may be lessened by using more advanced forms of mathematical
programming, in particular risk programming as outlined in Chapter 11 and by Hardaker, Huirne and
Anderson (1997, Ch. 9).
9.7 REFERENCES
Brooke, A., D. Kendrick and A. Meeraus (1992). General Algebraic Modelling System: A User's
Guide, GAMS Release 2.25, World Bank, Washington, D.C.
Dent, J.B., S.R. Harrison and K.B. Woodford (1986). Farm Planning with Linear Programming:
Concept and Practice, Butterworths, Sydney.
Doll, J.P. and F. Orazem (1984), Production Economics: Theory with Applications, 2nd end,
Wiley, New York.
GAMS Development Corporation (1997). GAMS (v2.25), GAMS Development Corporation,
Washington, D.C.
Hardaker, J.B., R.B.M. Huirne and J.R. Anderson (1997). Coping with Risk in Agriculture, CAB
International, Wallingford.
Hazell, P.B.R. and R.D. Norton (1986). Mathematical Programming for Economic Analysis in
Agriculture, Macmillan, New York.
Heady, E.O. and W. Candler (1958). Linear Programming Methods, Iowa State University Press,
Ames.
Lindo Systems Inc. (1994). What'sBest! User's Guide, Lindo Systems Inc., Chicago.
Mao, J.C.T. (1969). Quantitative Analysis of Financial Decisions, Macmillan, London.
Rae, A.N. (1994). Agricultural Management Economics: Activity Analysis and Decision Making,
CAB International, Wallingford.
Rickards, P.A. and D.J. McConnell (1967). Budgeting, Gross Margins and Programming for
Farm Planning, Professional Farm Management Guidebook No. 3, ABRI, University of New
England, Armidale.
Schrage, L. (1991). UNDO: An Optimization Modeling System, 4th edn, Wadsworth, Belmont.
Winston, W.L. (1991). Introduction to Mathematical Programming Applications and Algorithms,
PWS-KENT Publishing Company, Boston.
10. PLANNING FARM SYSTEMS OVER
TIME
10.5 SUBJECTIVE (INTERNAL) AND OBJECTIVE (EXTERNAL) INTEREST RATES
10.8 ANNUITIES: EVALUATING REGULAR COST AND RETURN STREAMS OVERTIME
10.9 PERPETUITIES: EVALUATING REGULAR COST AND RETURN STREAMS WITH AN
INDEFINITE LONG LIFE
10.11 SUMMARY OF PROCEDURES FOR FINANCIAL EVALUATION OF LONG-TERM
INVESTMENTS
10.15 REFERENCES
'... and that old common arbitrator, Time ... '
Shakespeare, Troilus and Cressida
So far, except for consideration (i) of resources providing a flow of services over time in Section
3.3.4, (ii) of depreciation as an element of capital fixed costs in Section 5.4 and (iii) of the time
frames relevant to whole-farm planning in Section 9.1, the time dimension of whole-farm systems and
their subsystems has not been considered. It has been assumed that all farm activities occur
concurrently within a time frame of a single season or year so that their performance can be evaluated
and compared without reference to the length of time over which their resources are tied up or their
output generated. However, planning and evaluation of activities extending beyond the coming season
or year generally require that their time dimension (and associated uncertainty) be explicitly
recognized and taken into account - and the more so, the longer the timespan of the activity, regardless
of whether it is of a resource- or product-generating nature or a capital investment in the farm service
matrix.

Time plays a pervasive role in all production activities but in none more so than agriculture. The
reason for this is that, because of its biological nature, agriculture is very time-dependent. Compared
with industrial production, agricultural production both takes more time and is generally time-related
in terms of the seasons. Thus, e.g., while grain production requires some three to six months of time
and the correct seasons, factory production of a car requires only a day and is independent of such
seasonal elements as rainfall, temperature, day length etc.
In terms of farmers' decision making, the influence of time occurs in five ways (Dillon and Anderson
1990, Ch. 6; Doll and Orazem 1984, Ch. 7). First, there is what can be called the flexibility effect. By
its nature, time gives the farmer flexibility in the form of options that would not be available if
production were instantaneous. These options relate to the operation of both resource- and product-
generating activities. They relate to the fact that the output of any activity is dependent on both the
pattern of input injections over time and the pattern of output harvests over time. Thus the yield of a
tea crop varies with both the frequency and level of fertilizer application as well as the time-pattern
and intensity of plucking. Within the timespan of any particular activity, decisions about its operation
(including marketing of the product, if relevant) can be revised or new decisions taken, perhaps on a
sequential contingency basis relating, e.g., to market or climatic conditions. In particular, relative to
production activities, time gives the farmer the opportunity to sequence input injections and/or output
harvests in varying ways. Thus the farmer has to decide not only the amount of fertilizer, irrigation
water, livestock feed etc. to be used, but also the pattern of its use over the coming production period.
Should he or she, e.g., apply all of a crop's fertilizer at sowing or should its application be spread
throughout the growing period or should it be applied only if rain is imminent? While, theoretically,
such questions fall in the domain of response analysis (Chapter 8), they are not easily answered
analytically (Dillon and Anderson 1990, Ch. 6). Generally, they are decided on the basis of trial-and-
error experience combined with an appreciation of the constraints imposed by agronomic and
biological logic. Thus farmers know that it is best if fertilizer application is complemented by rain
and that it is best if livestock receive feed daily rather than less frequently.
Second, in many activities time may be a variable input under the farmer's control, i.e., a decision
variable to be manipulated by the farmer through his or her choice of a time-length of run for the
production process (Dillon and Anderson 1990, Section 6.8). This is particularly the case for
livestock-fattening and aquaculture activities, as well as for many vegetative crops, and also relative
to the replacement decision for long-term perennial tree crops and woodlots. Such time-dependence
of output, of course, always exists in agriculture. Often, however, it can be ignored because the input
of time required is fixed as, e.g., in the case of most grain crops once a particular cultivar has been
chosen for planting. Note also that time may have an indirect effect on production through the
influence of input-carryover effects as, e.g., when there is a carryover of nutrients from fertilizer
application in prior production periods. The extent of such carryover is a function of the time elapsed
and may influence the farmer's current decision on input use (Dillon and Anderson 1990, Section
6.9.4).
The third influence of time is that, again by its nature, it introduces uncertainty into farmers' decision-
making. Because of the complex processes of change, evolution and growth that accompany the
passage of time, farmers cannot be sure of the future either in terms of the outcomes of today's
decisions or of the opportunities (including new products and technologies) that may become
available or of the developments that may occur in their family, social, political and economic
environments. Given the information available to him or her, the best that a farmer can do is to make
those decisions which are expected to best achieve his or her goals. Until it becomes possible to
remember the future (i.e., time becomes reversible!), there can be no guarantee that these decisions
will, with hindsight, turn out to have been the best that could have been made. While uncertainty is an
inexorable consequence of future time, for simplicity in exposition, planning over time is here
considered without regard to the presence of uncertainty. Planning of farm systems under uncertainty
is considered in Chapter 11.
The fourth influence of time arises from the fact that time is a resource which offers options in the
way other resources can be used. Thus a farmer might either invest savings from this year's income in
a bank at some rate of interest or reinvest these savings in the farm. Likewise, year by year over its
potential lifespan, the owner of a plot of coffee has to decide whether it would be better to keep the
existing stand for another year or to replant it with a new stand or to replace it by some other activity.
Such choices imply alternatives in the use of time and hence an opportunity cost for time. Thus the old
adage that 'time is money'. This opportunity cost of time is measured as a rate of interest (or,
equivalently, as a time-preference discount rate reflecting the degree to which future payments need
to be discounted to give their equivalent present value). This interest rate may be determined
objectively as the rate at which money would grow over time if invested in the best available
alternative. Often this objective interest rate can conveniently be taken as the market rate, i.e., the
price that has to be paid to borrow money per unit of time. Such an objective interest rate is useful for
purposes of economic rationalism, i.e., to assess in money terms the actual time-opportunity cost of
alternative investments. More generally, however, farmers measure the opportunity cost of time
subjectively (as is their right) in terms of their personal time-preference interest or discount rate
which (as outlined in Section 10.5.2) corresponds to their rate of time-preference tradeoff between
present and future income.
Fifth, over time, the value of money, i.e., its purchasing power, may change. Usually this takes the
form of inflation whereby the value of money decreases so that fewer goods can be bought than
before with the same amount of money. Sometimes, but only rarely, deflation or an increase in
purchasing power may occur. In the presence of inflation or deflation, the rate of interest needs to be
corrected to allow for their effect. This correction is easily made: if the nominal rate of interest is i*
per annum and w denotes the annual rate of inflation (w > 0) or deflation (w < 0), the corrected or
real rate of interest i is [(1 + i*)/ (1 + w) - 1] per annum. Thus, if there is an annual rate of deflation
of four per cent (i.e., w = -0.04) and the nominal rate of interest is ten per cent, the real annual rate of
interest will be [(1 + 0.10)/(1 - 0.04)] - 1 = 0.146 or 14.6 per cent. Note that if the rate of inflation w
exceeds the nominal rate of interest i*, the real rate of interest or growth will be negative. Except for
Example 1 in Section 10.4.1 below, throughout the remainder of this presentation it is assumed that
the rate of inflation or deflation is zero so that i* = i.

When time is a variable input to an activity, the farmer has to decide how much of it to use, i.e., for
how long to let the activity run before replacing it either with a new run of the same activity or a
different activity. In making such decisions, two principles should be borne in mind. The first is that a
dollar in the future is generally worth less than a dollar today; and the further into the future the dollar
is, the less it is worth today. The objective rationale for this is that to receive a dollar in the future,
less than a dollar needs to be invested today. It may also be justified subjectively on, e.g., the basis
that 'a bird in the hand is worth two in the bush'. This principle leads to the need to use interest-rate
or discounted cash-flow procedures in the planning and evaluation of activities extending more than a
year or two into the future.
The second principle relevant to the optimal use of time is that for processes/activities whose time
length t is at the discretion of the farmer, his or her choice of t will be optimal when (a) the present
value of the activity's marginal profit per unit of time at t is equal to (b) the amortized flow of profit
which could be obtained by allocating that last unit of time to the best available alternative activity
(which may be a new run of the same activity) (Dillon and Anderson 1990, Section 6.6.2). Part (a) of
this criterion corresponds to the additional gain to be made by continuing the current activity for one
more unit of time; part (b) is the cost incurred in the form of forgone profit by not using this last unit
of time for the best alternative activity. Thus equality between (a) and (b) implies that the basic
requirement for optimality in resource use and profit maximization, i.e., that marginal revenue equals
marginal cost, is satisfied. Conversely, so long as (a) is greater (smaller) than (b), the activity should
be continued (replaced). This criterion applies whether the time-length decision variable t is a matter
of weeks (as in the case of broilers) or of months (as in pig fattening) or of years (as for tree crops).
However, when t is only a matter of weeks or months so that time preference is not significant, the
criterion for optimal choice of t simplifies to the requirement that marginal profit per unit of time be
equal to average profit per unit of time (Dillon and Anderson 1990, Section 6.6.1; Doll and Orazem
1984, Ch. 7). Application of this simpler criterion is illustrated in Figure 10.1 which depicts a single
run of a repetitive time-dependent production process. Profit or net gain as a function of the time-
length t of a run of the activity is shown by the curve OAB. Maximum profit from a single run of the
activity occurs at B with t = OH. In contrast, for repeated runs of the activity, maximum profit per unit
of time occurs at A where the slope of the profit curve (i.e., marginal profit per unit of time) equals
maximum average profit per unit of time AG/OG and t = OG. Thus the criterion for optimal t is
satisfied at A. Clearly OG is less than OH. This reflects the fact that maximizing profit from a
sequence of runs of an activity implies that the time-length of each run should be shorter than if the
activity was only to be run once. The logic of this is that, in a particular run, as inputs are used
beyond t = OG, marginal profit per unit of time is less than the maximum average profit per unit of
time that could be obtained by using these inputs in a new run of the activity beginning at time G.
FIGURE 10.1 - Profit in relation to Time for a Particular Run of a Repetitive Production
Process having Time as a Decision Variable in the Absence of Time-preference
The influence of time through its opportunity cost on the long-term planning of farm systems is the
topic of this chapter. For simplicity, future outcomes are assumed to be known with certainty. In large
part, assessment of time opportunity-cost effects involves the use of a variety of interest-rate or
discounted cash-flow procedures which are standard to financial and actuarial analysis - see, e.g.,
Alien (1956, pp. 228-237), Chisholm and Dillon (1988), Mao (1969, Chs 6 and 7), Rae (1977, Chs
9, 10 and 11), Robison and Barry (1996) and Thuesen (1957, Chs 3 and 4). If done by hand, these
interest-rate procedures can involve much tedious calculation and chance of error. However, they are
readily available in general spreadsheet programs for personal computers such as, e.g., Microsoft
Excel.

Possible types of farm activities were outlined from a planning perspective in Section 9.3.1. In terms
of their timespan, all these types of activities fall into one of the following three groups:
Short-term activities have a useful lifespan of a year or less. Typical examples are annual crops such as grains, pulses and oilseeds,
vegetables etc. and such livestock activities as pigs, broiler chickens and pond fish when these are produced on an 'all in-all out'
batch basis. However, when livestock activities are self-generating (i.e., more-or-less permanent herds of cattle, flocks of sheep
etc.), they are better regarded as long-term activities.
Intermediate-term activities fall somewhere between short- and long-term activities; the dividing
lines are not fixed. Their lifespan often depends on how they are managed. Thus most fodder crops
can be grown either as a one-year annual or as a semi-perennial crop maintained over two, three, four
... years. This also applies to some grains: e.g., sorghum in South Asia is commonly ratooned to give
three, four, five crops over two to three years and thus becomes an intermediate crop. Likewise,
sugarcane is a short-term crop in Java, an intermediate two- to four-year crop in Australia, but a long-
term crop of 15 years or more in southern Sri Lanka.
Long-term activities are those having a long or indefinite useful life. The lower limit is not precisely
defined and the upper limit ranges from around 25 years for coffee and cacao to 65 to 80 years for
coconut to over 100 years for the older varieties of tea and nutmeg.
Resource-generating activities also fall into these same classification groups, according to the length
of time over which the resources are generated. Thus investment in the construction of a farm
reservoir would be regarded as a long-term activity because it would generate a resource, irrigation
water, over its life of 30, 40... years. However, the time dimension of activities is not always
obvious: e.g., investment in the construction of a chicken shed might be a short-term activity if it is
built only roughly using coconut atap intended to last only one year, or an intermediate activity if it is
built using higher quality sago atap lasting four years, or a long-term activity if it is built using mud
bricks.
In both ex ante planning and ex post evaluation of the results of short-term activities, the time
dimension may often be ignored. This will be the case when the purpose is comparison of activities
which all have approximately similar lifespans: e.g., when the economics of a four-month paddy crop
are to be compared with a six-month maize crop. Here small time differences do exist but their
recognition would make little difference in the comparative evaluations. The results can be
expressed, e.g., as 'crop net returns of $500 per ha vs $400 per ha', without specifying that the first
crop occupies the land for four months while the second uses six months of land resource.
A second situation occurs when all the activities are again 'short-term' but there are nevertheless
significant time differences between them; e.g., when a four-month paddy crop is to be compared with
an 11-month cassava crop. Obviously it would now be misleading to express the comparative results
as, e.g., '$500 per ha from paddy vs $700 per ha from cassava'. Here a more accurate comparison
would introduce time as a specific dimension of returns, in which case the real comparison would be
expressed as $500/4 = $125 per ha-month from paddy vs $700/11 = $64 per ha-month from cassava,
which conveys quite a different picture.
At high latitudes and elevations, or where, because of limited water supply, the growing season is
short, usually only one crop annually is possible. This fact will be understood and there is no need to
attach a time dimension to the analysis. But in the wet tropics and where year-round production is
possible, specification of a time dimension in the analysis of activities is often mandatory.
There is another situation where time can be ignored. This is in the analysis of the seasonal or annual
economics of long-term activities, e.g., assessment of the current annual returns from tree crops
(coconut, rubber etc.) and from time-stable livestock activities such as a dairy herd. In this type of
analysis these long-term activities are treated as if they were short-term activities. Using such
methods as presented in Chapter 7, they are evaluated relative to some short-term period; the fact that
the respective activities might have had quite different cost/return profiles in the past (and may again
have in the future) is irrelevant for the purpose of evaluating their current performance.
The planning of intermediate- and long-term activities requires a different approach. Here the
problem is to evaluate the likely future performance of, e.g., a new coffee activity over its entire
future life of up to 25 years or a coconut activity over its future life of up to 70 years. However, in
some situations ex ante evaluation may need to extend over only part of the activity's future life, e.g.,
over only the first 15 years of a new or proposed coffee crop, or over the next ten years of an old
coffee crop.
Long-term ex ante evaluation is required for a range of purposes, e.g.:
(i) to determine the economics of some planned or potential long-term crop or livestock activity without reference to other
comparative activities. Here the operating objective is to determine expected profitability (future expected returns vs future
anticipated costs) and the time profiles of these future return and cost streams, e.g., to determine the future break-even point of
the investment.
(ii) to allow comparisons among alternative crop or livestock investment possibilities. Here the
operating objective is to bring the cost and return streams of all the alternatives to some common
basis in time so that, e.g., the average annual return from a 20-year crop can be compared with that of
a 35-year or a 65-year crop.
(iii) to obtain similar comparisons between long-term crops on the one hand and sequences of
alternative short-term crops on the other.
(iv) to determine the optimal age at which to replace a long-term crop by a new activity or to
determine whether an existing activity should be replaced by a new long-term activity.
Procedures for evaluating long-term activities are discussed below. In essence, they reduce to either
(a) the process of discounting, i.e., the transference of future cost and return payments backward
through time to the present or, less frequently, (b) the process of compounding, i.e., transferring
present and future costs and returns forward through time to a common future point in time. This is
done in order to bring streams of payments having different time profiles to some common basis of
equivalence so that they can be compared.


In its simplest and most usual form, compounding is the procedure of taking some initial amount or
present value (PV) forward through time to find that future amount An which is equivalent to PV (i.e.
to which PV will grow) when PV increases at some uniform annual rate i over some specified
number of years n. Note that PV may be positive (i.e., a credit or a revenue amount) or negative (i.e.,
a debit or a cost amount). Given that in the first year PV will grow to (PV+iPV) = PV(1+i), in the
second year from PV(1+i) to [PV(1+i)+iPV(1+i)]= PV(1+i)², and so on, the general formula for
compounding is thus
An = PV(1+i) n.
This procedure of taking some initial sum through time to find its future equivalence can be applied to
a wide range of problems. Compounding factors (1+i)n are listed in Table 10.1 for a range of
interest or growth rates i and time periods n. Note that, throughout the presentation of this chapter, it
is assumed that periods are measured in years so that i and n are on an annual basis, and that the
compound interest rate i is constant over time. It is also assumed that any payments made (i.e., costs)
or received (i.e., revenues) occur at year's end. If these various assumptions do not hold, the
procedures outlined need to be adjusted appropriately1 (Chisholm and Dillon 1988).
1 In particular, if compounding (or discounting) is to occur q times per year and the interest rate per period of length 1/q years is r,
then the annual interest rate i equivalent to r is given by:
i = (1+r) q - 1.
Conversely, if i and q are given,

r = (1+i) 1/q - 1.
If the annual rate of interest is i and compounding (or discounting) is carried out q times per year,
then:
An = PV [1 + (i/q)]nq.
Example 1. If a family places $100 in a savings account or invests it on the farm or lends it to a
neighbour, in each case at an earning rate of nine per cent annually, in the absence of inflation or
deflation this $100 sum will in four years increase to an amount of A4 = $(100)(1 + 0.09)4 or, taking
the compounding factor (1 + 0.09)4 = 1.412 from Table 10.1, A4 =$(100)(1.412) or $141. Should
there be an annual inflation rate of six per cent, the real rate of interest will be [(1.09)/(1.06)] - 1 =
0.028 or approximately three per cent. In real terms, i.e., in terms of today's dollars (rather than the
inflated dollars of four years' time), the $100 will grow to $(100)(1.03)4 = $112.
Although most commonly used in a financial context, the compounding factors of Table 10.1 are also
applicable to non-financial growth problems.
Example 2. As a result of his or her development program, a farmer's annual paddy yield, now three
tonnes per ha, is increasing at an average annual rate of six per cent. What yield can be expected in
five years' time? Applying the compounding formula, A5 = 3(1 + 0.06)5 = 3 x 1.338 (Table 10.1), the
farmer can expect four tonnes per ha.
Often both the initial sum or PV and the equivalent future amount An are known and the rate of
increase is to be determined.
TABLE 10.1 - Compound Growth Factors (1+i)n
Source: Chisholm and Dillon (1988, Appendix I).
Example 3. Paddy yields on a sample of farms are now six tonnes per ha; ten years ago they were
only 3.5 tonnes. The average annual rate of increase i has therefore been such that 6 = 3.5(1+i)10 or,
rearranging, 1.71 = (1+i)10. From Table 10.1, this equality condition is met at a growth rate i of about
0.055 or 5.5 per cent.
Discounting is the reverse of compounding. It is the procedure for finding the present value, denoted
PV, equivalent to some known future amount An to be paid or received n years from now. As would
be expected from rearranging the formula for compounding, the formula for discounting is:
PV = An(1+i)n
where i is the annual discount rate or interest rate. Discount factors 1/(1+i)n = (1+i)-n are the
reciprocals of the compounding or growth factors of Table 10.1. Values of the discount factor
1/(1+i)n for a range of i and n values are given in Table 10.2.
Example 4. A farmer has fuelwood trees which will mature in four years' time and will then be worth
Rs 500. What is their PV if the farmer's subjective discount rate (see Section 10.5 below) is nine per
cent? Under these conditions the trees have a PV of Rs (500)/(1 + 0.09)4 = Rs (500)(0.708) (using
Table 10.2) or Rs 354.
Example 5. In Example 2 above, crop yield was increasing. But on some other farm, because of
continuing soil erosion or salinity etc., yield may be decreasing. If present yield is four tonnes and
declining at an annual rate of six per cent, what will yield be in five years' time? This can be found
approximately by using the reciprocal of the relevant value from Table 10.1 as PV = 4 x 1/(1 + 0.06)5
= 4 x 1/1.338 or, more directly, by using the relevant discount factor from Table 10.2 as 4 x 0.747 =
2.99 tonnes. Note that this example involving a negative growth rate has been conveniently
approximated by reversing the time order and thus converting the problem to one involving positive
growth. The accurate solution to this problem of negative growth would be found as A5 = 4(1 - 0.06)5
= 4 x 0.945 = 4 x 0.734 = 2.94 tonnes.
Many activities involve a stream of future income and/or costs rather than simply a single future item
of income or cost.
Example 6. A farmer has an old kitul palm which is nearing the end of its useful life, but which will
still yield toddy to the value of Rs 100, 70 and 40 over the next three years and which, at the end of
the third year, can be cut down and sold as fuelwood for Rs 400. However, the village merchant
would like to buy the palm immediately for making axe handles. If the farmer sells now and has a
discount rate of nine per cent, what minimum amount should he or she charge the merchant? In the
farmer's hands, the palm represents the following stream of nominal income values by years:
Year (n): 0 1 2 3 3
Nominal income (Rs): 0 100 70 40 400
Discount factor 1/(1 + 0.09)n: 1 0.917 0.842 0.772 0.772
Discounted income equivalent (Rs): 0 92 59 31 309

Assuming the nominal income payments are received at the end of their respective years, as shown
above their PV can be ascertained by applying the relevant discount factors from Table 10.2. To the
farmer the PV of this activity is thus Rs (0 + 92 + 59 + 31 + 309) = Rs 491. He or she would have to
charge the merchant at least this price to obtain an income equivalent to that obtainable from
continuing to tap the palm for a further three years.
Note that in the above cash-flow tabulation. Year 0 has been included for completeness. It
corresponds to the immediate present (end of Year 0 or start of Year 1) when, in some situations,
payments may occur.
10.5 SUBJECTIVE (INTERNAL) AND OBJECTIVE (EXTERNAL) INTEREST

RATES

So far, the opportunity cost of time as measured by the rate of interest or discount i has been assumed
to be given. Before proceeding further, it is useful to consider the basis and justification for using
some particular rate rather than another. The appropriate rate to use in any particular problem may be
set by subjective (internal) factors or by objective (external) factors (Pearse and Turner 1990, Ch.
14). Some problems will involve consideration of both types of rate (as in Example 9 below).
Objective interest/discount rates are set by factors external to the value structure of the farm
household. Money which can be put in a bank account at ten per cent or borrowed in the souk at 30
per cent is an obvious example. The farm family can only accept or reject such market-set rates. Or
income might be re-invested in the farm where it will return 15 per cent and this also is determined
by factors external to the household (although they are internal to the farm). This on-farm rate is also
fixed, at least in the short run, although obviously the family, as farmers, can work to change it over
time.
Objective or external rates may be set by such external factors as the local supply and demand for
credit; or, if invested on the farm, by crop yields and prices; or by government regulation, custom,
tradition or religious law. Being external and thus not under the control of the farmer, the future
values of objective interest rates are subject to uncertainty. Such uncertainty is considered in Section
11.16.
Subjective interest rates are set internally by the value structure of the farm household itself and thus
vary both over the life of the household and between households. Whereas objective rates are given
and explicit, subjective rates are implied in what goals the farm family pursues - its preference for
consumption or saving; whether it prefers one activity over another; whether it judges a high-return
but exploitive farm plan as superior or inferior to a low-return but sustainable plan. All these various
attitudes, and their expression as decisions affecting the use of resources and outputs, can be
quantified - at least implicitly - in terms of the farm family's rate of time-preference tradeoff for
present vs future income, consumption or other rewards (Doll and Orazem 1984, pp. 251-255).
The subjective time-preference interest rate or discount rate is determined by many factors. Beyond
the general influence of the external interest-rate environment, some of the more important of these are
probably the following:
(i) Poverty or wealth in relation to a family's material needs. Other things equal, a poor family will usually have a greater
preference for near-term income than for income that is to be received at some distant time in the future. It will therefore have a
high discount rate for future income as compared with more immediate income and will tend to adopt a farm plan which yields
short-term rather than long-term returns. If the degree of poverty is great, the family will possibly pay no attention to the long-
term sustainability of its farm system. Only after today's needs are met can there be concern about resources for tomorrow.
Poverty can thus be a significant cause of resource degradation (Chopra and Rao 1992; von Braun 1992).
(ii) Opportunities for making income-increasing investments, on or off the farm. Where good
investment opportunities exist, the family will have a high demand for capital and other resources to
exploit them. If borrowing is limited or too expensive, the family's time-preference for current farm
income, from which the additional capital has to be saved, will be high. Where investment
opportunities are few (e.g., on the already highly developed paddy farms of Java), there will be low
demand for investment resources and time-preference will tend to be relatively low (although this
might well be offset by (i) above or the following factors).
(iii) Age and the planning horizon. An individual family might not be much interested in future
relative to near-term income because the decision makers are old. Their time-preference will be
relatively high but again this might be cancelled by the other factors listed.
(iv) Social and cultural environment. This factor establishes the general framework within which the
other factors operate. In some societies, saving and the accumulation of wealth is highly regarded; in
others, social status is achieved through present consumption, lavish expenditure and generous giving.
(v) Personal characteristics of individual families. These characteristics occur within the
framework of (iv) and refer to relative personal values as these determine attitudes to change and
development, saving, investment, risk etc. Basically they are of a psychological nature.
TABLE 10.2 - Discount Factors 1/(1+i)n
Source: Chisholm and Dillon (1988, Appendix II).
To summarize, the way in which a family manages its farm resources and chooses between activities
which have a short- or long-term payoff implies some quantitative level of preference for near-term
income vs future income. This time-preference is usually positive: most families attach a higher
subjective value to $100 to be received today than to $100 to be received in a year's time. Such
preference is reasonable since $100 today can be invested to yield an amount of $(100)(1+i) in a
year's time. However, time-preference might sometimes be negative: some families will place a
higher value on the conservation of their farm for future support of their children than on their own
immediate consumption of income (above some necessary minimum). The direction and degree of
time-preference will be determined by factors (i) to (v) above. In turn, the direction and degree of
time-preference can be measured in terms of the subjective interest rate which time-preference
implies. Once this subjective interest/discount rate is measured - by implication of what farmers
actually do or the choices they actually make - it can be applied to a wide range of planning problems
on their individual farms.
Example 7. The farmer of Example 4 could cut his or her trees in four years' time and then sell them
for Rs 500. Suppose he or she also has the option of selling the trees now, at the lower price of Rs
300 because they are still immature. If he or she in fact makes this latter choice, what subjective
discount rate is implied by this decision? In the farmer's eyes, Rs 500 in four years' time is equivalent
to or has a PV of Rs (500)(1+i)-4 = Rs 300 so that the discount factor (1+i)-4 = 0.6. For a term of four
years this discount factor of 0.6 is found by interpolation from Table 10.2 to correspond to a
(subjective) interest rate of 0.14 or 14 per cent. But the farmer might be willing to sell the trees now
for an even lower price. Suppose this final rock-bottom price is Rs 250. His or her real discount rate
then implies a discount factor of 250/500 = 0.5, which for a four-year term is found from Table 10.2
to correspond to an interest rate of (about) 19 per cent. It will be noted that this decision by the
farmer has nothing directly to do with the discount or interest rates which prevail in the outside
world. The equivalence of Rs 250 now and Rs 500 in the future exists only in the farmer's mind, and
it arises from some combination of the underlying factors (i) to (v) above which are operative for this
particular individual or household. If the farmer were poorer or richer, or had more or fewer
opportunities for investment, or lived in a different socio-cultural environment, he or she might well
have some other subjective discount rate.
Example 8. If three families were to be asked whether they would prefer to sell a calf now for $100
or wait and sell the full-grown ox at a higher price in four years' time, they would probably reply that
(because of their positive time-preferences) they would in general prefer the money now, but that it
depended on how much higher than $100 the future price was to be. Family A, a very poor family
concerned with meeting its immediate needs, might say that it is willing to wait four years for the
income only if it could then get $250 for the ox. Family B, which has good opportunities for
developing its farm if it invests money now, might be prepared to wait if the future expected value of
the ox is at least $200. Family C, which is well off, might say it would be happy to wait if a future ox
price of only $150 could be anticipated. Given this information and using the relation An = PV (1+i)n,
by interpolation from Table 10.1 the implied subjective interest rates for the families A, B and C are
respectively 0.26, 0.19 and 0.11.
Example 9. In some situations it is necessary to consider both subjective and objective interest rates.
Suppose, e.g., the government plans to introduce some (unrelated) agricultural development program
in which the farmers of Example 8 can participate if they are willing to bear the cost. The program is
guaranteed to give participants an annual return of 15 per cent (an external rate) on their outlay. If
families A, B and C are equally typical, how many families could be expected to participate? A
comparison of their subjective interest rates with the external rate of 15 per cent indicates that only
well-off families like family C (or one third of farmers) might be interested. For families A and B,
with their respective discount rates of 26 and 19 per cent, the guaranteed future return of 15 per cent
is not high enough to be attractive. Respectively, they would need a return of at least 26 or 19 per
cent.

Discussion so far has concerned the equivalence of future revenues or returns. Exactly similar
procedures are used respectively to determine the present value of future costs or, less frequently, the
future value of present costs. There are three main situations in which such calculations are
necessary.
First, in order to compare the future income stream from a long-term activity against its costs, and so
determine the activity's net return over its lifetime, it is necessary to bring both to the same time basis.
The most convenient common basis is usually the PV of each of these streams (although other time
bases are also possible.)
Example 10. A farm activity will mature and return the farmer Rs 1 000 in four years' time. It will
also require outlays on inputs of Rs 80 now and Rs 100 in three years' time. What are activity net
returns if the relevant interest rate is nine per cent? Problems such as this are conveniently sketched in
an input-output time chart as shown in Figure 10.2. On an annual accounting basis (Chapter 7), the
time profile of net returns to the activity over the next four years would be Rs -80, 0, 0, -100, 1 000.
But if concern is with the performance of the activity over the whole of its life rather than in its one-
year phases, the PV of total returns is Rs (1 000)(0.708) = Rs 708 and the PV of total costs is Rs
[(80)(1.000) + (100)(0.772)] = Rs 157. The activity therefore has a net worth of Rs (708 - 157) = Rs
551 in terms of PV. Note that, as made clear by the time chart of Figure 10.2, in applying interest-rate
procedures it is important to specify whether payments occur at the start or end of the relevant time
periods.
FIGURE 10.2 - Example of an Input-output Time Chart
The second situation involving the discounting of future cost occurs in making provision for the future
replacement/repair of items of physical farm capital (Section 5.4).
Example 11. An irrigation channel will require desilting costing Rs 1 000 in five years' time. What
provision should be made now for this future expenditure if the interest rate is nine per cent? Using
Table 10.2, a depreciation fund of Rs 650 if established now would generate the necessary Rs 1 000
in five years' time.
The third situation arises where a single future stream of costs is to be evaluated, or where two or
more such streams are to be compared.
Example 12. A farmer has budgeted out the likely per unit costs of establishing passionfruit and
cinnamon activities. These data are listed below. Given the farmer has a time-preference rate of ten
per cent and assuming that the respective costs are incurred at the end of each year, which activity has
the higher establishment cost?
Establishment Cost at End of Year (n) Nominal Total PV

Activity
0 1 2 3
Passionfruit (Rs): 0 1 200 500 300 2 000 1 729
Cinnamon (Rs): 0 300 900 800 2 000 1 617

In nominal (i.e., undiscounted) terms, both crops have the same cost of Rs 2 000 over the
establishment period of three years. However, because of the farmer's time-preference and the
difference in sequencing of the respective cost streams, the PV of the cost of passionfruit
establishment (Rs 1 729) is higher than that of cinnamon (Rs 1 617).

Planning of a long-term activity involves finding the PV of the activity's stream of annual net returns
(i.e., annual 'return minus cost' differences). This is known as discounted cashflow analysis.
Example 13. The separate annual cost and return streams for the first seven years of a proposed
cardamom activity are shown in lines 1 and 2 of Table 10.3 and annual return minus cost differences
or net returns in line 3. Note that net returns are constant (in nominal terms) in this example for Year 6
onwards. It is necessary to apply the appropriate discount rate of, say, ten per cent only to line 3. The
relevant discount factors (from Table 10.2) are shown in line 4. Extended to the full 30-year life of
the activity, financial evaluation might be largely in terms of line 6 which shows cumulative total
discounted net returns to the end of each year. (However, as discussed in Section 10.13, there are
also other criteria pertinent to evaluation of long-term investments.) Note that Year 7 is the first year
having a positive cumulative PV, i.e., it will take seven years before the PV of the returns' stream
exceeds that of the costs' stream. The break-even period for this cardamom investment is thus seven
years.
TABLE 10.3 - Example of Discounted Cash-flow Analysis
Year
Item
0 1 2 3 4 5 6 7
(1) Costa ($) 500 300 900 800 200 200 200 200 ...
(2) Returna ($) 0 0 0 100 500 1 000 1 600 1 600 ...
(3) Net return ($) -500 -300 -900 -700 300 800 1 400 1 400 ...
(4) Discount factorb 1.000 0.909 0.826 0.751 0.683 0.621 0.564 0.513 ...
(5) PV of net return ($) -500 -273 -743 -526 205 497 790 718 ...
(6) Cumulative PV ($) -500 -773 -1 516 -2 042 -1 837 -1 340 -550 168 ...
a Payments are assumed to occur at year end.

b From Table 10.2 with i = 0.10.
As noted previously, discounted cash-flow evaluation of a long-term activity will extend over the full
30-, 40-, 50-... year life of the activity if the total crop cycle is of interest. Or it might be limited to
some shorter period of only, say, five or ten years. This will depend on analytical circumstances and
the planning horizon of the farmer which might be much less than the lifespan of a very long-term
crop.
10.8 ANNUITIES: EVALUATING REGULAR COST AND RETURN STREAMS

OVERTIME

So far the costs and returns used in the examples have been either single discrete quantities or some
irregular series of payments occurring at points in the life of the activities. Fortunately, as in the
example of Table 10.3, many long-term activities become fairly stable after their establishment years;
from this point on they often generate rather uniform streams of annual costs and returns. From that
point in time when they become uniform - or can be assumed to be uniform without excessive
distortion of the facts - they can be evaluated as annuities. This greatly reduces the required
calculation effort. Indeed, it often makes practical sense to regard cost and return streams as annuities
even when, strictly speaking, they are not.
An annuity is a series of equal annual payments which occur over some medium or long but limited
time period. These payments are usually of a future nature and may be either incoming (i.e., revenues
and thus positive) or outgoing (i.e., costs and thus negative). Annuities are particularly useful in
evaluating long-term production activities or investments. The usual problems to which annuities are
applied involve either (a) finding the final terminal amount which the annuity will generate over its
life or (b) finding the PV of this future amount or, equivalently, of the annuity.
The terminal value An of an annuity A payable at the end of each year for a period of n years is the
value to which the series of annual payments will accrue if, as each is received, it is invested at the
compound rate of interest i until the end of Year n. This terminal value is obtained as:
An = A[(1+i) n - 1]/i.
Values of the annuity factor [(1+i)n - 1]/i are easily calculated using the compound growth factors of
Table 10.1.
This terminal value An actually consists of two components: (i) the total of all successive uniform
payments over the life of the annuity, i.e., an amount equal to nA, and (ii) the total interest earned by
all of these payments on the assumption that each payment is invested at some uniform interest rate
over the remaining life of the annuity. Thus the terminal value of a $100 annuity over five years at
nine per cent would consist of $100 received at the end of the first year with interest on this for four
years; plus $100 received in the second year with interest on this for three years etc. The total amount
by the end of the fifth year would be $599 made up of $500 from the annual payments plus $99
interest earned on these payments. This terminal value is calculated, using the relevant annuity factor
[(1 + 0.09)5 - 1]/0.09 = 0.539/0.09 = 5.989, as A5 = $(100)(5.989) = $599.
Example 14. A stand of coconut palms is expected to yield an annual net return of $200 over the next
20 years. As each of these income payments is received, the farmer will invest it at a rate of eight per
cent. At the end of the 20 years, what sum will have been accrued by this coconut + investment
activity? This terminal value is found as A20 = $(200)(1.0820 - 1)/0.08 or $(200)(45.75) = $9 150.
Note that this value is in nominal dollars of 20 years hence. It is not in terms of dollars in hand today.
For a presumed interest rate i, the annual payment or annuity A equivalent to a lump sum An due at the
end of n years is obtained as:
A = Ani/[(1+i) n - 1].
This is known as the sinking-fund formula since it gives the amount A which should be put aside at
the end of each year 1, 2, 3... n and invested at an annual interest rate i to the end of Year n (i.e.,
invested in a 'sinking fund') in order to meet a lump-sum commitment due at the end of Year n. This
lump-sum commitment might be, e.g., a debt, a dowry or a sum needed to replace worn out or
obsolescent capital items. Values of the sinking-fund factor i/[(1+i)n - 1] are easily calculated using
the compound growth factors of Table 10.1.
For most purposes, whether an annuity represents a stream of future net income (as in Example 14) or
of costs, the value of the annuity will need to be assessed on the basis of its PV. Thus, if the coconut
activity of Example 14 is to be compared with other activities which have different lengths of life and
different time profiles of costs and returns, it is necessary that all these activities be assessed on some
standard time basis. The most convenient such basis is usually their PV. This is calculated as:
PV = An(1+i) n = A[(1+i) n - 1]/[i(1+i) n]
where, as before, payments of size A occur at the end of each year over the n-year life of the annuity
and i is the presumed rate of interest. The annuity discount factor, represented by the quotient
[(1+i)n - 1]/[i(1+i)n] in the above equation, is given in Table 10.4 for a range of i and n values.
Example 15. What is the PV of the 20-year stream of coconut + investment income of the previous
example? As noted, this is a 20-year annuity of $200 at eight per cent. The PV of this is found by
applying the appropriate annuity discount factor from Table 10.4: PV = $(200)(9.818) = $1 964. At
first glance there is such a large difference between the terminal value of this annuity ($9 152.
Example 14) and its PV equivalent of only $1 964 that a check on the results might seem warranted.
For this the $9 152 amount, which is due in 20 years' time, can be discounted directly using the
appropriate discount factor from Table 10.2: PV = $(9 152)(0.214) = $1 958. Apart from a slight
difference due to rounding, no error is present. The large difference between the value of $9 152 in
20 years' time and the value of $1 964 today indicates dramatically the effects of time and the
discount rate in the valuation of long-term activities.
Example 16. If the cardamom proposal of Table 10.3 has a lifespan of 30 years, what is its PV if i =
0.10? To the end of its break-even period, Year 7, the activity has a PV of $168. Thereafter, the
activity constitutes an annuity of $1 400 for n = (30 - 7) = 23 years. The value of this annuity
discounted to the start of Year 8 (or end of Year 7) is $(1 400)(8.883) = $12436. Discounting this
value back a further seven years to the present, its PV is $(12 436)(0.513) = $6 380. Thus the full PV
of the 30-year cardamom activity is $(168 + 6 380) = $6 548.
Example 17. An ageing family needs funds but at the same time wants to reduce its work load. Its
members will not sell their land, but are considering selling leases to operate the following activities
on their land: (a) a stand of coconut palm with 15 years of useful life remaining and returning $100
net annually; (b) a block of coffee with ten years of useful life remaining and returning $50 net
annually; and (c) a grove of fuelwood trees which will be ready for harvesting in six years' time and
will then be worth $150. The old coconut palms will also have a residual value of $120 as
construction timber in the 15th year. If the family specifies a discount rate of nine per cent, what are
reasonable prices at which to sell the leases for these respective activities? For the coconut activity,
annual net returns constitute a 15-year $100 annuity with, from Table 10.4, a PV of $(100)(8.061) =
$806 while the sale of timber for $120 in Year 15 has, using Table 10.2, a PV of $(120)(0.274) =
$33. The price for the 15-year coconut lease should thus be at least $(806 + 33) = $839. Annual net
returns from the coffee activity constitute a ten-year $50 annuity with a PV of $(50)(6.418) = $321 so
the price of the ten-year coffee lease should be at least this much. Likewise, the sale of fuelwood in
six years' time for $150 implies a six-year lease price of no less than $89 for this activity. These
prices would only be attractive to buyers who had a discount rate of less than nine per cent. Thus a
buyer with a discount rate of seven per cent would value the 15-year coconut lease at $(100)(9.108)
+ (120)(0.362) = $954 which compares favourably with its suggested offer price of $839.
Note that if the family in the above example were to sell its land, as distinct from the activities
supported by that land, and if it can be assumed that such land will generate a more-or-less uniform
and constant stream of future income, then they would be selling a perpetuity rather than an annuity.
Perpetuities are discussed in the following section.
10.9 PERPETUITIES: EVALUATING REGULAR COST AND RETURN STREAMS

WITH AN INDEFINITE LONG LIFE
A perpetuity is an annuity that continues on forever. Strictly speaking, there is no such thing;
however, many cost and/or return streams can be assumed to be perpetual and this greatly facilitates
their evaluation. The PV of a perpetuity is obtained as:
PV = R/i
where R is the perpetual annual uniform payment (either revenue or cost) and i is the relevant interest
rate.
Example 18. If the appropriate interest rate is ten per cent, what is the PV of a hectare of paddy land
that annually earns a net return of $500? Assuming that the land is going to remain in this use
indefinitely, its annual yield of $500 is an annuity without apparent end, thus a perpetuity. Its PV is
$500/0.10 = $5 000.
Example 19. Assuming an interest rate of ten per cent, what is the PV of a hectare of young seedling
tea expected to yield a net return of $600 annually? Seedling tea has a very long lifespan and may
well continue producing for 80, 90,... years. Its income stream is not a perpetuity but is close to being
so. Its value therefore can be estimated as $600/0.10 = $6 000. In fact, whether the tea is regarded as
a perpetuity or as a very long-term annuity would not make much practical difference. Valuing it as a
50-year annuity would give a PV of $(600)(9.915) or $5 949 which, given the uncertainties attached
to such a long period, is in a practical sense about the same as $6 000.
TABLE 10.4 - Annuity Discount Factors [(1+i)n - 1]/[i(1+i)n] for obtaining the Present Value of
an Annuity: PV = A[(1+i)n - 1]/[i(1+i)n]
Source: Chisholm and Dillon (1988, Appendix III).

Perpetuities are used widely in the valuation of land and farms. There are three main alternative
bases for such valuation: market value; value based on cost of production of the land or similar asset;
or economic value based on future productivity of the resource.
Obtaining a market-based value is purely empirical: a piece of land is worth exactly what some
'willing buyer' is prepared to pay for it. If a willing buyer is not on hand, then reference is made to the
prices which other willing buyers have recently paid for similar land. Such a market-based value
might approximately correspond to productivity-based value (below), or differ considerably from it.
In commercial Western societies there is usually a fairly close relationship between market-based
and productivity-based values. But in most Asian societies people are usually willing to pay higher
prices for land than would be justified by reference to its productivity alone. This is because they are
purchasing land for its other attributes as well, i.e., its properties as a secure store of wealth, as an
insurance policy against bad times or for the social prestige which land ownership confers.
In agricultural societies where a land market has not yet emerged and where there is little market
trading of agricultural products, the value of agricultural land is probably best determined on the
basis of its cost of production - which usually implies the value of the labour required to create new
farms from a virgin environment (as, e.g., in new settlement areas of Kalimantan and West Irian).
In commercial situations, land and farms are usually valued according to an estimate of the value of
their future productivity. Here 'value' has an economic basis rather than a 'market opinion' or cost-of-
production basis. As noted in Example 18, this future productivity is regarded as a perpetuity. (But
nonetheless the legal profession in most Western countries prefers valuations based on market price.)

From Section 10.8.1 the terminal amount An to which an n-year stream of uniform future payments
will grow is found by evaluating it as an annuity; the PV of this terminal amount (and thus of the
annuity) can then be found by discounting it. Or the PV of the annuity can be found directly by using
the annuity discount factors of Table 10.4. Amortization is the reverse of this procedure. A PV or
lump sum is amortized - i.e., liquidated or literally 'killed' - by taking it forward through time as an
equivalent stream of equal (usually annual) payments. The formula for finding the size of each of these
equal future payments A is:
A = PV[i(1+i) n]/[(1+i) n-1]
where A is the annual payment and PV is the sum to be amortized over n years at an interest rate of i.
As would be expected, this formula is the reverse of the formula for the PV of an annuity A. The
amortization factor [i(1+i)n]/[(1+i)n - 1] is simply the reciprocal of the annuity discount factors of
Table 10.4.
Probably the most common use of amortization is in calculating the repayment schedule of a fixed-
term loan, i.e., 'killing off the amount borrowed by repaying it in equal annual instalments.
Example 20. A farmer borrows a sum of $400 at nine per cent compound interest and agrees to repay
it with a series of equal annual payments over 15 years. What will be the size of each uniform
payment A? From Table 10.4 the amortization factor is 1/8.061 = 0.124. Each payment will thus be
$(400)(0.124) = $49.60. Moreover, since each uniform payment consists of some amount of principal
and some amount of interest, and since the farmer will repay a total of $(49.60)(15) = $744 while he
or she borrowed only $400, the farmer will pay a total interest charge of $(744 - 400) = $344 on this
loan.
Amortization is especially useful in bringing irregular series of future costs or returns to the
equivalent basis of a uniform series. In this way, series having different time profiles can be
compared as uniform streams rather than simply on a lump-sum PV basis.
Example 21. A farmer can install either of two alternative irrigation systems: system A which
involves constructing a dam and pumping water through sprinkler pipes; or system B which involves
pumping directly from a creek and using gravity-flow surface application. Both systems will yield the
same increase in crop returns, $500 per year. But these systems will have different construction and
future maintenance costs for periodic desilting of channels, replacement of pipes and pumps etc. The
farmer wants to know: (a) the relative annual cost of each system and (b) the expected increase in
annual net returns using each system. Two other pieces of information are necessary: the length of the
farmer's planning horizon and the appropriate interest rate. Although both systems would - if
maintained and periodically renovated - have an indefinite life, the farmer specifies the planning
horizon to be 20 years and the relevant interest rate as 15 per cent. The time-schedule of construction
and periodic system maintenance costs is given as follows:
Planning year: 0 1 7 10 15
System A costs ($): 0 1 000 200 600 100
System B costs ($): 600 500 0 400 400
Discount factor: 1 0.870 0.376 0.247 0.123

The first step is to bring each cost stream to its equivalent total PV by discounting each future cost
item (using Table 10.2) and summing. These total PV sums and their amortized equivalents are:
Total PV of Costs Amortization Factor Amortized Annual Value of Costs
System A: $ 1 106 0.160 $ 177
System B: $1 183 0.160 $ 189

Increased crop returns from both systems are already on an annual basis, $500 per year. To compare
system costs with these returns it is therefore convenient to bring costs to this same annual basis.
System A has a nominal total cost of $(1 000 + 200 + 600 + 100) = $1 900 but when these separate
items - which occur at different points over time - are brought to a common basis (by discounting),
they have a total PV of $1 106. But returns are on an annual basis; so to compare returns vs costs, the
costs of $1 106 must be 'spread out' or amortized over the 20 years of the investment. The annual
equivalent of $1 106 is $(1 106)(1/6.259) = $177 (using Table 10.4). The annual-equivalent costs of
system B are found in the same way as $189; and both can be compared with the annual increase in
returns of $500. Thus system A implies an annual increase in net returns of $(500 - 177) = $323
while system B yields $311. If the farmer wants only to compare the two alternatives on a total basis,
i.e., over the full 20-year planning horizon, amortization is not necessary. The comparison can be
based on the total PV of the cost streams, $1 106 and $1 183, vs the PV of the associated returns
stream of $500. $500 annually over 20 years is an annuity: its PV for i = 0.15 is $(500)(6.259) or $3
129 (Table 10.4). The comparison then is $ 1 106 vs 1 183 vs 3 129.
Amortization can also be used to calculate the regular amount that should be invested at the end of
each year in a depreciation fund to cover the replacement cost of capital items as they become worn
out or obsolescent.
Example 22. A farmer's irrigation pump will need to be replaced ten years from now at a cost of $1
000. How much should the farmer invest each year in a depreciation fund to finance this replacement
if the relevant interest rate is eight per cent? The amount to be invested at the end of each year
constitutes a ten-year annuity and may be found directly by applying the sinking-fund formula of
Section 10.8.2. Thus A = $(1 000)(0.08)/(1.0810 - 1) = $69. Alternatively, as shown in Figure 10.3,
the annual depreciation cover may be found by first finding the PV of the $1 000 replacement cost and
then amortizing this amount to determine the needed equivalent annuity. Thus PV = $(1 000)/(2.159) =
$463.18 so that A = $(463.18)(0.08)(2.159)/(1.159) = $69.
10.11 SUMMARY OF PROCEDURES FOR FINANCIAL EVALUATION OF LONG-

TERM INVESTMENTS
The interest-rate procedures outlined in preceding sections are summarized in Figure 10.3 which
shows schematically the relationships between PV, A and An assuming an annual interest rate of i and
a period of n years. Application of these discounted cash-flow procedures to the evaluation of a long-
term farm activity or investment implies the following sequential steps:
(i) Determine the time-perspective from which the activity is to be evaluated: whether ex ante, limited ex ante, current year or ex
post.
(ii) Define the time period over which the activity is to be analysed. This might or might not
correspond to (iii) below.
(iii) Define the time profile of the activity. This is conveniently done in terms of a yield curve
sketched over the life of the activity (as in Figure 10.4 below). Data for such yield curves may be
available from records or obtained by means of a survey of farms with comparable conditions.
(iv) Based on this time profile, prepare a set of budgets giving the cost/return streams for the activity
over the time period to be analysed.
(v) Bring all relevant cost/return streams from (iv) to a common basis in time using an appropriately
chosen interest or discount rate.
(vi) Select and apply appropriate evaluation criteria to (iii), (iv) and(v). There are several
alternative criteria (Section 10.13 below).
FIGURE 10.3 - Relationships between a Terminal Value (An) due at the End of Year n, its
Present Value (PV) and an Equivalent Annuity (A) for an Interest Rate of i
Before applying these steps to an illustrative example, the time-perspective of evaluation warrants a
little further discussion. An ex ante perspective is required if the problem is the forward planning of
an activity over its entire future life. A limited ex ante perspective is appropriate if planning
concerns only some part of the future life of the activity: the first 20 years of life of a 70-year coconut
crop, or the next ten years of an already mature nutmeg crop etc.
An ex post perspective is required if the problem is to evaluate the past performance of some existing
or past crop, over all or pan of its lifespan. Such backward-looking analysis is usually limited to
historical studies of old crop stands, but the purpose might be to obtain agro-economic guides to the
planting of future crops. Ex post analysis in the form of comparative analysis is routinely applied to
short-term crops and farms growing such crops (Chapter 7). It is also applied to the immediately past
phases of long-term crops (current-year perspective, below). It is not often applied to full evaluation
of all past phases of long-term crops (e.g., for a fall evaluation of a tea crop planted in 1910), but it is
often applied to some particular aspect of the historical past; e.g., the only way to determine the likely
future impact of typhoons on yields and returns of coconut in parts of the Philippines is to analyse the
effects of past typhoons on past crops - and the farther back in time such studies can be extended the
more valid will be the guides obtained for future planning (Chapter 11).
A current perspective is required if all past and likely future events can be ignored and attention
restricted to what happened to an activity in the immediate past few months or year and what will
happen in the next few months or year. The time dimension is ignored in such analyses.

The general steps (i) to (vi) above are now applied to the evaluation of a proposed cardamom
activity.
(i) Time-perspective: A proposed cardamom activity is to be evaluated (a) to determine the probable economic performance of the
investment measured in financial terms (money returns in relation to money costs) and the break-even point in time when
returns will have covered costs, and (b) to obtain these data in a form which will allow financial comparison of this activity with
other possible investments. The time-perspective is thus ex ante.
(ii) Time period: For planning purposes, the cardamom activity is postulated to have an economic
lifespan of 25 years from planting. It is judged by the farmer that it would then be more profitable to
replace the cardamom rather than continue it for another year.
(iii) Time profile: The time profile of a long-term crop such as cardamom is best constructed in two
stages. A yield curve (Figure 10.4) is sketched over the period extending from initial land preparation
to the end of the plants' useful life; then - as per (iv) below - budget tables of inputs, costs, outputs
and returns are prepared on the basis of this curve. The yield curve allows the sequential economic
crop phases to be distinguished and cost and return items identified as, e.g., from the example
cardamom yield curve of Figure 10.4:
Phase 1 (Years 1 to 3): land preparation, planting and annual maintenance costs; no returns.
Phase 2 (Years 4 to 15): routine maintenance plus harvesting, processing and marketing costs; yields
annually increasing.
Phase 3 (Years 16 to 30): same as Phase 2 but yields decreasing.
(iv) Activity budgets: Three budgets are needed. The first, a base budget as presented in Table 10.5,
details the costs of crop establishment and maintenance on a per unit (here per ac) of activity basis
over the years from planting up to when crop maintenance costs become stable, here from Year 1 to
Year 4 inclusive. Because labour is usually the most important input, it is a useful practice to show
labour units explicitly, not just labour cost. (The budget can then easily be up-dated by applying new
unit-labour costs as these might change in future years.) For most long-term crops, annual maintenance
inputs/costs will be fairly uniform from the year in which the crop begins to produce (its year of
maturity) to the end of its useful life, here Rs 930 annually as shown in Table 10.5 for Year 5
onwards.
The second base budget, as shown in Table 10.6, refers to the unit costs of harvesting, processing and
marketing the variable amounts of annual output. This is a 'side' budget which specifies the above
costs on a per unit of output basis, not on a per unit of land basis (as was done in Table 10.5),
because the levels of output per ac to which these unit costs are applied will change annually
according to the yield curve of Figure 10.4. (Of course, if the yield curve is fairly constant over the
crop's lifespan, output and returns can also be budgeted on a per ac of crop basis.)
The third budget, as presented in Table 10.7, is a tabulation of returns and costs for each year of the
activity. It is derived from the base data of Tables 10.5 and 10.6 and Figure 10.4. The object is to
obtain line 5, the stream of annual net returns. These net returns are nominal undiscounted amounts. In
themselves they have little significance until they are transformed to a common basis in time.
(v) and (vi) Evaluation of equivalent net returns: Bringing the cardamom cost and return streams of
Table 10.7 to a common basis in time and criteria for their evaluation are considered in the following
Section 10.13.
FIGURE 10.4 - Yield Curve for Cardamom Investment Example
TABLE 10.5 - Inputs and Costs for Establishment and Maintenance of Cardamom (per Acre
Basis)
Establishment Year 5 onwards

Item
Year 1 Year 2 Year 3 Year 4
Labour units (days/operation/year):
Thin jungle 30
Clear planting points 20
Roads, paths, drains 10 5 5 4
Plant shade 2
Plant cardamom 35
Weed 25 20 20 20 20
Fertilize 10 10 10 14 14
... . . . . .
... . . . . .
Pest control 2 2 3 3 5
Total labour units: 140 52 64 51 60
Total labour costs (Rs): 1400 520 640 510 600
Material and services (cost/operation/year) (Rs):
Plants 250 50
Chemicals, sprays 20 30 50 50
Fertilizer 100 120 120 240 240
Transport 50
... . . . . .
... . . . . .
Tools 40 30 20 20 20
Total material and services costs (Rs): 480 240 200 350 330
Total all costs (Rs): 1880 760 840 860 930

Evaluation criteria for farm systems and subsystems were discussed in Section 6.2 (stability,
diversity, flexibility, sustainability etc.). These apply to all activities, enterprises and whole farms. In
particular, relative to long-term investments, sustainability is of crucial importance from both a
private and a social perspective. It is in both the farmer's and society's interest that her or his
activities do not cause the degradation of resources either on the farm itself or external to the farm.
Long-term investments should therefore always be evaluated from a sustainability and environmental
perspective (Markandya 1994). There are also other selection/evaluation criteria of a financial nature
which apply specifically to long-term activity/system investments. The most important of these are
listed in Table 10.8 and briefly discussed in the context of the cardamom example of Table 10.7.
TABLE 10.6 - Derivation of Cost of Harvesting and Processing Cardamom (Rs per Cured
Pound)
Basis: a gang of pickers harvesting 440 lb of green cardamom per day
Cost/440 lb green
(Rs)
Harvesting
Field supervisor 15
Contract picking 300
Processing
Labour for weighing and drying 70
Fuel for drying 140
Labour for grading and packing 10
Sacks 40
Total direct cost of harvesting and processing 575
Cost per lb of green cardamom 1.31
Cured-to-green recovery rate ('out-turn') 20%
Cost per lb of cured cardamom 6.55
TABLE 10.7 - Revenue and Cost Streams for Cardamom over 25 Years (Rs per Acre)
Year Nominal Total

Item
1 2 3 4 5 6 25
(1) Revenue 0 0 0 700 1 050 1 400 1 890 55 370
(2) E + M costa 1 880 760 840 860 930 930 930 23 870
(3) H + P costb 0 0 0 145 218 290 392 11 475
(4) Total cost 1 880 760 840 1 005 1 148 1 220 1 322 35 345
(5) Net revenue -1 880 -760 -840 -305 -98 180 568 20 025
a E + M = establish and maintain; cost from Table 10.5.
bH + P = harvest and process; cost calculated as cured yield (from Figure 10.4) multiplied by Rs
6.55 (from Table 10.6).
Financial evaluation of long-term farm activities or investments can be on the basis of a number of
factors, depending on the Mode, Field and purpose of the analysis (Section 2.1). In all except the
most simple cases, evaluation would be by reference to a set of factors rather than to only one. As
they are derived for the present cardamom example, these factors are listed in Table 10.8. Factors (1)
to (8) are relevant for on-farm or on-estate decision making (i.e., farm management in Field A).
Factors (9) to (11) would also be considered if this cardamom activity were part of some larger
public investment, such as a publicly funded settlement scheme (i.e., farm management in Field C). In
this latter case, the listed factors would be used, e.g., to decide whether the best crop on which to
base the scheme is cardamom or robber or tea etc. In the following discussion of the criteria of Table
10.8, an annual interest rate of ten per cent is assumed for discounting and amortization.
TABLE 10.8 - Example of Some Factors relevant to the Choice and Evaluation of Long-term
Investmentsa
For farms and estates (Farm Management in Field A) Cardamom
(1) Total life of the activity from establishment: 25 years
(2) Time to maturity (i.e., initial yield): 4 years
(3) Time to full maturity (i.e., maximum yield): 15 years
(4) Operating break-even point: Year 6
(5) Investment break-even point: Year 14
(6) Activity establishment costs per acre (Years 1 to 4): Rs 4 340
(7) Activity net PV per acre: Rs 2 641
(8) Amortized annual equivalent of net PV per acre: Rs 291
For public projects (Farm Management in Field C)
(9) Benefit-cost ratio: B/C = 1.22
(10) Foreign exchange benefit-cost ratio: B/C = 2.16
(11) Internal rate of return on the investment: IRR = 15.3

a Based on the cardamom example of Figure 10.4 and Tables 10.5, 6 and 7.
(1) Activity total life, here specified as 25 years, might or might not be relevant to a farmer/estate
manager. A long productive lifespan is generally a positive attribute, but not if the farmer's planning
horizon is limited; or if the pace of genetic improvement is such that an estate manager knows that,
whatever variety is planted now, it will be able to be replaced by a superior one in the relatively
near future (as, e.g., with rubber).
(2) Years-to-maturity, here four years, would be an important consideration on poor farms if the
activity under consideration is a food crop (jak, breadfruit etc.). On very poor farms it might well be
the dominant one; factors (4), (5), (7), (8) might be irrelevant in this case.
(3) Years to full production maturity, here about 15 years, is also sometimes a critical factor, e.g.,
inhibiting the planting of some of the palms which are notoriously slow developers, regardless of
how productive they might eventually become and how low their cost of establishment.
(4) Years to operating break-even point, here Year 6, when current annual return first exceeds
current annual costs. This would be an important consideration on farms/estates experiencing cash-
flow problems.
(5) Years to investment break-even point, here Year 14 when, for the first time over the activity's
life, the cumulative PV of total costs is exceeded by the cumulative PV of total returns. This factor,
too, would be important for farms/estates having cash-flow problems. Where the investment funds
must be borrowed, the interest burden on this debt over the period up to when the investment reaches
its break-even point might well be the single most important factor inhibiting development.
(6) Establishment costs, particularly for labour in the initial year(s), might encourage or discourage
adoption of the activity independent of its possible attractiveness in future years.
(7) Activity net PV is obtained by summing the discounted values of the items of line 5 of Table 10.7.
This is the first of the financial values by which the activity might be evaluated. As derived in column
(8) of Table 10.9, the PV of the proposed cardamom activity is Rs 2 641 per acre.
(8) Amortized net PV is factor (7) transformed to an annual basis over the life of the investment. Here
it is Rs (2 641)(0.110) = Rs 291 (using the amortization factor for ten per cent over 25 years, 1/9.077
= 0.110, from Table 10.4).
TABLE 10.9 - Example of Worksheet for obtaining the Benefit-cost Ratio and Internal Rate of
Return of a Long-term Investmenta
Total Cost (Rs/ac) Total Return (Rs/ac) Net Return (Rs/ac)
(1) (2) (3) (4) (5) (6) (7) (8)

Year
Discounted Nominal Discounted Nominal Discounted
Nominal
at 10% at 13% at 10% at 13% at 10%
0 0 0 0 0 0 0 0 0
1 1 880 1 709 1 664 0 0 0 -1 880 -1 709
2 760 628 595 0 0 0 -760 -628

3 840 631 582 0 0 0 -840 -631
4 1 005 686 616 700 478 429 -305 -208
5 1 148 713 623 1 050 652 570 -98 -61
6 1 220 688 586 1 400 790 672 180 102
7 1 293 663 550 1 750 898 744 457 235
8 1 365 636 513 2 100 980 790 735 344
9 1 452 616 483 2 520 1 069 839 1 068 453
10 1 535 591 452 2 940 1 133 866 1 405 542
. . . . . . . . .
. . . . . . . . .
25 1 322 122 62 1 890 174 89 568 52
Total 35 345 12 121 9 539 55 370 14 762 10 675 20 025 2 641
a Based on the cardamom example of Table 10.7.
Other factors can be obtained from extensions of Table 10.9, e.g., assuming that borrowed funds are
to be used, additional relevant factors might be the total amount of the necessary loan and its total
interest cost. (Note, however, that many farmers do not like borrowing, or will not or cannot borrow
beyond a certain limit regardless of how profitable the investment might eventually turn out to be.)
The remaining three factors of Table 10.8 are relevant for the evaluation of large public investment
projects, i.e., to farm management in Field C rather than Field A (Section 2.1.7).
(9) The benefit-cost ratio (B/C) is used primarily for making rapid comparisons among roughly
similar alternative investments - see, e.g., Gramlich (1990) and Sinden and Thampapillai (1995). In
the present example 'benefits' are simply the activity gross return stream, line 1 of Table 10.7. This is
transposed to column (4) of a new table, Table 10.9. Here each annual gross return item is discounted
at ten per cent and the results shown in column (5). They are then summed to obtain total discounted
gross return over 25 years of Rs 14 762. The same is done for the annual cost stream of line 4 of
Table 10.7 and the results shown in column (2) of Table 10.9. These also are then summed to obtain
total discounted activity cost of Rs 12 121. The B/C ratio for i = 0.10 is then 14 762/12 121 = 1.22.
As calculated, this ratio is purely a measure of an investment's financial attractiveness. It is of limited
validity as a measure of an investment's total desirability in relation to factors (1) to (8), or factor
(10) below, or - most importantly - in terms of the environmental or social merit or impact of an
investment.
(10) The foreign exchange B/C ratio is the same as (9) except that only those cost and return
components of the investment which require or generate foreign exchange are now considered. One
would proceed by inspecting the base budget Tables 10.5 and 10.6 and identifying all those items
which must be imported - machinery, fuel, fertilizer etc. - and arranging these as a new 'foreign
exchange costs' line in Table 10.7. That part of total output which is to be exported would be
similarly identified to provide a new 'foreign exchange returns' line. The analysis then proceeds as
before. As shown in Table 10.8, the proposed cardamom activity has a foreign exchange B/C ratio of
2.16 as compared with its standard B/C ratio of 1.22. The proposal is thus much more attractive in
foreign exchange terms.
Of course, small farmers themselves will not usually be concerned with the foreign exchange aspects
of an activity. But commercial estates in some countries are allowed to retain some proportion of
foreign exchange earnings and this can greatly influence their choice in enterprise selection. For
analysis in support of public agricultural development projects, the foreign exchange dimension is not
infrequently the dominant factor in evaluation.
(11) Internal rate of return (IRR). This is that rate of discount which makes the PV of an investment's
cost stream equal to the PV of its revenue stream so that the net PV of the investment is equal to zero
(Sinden and Thampapillai 1995, Ch. 9). The unknown is the discount rate. (By contrast, in finding the
net PV or the B/C ratio of an investment, the discount rate is specified.) As for other types of
discounted cash-flow analyses, calculation of IRR is provided by personal computer spreadsheet
programs such as, e.g., Microsoft Excel. The procedure for calculation of IRR by hand is illustrated
in Table 10.9. Some trial rate, say ten per cent, is first applied to the nominal (undiscounted) cost and
return streams and columns (2) and (5) are obtained, the totals of which are respectively Rs 12121
and Rs 14 762. Discounted returns are considerably greater than costs at ten per cent; therefore some
higher rate is now applied, say 13 per cent. The results of this are shown in columns (3) and (6).
Discounted returns of Rs 10 675 are still higher than discounted costs of Rs 9 539 but not by much.
The equality condition is then found graphically by plotting PV against i as shown in Figure 10.5. The
point of intersection of the discounted cost and return curves indicates an IRR of 15.3 per cent.
FIGURE 10.5 Graphical Determination of IRR
One note of caution is necessary. In analysis to obtain B/C ratios and IRR values at farm or estate
level, objectivity and honesty on the part of the analyst can usually be assumed. This may not be the
case where these apparently objective procedures are applied in support of investments intended
incidentally to serve some vested interest or to justify investment decisions which have already been
taken. By fiddling with assumptions concerning future costs, prices, the time profile of the activity
etc., it is possible (within limits) to come up with whatever B/C ratio or IRR will best serve the
purpose of the analyst or client. Not infrequently these measurements are predetermined and the
technical-economic conditions of the activity (prices, yields etc.) then tailored to achieve them.
(Among practitioners of such art, an IRR range of 17 to 27 per cent seems to be preferred ground.)

Three major factors make whole-farm planning over time difficult and not as reliable as would be
wished. Indeed, beyond the immediate short term of the coming season or year, in situations of
dynamic change induced by research and development these factors may often make fruitful farm-
system planning virtually impossible. Two of these factors have already been mentioned (Section
10.1): first, the flexibility effect whereby time gives the farmer options for the sequencing of
activities' input injections and output harvests within years and of activities between years and,
second, the inevitable presence of uncertainty. The effect of these factors is compounded by the fact
that the longer the planning period, the greater the number of options to be considered and the amount
of uncertainty to be faced. This uncertainty facing farmers in their planning can have many sources. It
may relate to Nature (yields as affected by the vagaries of climate, pestilence etc.), to the economic
environment (technology, prices, interest rates, inflation), to government policy (subsidies, quotas,
labour laws etc.), to personal matters (health, family relationships) or to the socio-political
environment (social unrest, civil war etc.). Particularly for small farms of Type 1 (subsistence) and
Type 2 (semi-subsistence), such uncertainty is multiplied in situations of change where non-
traditional farming methods and activities are becoming available or induced as a consequence of
agricultural research and/or social change (improved education, better communications, population
pressure etc.). Prior to these modem developments, such small farms by trial and error over the
centuries had established traditional farm systems well-oriented to their needs and in equilibrium
with their natural and social environments. Clearly, disturbance of these traditional systems brings
with it significant uncertainty in planning.
The third factor that makes whole-farm planning over time difficult is really due to the combined
effect of the above two factors. It is known as the curse of dimensionality. By this is meant the
exponential expansion in the number of possibilities to be considered as the length of the whole-farm
planning horizon increases. Thus, e.g., suppose a farmer can grow any of six crops. Assuming that
only one crop can be grown at a time (i.e., mixed cropping is not allowed), planning for a one-year
horizon would involve consideration of six crop options; for a two-year horizon, six by six or 36
two-year crop-sequence options; for a three-year horizon, 63 or 216 such possibilities. Clearly, even
without consideration of crop mixtures or of uncertainty with its array of possible outcomes, the
whole-farm planning problem can rapidly become so large as to be unmanageable or to be so
complicated that the cost of planning exceeds its benefit. Thus, while multi-period mathematical
programming techniques may be applied to the long-term planning of whole-farm systems (Rae 1977,
Ch. 10), it is doubtful if the benefits are worth the cost on an individual farm basis. This is
particularly so if significant uncertainty is present and the planning horizon is longer than two or three
years. The exception to this is if the whole-farm system is so simple as to not suffer the curse of
dimensionality in its planning. This is most likely to be the case for mono-crop estates or small
specialist farms growing such perennial tree crops as rubber, coconut, tea, coffee etc. and for small
or large specialist single-product farms producing such products as milk, broilers, flowers, fish etc.
Subject to appropriate consideration of uncertainty (as considered in the following chapter), such
mono-crop or single-product systems can be planned using the discounted cash-flow procedures and
evaluation criteria outlined in this chapter. Likewise, as indicated in this chapter, these procedures
for investment planning and evaluation can be applied to the planning of farm subsystems or
individual activities to the extent that they do not suffer as much from the curse of dimensionality as
do whole-farm systems.
Given the curse of dimensionality, the best approach to planning over time for non-specialist (i.e.,
mixed) farms, either small or large, is to adopt a rolling planning strategy. Under this approach a
plan is drawn up each year for the next two or three years. Each year the first year of the previous
plan is discarded (since it relates to the year that is passing) and a further year into the future is added
to the plan, i.e., the plan is rolled forward a year. In essence, the plan drawn up in the previous year
is revised in the present year with account taken of whatever relevant new information is available.
The process is depicted in Figure 10.6 for a two-year rolling planning horizon: during Year 1, Plan 1
is drawn up covering Years 2 and 3; during Year 2, Plan 2 is drawn up covering Years 3 and 4; and
so on for Plans 3, 4,... These plans may be of any desired degree of complexity ranging from a basis
in whole-farm budgeting to sophisticated mathematical programming approaches and, as need be, they
may include decisions about the initiation, continuance or replacement of longer-term activities (such
as perennial tree crops) extending beyond the planning period.
FIGURE 10.6 - Schematic Depiction of a Rolling Planning Strategy
10.15 REFERENCES
Alien, R.G.D. (1950). Mathematical Analysis/or Economists, Macmillan, London.
Chisholm, A.H. and J.L. Dillon (1988). Discounting and Other Interest Rate Procedures in Farm
Management, Professional Farm Management Guidebook No. 2, ABRI, University of New England,
Armidale.
Chopra, K and C.H.H. Rao (1992). The Links between Sustainable Growth and Poverty', Quarterly
Journal of International Agriculture 31(4): 364-379.
Wiley, New York.
Gramlich, E.M. (1990). A Guide to Benefit-Cost Analysis, 2nd edn, Prentice-Hall, Englewood
Cliffs.
Markandya, A. (1994). 'Criteria, Instruments and Tools for Sustainable Agricultural Development',
in A. Markandya (ed.), Policies for Sustainable Development: Four Essays, Economic and Social
Development Paper 121, Food and Agriculture Organization of the United Nations, Rome, pp. 1-70.
Mao, J.C.T. (1969). Quantitative Analysis of Financial Decisions, Macmillan, London.
Pearce, D.W. and R.K. Turner (1990). Economics of Natural Resources and the Environment,
Harvester Wheatsheaf, New York.
Robison, L.J. and P.J. Barry (1996). Present Value Models and Investment Analysis, The
Academic Page, Northport.
Sinden, J.A. and D.J. Thampapillai (1995). Introduction to Benefit-Cost Analysis, Longman,
Melbourne.
Thuesen, H.G. (1957). Engineering Economy, 2nd edn, Prentice-Hall, Englewood Cliffs.
von Braun, J. (1992), 'Agricultural Growth, Environmental Degradation, Poverty and Nutrition:
Links and Policy Implications', Quarterly Journal of International Agriculture 31(4): 340-363.
11. PLANNING AND MANAGING FARM
SYSTEMS UNDER UNCERTAINTY
11.1 UNCERTAINTY
11.2 RISK
11.17 REFERENCES
'Life is risky. We cannot remember the future.'
Anderson and Dillon (1992)
Uncertainty and risk go hand in hand with farming. They are a pervasive feature of the farm
environment. How to handle the risks often associated with uncertainty is the most difficult aspect of
farm-system planning and management. It is the subject of this final chapter. Risk and uncertainty
were ignored in preceding chapters. Input prices, output prices, input-output coefficients and yields
were all assumed to be known with certainty. This is clearly a very strong assumption, particularly
relative to long-term planning. For short-term planning and management, however, the assumption of
certainty may often be a reasonable approach. In general, the shorter the planning period, the less the
uncertainty. Likewise, the planning and management of a farm subsystem involves less uncertainty
than does that of a whole-farm system. Too, much of the uncertainty and risk relating to short-term
planning and management of farm systems is of a repetitive routine nature. Over time, farmers have
learnt how to accommodate such routine risk through the use of traditional risk-mitigation strategies
and a cautionary approach. This is particularly so for resource-poor small farmers of Type 1
(subsistence) and Type 2 (semi-subsistence) whose planning and management might generally be
characterized as prudent, circumspect and cautious. Being resource poor and directly dependent on
their own production of food staples, Type 1 and (to a lesser extent) Type 2 farmers and their
families, more so than others, are at risk of malnutrition, starvation and death if they fail to meet the
risks they face from Nature. Caution comes naturally when one's life or family is at stake.
11.1 UNCERTAINTY
Uncertainty is a pervasive feature of life. It may sometimes exist about the past and this may be
relevant to the future. More generally and more importantly, it exists about the future which, unlike the
past, cannot be remembered. By its very nature, the future implies uncertainty as a consequence of the
irreversible nature of time. Uncertainty is also a function of time in that the further into the future the
concern, the greater the uncertainty. There is less uncertainty about tomorrow than there is about a
year hence - and less about a year hence than about ten years hence. Consequently, long-term farm-
system planning is far more hazardous than seasonal or annual planning.
Uncertainty is defined as imperfect knowledge. In the context of farm management, it is relevant to
managerial decision making about the planning and running of the farm system. Decisions which,
because of the influence of forces beyond the farm manager's control, do not have a single sure
outcome are known as uncertain decisions. Such decisions have an array of possible outcomes
which, it is argued, can be specified/portrayed/viewed by the decision maker in the form of a
subjective probability distribution corresponding directly to his or her personal degrees of belief in
the occurrence of the possible outcomes. So, if the farmer believes there are n possible outcomes,
denoted O1, O2,... On to a particular decision, he or she can attach a subjective probability of
occurrence P(Oi) to each outcome for i = 1, 2,... n. To be logical, these probabilities must follow the
usual rules of probability, viz., the set of outcomes O1, O2,... On should be mutually exclusive and
encompass all possible outcomes of the decision so that (1) P(Oi) is in the range from zero to one,
i.e., 0£ P(Oi)£ 1; (2) the probability of the i-th or j-th outcome occurring should equal the sum of their
individual probabilities of occurrence, i.e., P(Oi or Oj ) = P(Oi)+ P(Oj ), and (3) the probability that
one of the outcomes will occur is unity, i.e., P(O1 or O2...or On) = P(O1)+ P(O2)+...+P(On) = S
P(Oi) = 1
These same rules, cast in continuous form, also apply when, unlike the discrete outcome situation
given above, the set of possible outcomes constitutes a continuous set. For example, in considering
crop yield outcomes in relation to nitrogen fertilizer application, the farmer might consider possible
outcomes either in terms of a number of discrete yield intervals or in continuous form as implied by
the continuous response function Y = f(N, R) where Y denotes yield per ha, N is nitrogen fertilizer
applied per ha and R is rainfall, the latter being a variable beyond the farmer's control which causes
yield to be uncertain. The situation is exemplified in Figure 11.1. In the top diagram, response curves
are shown for yield response to nitrogen when rainfall is at its best (upper curve) or at its worst
(lower curve), i.e., response curves depicting Y = f(N/R) with R, respectively, at its best and worst
levels. The vertical distance between the two response curves indicates the range over which the
farmer believes yield may vary for any given level of N depending on the amount of rainfall received
by the crop. Thus for N = 50, Y may range between 1.5 (if R is at its worst or most ineffective level)
and 4.5 (if R is at its best or most effective level). The farmer is assumed (reasonably) to have
degrees of belief about the occurrence of these various possible outcomes. These personal degrees of
belief for Y values when N = 50 are expressed as subjective probabilities in the bottom half of Figure
11.1 - (a) in continuous form on the left and (b) in discrete form (for yield intervals of 0.5t) on the
right. The elicitation of such subjective probabilities corresponding to farmers' individual personal
degrees of belief is outlined in Section 11.10.3 below.
Relative to Figure 11.1, note that there would be a subjective probability distribution specifiable by
the farmer for any relevant level of N. These distributions would, of course, vary in range (and
therefore also in location and shape) for different levels of N. The diagram only illustrates the
farmer's subjective probability distribution for N = 50. If he or she were to choose to apply 50kg/ha
of N, his or her personal degrees of belief in the various possible outcomes would be as depicted in
the (continuous or discrete) probability distribution. If the farmer, therefore, were choosing between
alternative applications of 25, 50, 75 or 100 kg/ha of N, this choice would devolve or be equivalent
to a choice between the subjective probability distributions for outcomes corresponding to these
various levels of N. It is in this fashion that decision making under uncertainty corresponds to a
choice between subjective probability distributions. Ipso facto, the best choice a decision maker can
make under uncertainty corresponds to that choice whose subjective distribution of outcomes most
appeals or is most attractive to him or her.
It must be emphasized that the subjective probability distributions of outcomes referred to above are
personal to the decision maker. While the shape and location of these distributions may be influenced
by others, such as advisers or wizards, or by empirical information from the past or from other
locations, these subjective probabilities belong to the decision maker. Because it is his or her
decision and consequent responsibility, the degrees of belief pertinent to decision making must also
be the decision maker's, not someone else's. It follows that different decision makers may legitimately
have different degrees of belief and therefore different subjective probability distributions for the
same decision problem. Right or wrong are not words that can be used to describe subjective
probability distributions.
FIGURE 11.1 - Uncertainty in Crop-yield Response to Fertilizer and its Depiction (a) by a
Continuous and (b) by a Discrete Subjective Probability Distribution
11.2 RISK
While uncertainty is always present, risk may not be. Risk is only present when the uncertain
outcomes of a decision are regarded by the decision maker as significant or worth worrying about,
i.e., when they affect his or her wellbeing (Fleisher 1990, Ch. 2; Robison and Barry 1987, pp. 13-
14). For a decision problem under uncertainty whose outcomes are regarded as significant, i.e., for a
risky decision problem, the risk that is present is specified by the entirety of the decision maker's set
of subjective probability distributions for the choice-contingent outcomes that may occur. Only the
complete probability distribution (or set of distributions) for the possible (annual) outcomes of a
particular (long-term) choice can fully depict the risk which that particular choice entails for the
decision maker. Other measures of risk such as the range or mean and variance or coefficient of
variation of the distribution of outcomes are (except for special cases such as that of the mean and
variance for the Normal distribution or the range for the rectangular distribution) only partial
measures of risk.
Depiction of risk by the entire probability distribution of outcomes is somewhat different from
everyday usage of the term 'risk' as referring to possible 'bad', 'not so good' or 'negative' outcomes,
i.e., outcomes conventionally located on the left-side tail of the probability distribution. Such
outcomes are better referred to as downside risk. Conversely, upside risk refers to possible outcomes
conventionally located on the right-side tail of the distribution, i.e., to 'good', 'not so bad' or 'positive'
outcomes which the decision maker runs the 'risk' of not obtaining.

The risks faced by farm-system managers may be categorized as coming from two sources: (1) the
external environment as it impinges on the farm system and (2) the internal operational environment of
the farm system.
Major external sources of risk relate to uncertain turbulence in the natural, economic, social, policy
and political environments in which the farm system has to operate (Beal 1996; Fleisher 1990, Ch. 3;
Hardaker, Huirne and Anderson 1997, Ch. 1). Of most general relevance and importance are risks
associated with the natural environment. Because of its time-dependent biological nature,
agricultural production is directly dependent on Nature with all its uncertainties and forwardness
including, on the downside, the possible vicissitudes of short-term weather (droughts, floods, frosts,
storms, cyclones etc.) and long-term climate (such as climatic change including greater variability
due to changes in the Greenhouse Effect), not to mention such natural hazards as earthquakes,
tsunamis, volcanic eruptions, landslides, wildfires and the ever-changing incidence of pests and
disease. All these effects from Nature impinge on yields and, through their effect on market supply,
may impinge on prices, both locally and globally.
Risks associated with the farm system's economic environment relate to uncertainty about (1) market
(demand and supply) conditions and hence to prices for both inputs and outputs, (2) inflation and
interest rates (relevant to long-term planning), and (3) productivity through the availability and merit
of new technology. New technology is an interesting source of risk in that, as it is new, farmers will
lack experience of it and, being cautious, are likely to subjectively assess it as more risky and less
profitable than it possibly is.
The social environment is not, in general, a major source of farm-system risk. Over time, however,
change in education and lifestyles can impact on the availability and competence of farm labour
supply. More important is the possible risk of social upheaval and - in the extreme - war which may
devastate a country's farm systems, destroy the farm household's support network and - at worst - lead
to famine.
Even within a stable socio-political environment, changes in the government-decreed policy
environment may be a significant source of risk to farmers. Policy or institutional areas of particular
relevance are those relating to: commodity prices and marketing; the availability and cost of credit,
water rights and other inputs; the availability of public infrastructure; environmental standards; safety
and health standards; labour laws; land tenure; export and import regulations; exchange rate controls;
and general government reactions to ongoing market globalization with increasing competition and
pressures for deregulation and economic restructuring. There are two dimensions to such policy-risk
possibilities: first, uncertainty about what changes may be legislated and, second, uncertainty about
the extent to which legislated changes will be enforced. Policy risk may impact on the farm
household's income either through effects on yields, on total output level, or on input or output prices.
The final broad external source of farm risk comes from uncertainty about the political environment,
i.e., from any marked change in the political ideology holding sway such as, e.g., moves from a
socialistic centrally planned to a capitalistic free-market system or vice versa. Such changes are
likely to be dramatic in their effects on the type and scale of farming systems in use. They are
particularly pertinent to long-run farm planning.
All the above external sources of risk impinge more or less equally on those farm systems which
share the same external environment. In contrast, sources of risk internal to the farm-household system
impinge, in the main, only on the particular farm. The major internal sources of risk relate: (1) to the
health of the farm household's members; (2) to the inter-personal relations between farm-household
members as influenced by personality, changing values, attitudes and aspirations; and (3) to the
approach followed by the farm manager relative to (a) the conservation or degradation of farm
resources (leading to resource and ecological risk), (b) the use of credit to finance the farm's
operation and development (leading to financial risk), and (c) the inter-generational transfer of the
farm (leading to succession risk). While the first two of these internal sources of risk (i.e., health and
family relationships) relate particularly to the short term, the third set of sources is of more long-term
relevance.
Beyond the above sources of risk that impinge directly on the farm system itself, it should be noted
that the farm management analyst or adviser, regardless of the Field or Mode of operation (Sections
2.1.7 and 8, respectively), faces an additional source of risk. This is that, despite the best of efforts,
the data and/or models used in analysis and diagnosis may be inadequate or incorrect. He or she
should therefore always exercise caution in the interpretation of analysis and the provision of advice.
Too, the analyst should recognize that the risks faced by farmers from the above external and internal
sources are not only pervasive but also additive. This means that, in considering a new farm activity
or technique, the important question may be how it adds to (or subtracts from) the prevalent risk
(Hardaker, Huirne and Anderson 1997, Ch. 11).

Whatever its source, farm-system risk manifests itself as a set of possible outcomes across each of the
alternative choices the farmer may take in managing the farm system or its subsystems. So long as the
outcomes for each alternative are defined so as to be mutually exclusive and exhaustive, their chance
of occurrence can be specified by the farmer in the form of a subjective probability distribution. By
definition, these risky outcomes are of economic significance, i.e., they are significant net gains or
losses measured, as appropriate, either in financial or other terms. They may relate to the whole-farm
system of Order Level 10 or to lower-level subsystems, particularly crop and livestock enterprises
and activities of Order Levels 4 to 7 (Figure 1.2).
The critical economic variables through which uncertainty in the farm's external or internal
environment is translated into farm-system risk are those which determine the net economic outcome
of a decision. For commercial farmers, these variables are product prices, unit yields and aggregate
yields, the prices and quantities used of variable inputs, and fixed costs. For purely subsistence
farmers with no market contact, input or output price risk is irrelevant and risk to their subsistence
livelihood is manifested only through uncertainty in unit and aggregate yields. So, while commercial
farmers may suffer risk from all the sources discussed in Section 11.3 above, risk for subsistence
farmers is largely manifested through yield uncertainty arising from the vagaries of Nature and, if they
are ill-served politically, from the social upheaval of civil unrest and war. Today, however, purely
subsistence farms are rare. Nearly all small farms have some market contact. Small farms of Type 1
(subsistence) and Type 2 (semi-subsistence) are therefore generally mainly affected by Nature-
induced yield risk but, to varying degree, must also face risk arising externally from the economic,
social, policy and political environments, as well as internal risks related to the health and
relationships of farm-household members.
Risk in short-term decisions. Relative to short-run decisions, it is generally product prices and
yields which are uncertain and thus most important in determining risk. This is because, in the short-
run context of seasonal or annual decisions, the quantity of variable inputs used is usually under the
farmer's control and the price of variable inputs and the level of fixed costs are usually known to the
farmer at the time of decision, i.e., at the start of the production process; it is product yield and price
at the termination of the production process at harvest that are uncertain. The fertilizer-choice
situation depicted in Figure 11.1 provides an example.
Risk in long-term decisions. In the context of long-term decisions, all yield, price and cost variables
are likely to be uncertain. As illustrated in Section 10.12, yield profiles over the life of the crop form
the basis of investment planning for long-term crops such as tea, cacao, coffee etc. Such profiles, as
determined by agronomic factors, can be predicted with a high degree of confidence in the absence of
uncertain events. However, in any particular year, actual achieved yield may depart significantly from
expected yield due to the effect of such unpredictable events as cyclones, storms, tidal waves,
droughts, frosts, outbreaks of disease, attacks by monkeys or other pests. Compared with long-term
yield patterns, commodity prices of long-term crops are generally more variable and uncertain.
Empirically, long-term prices can behave in several ways. Some commodities have no apparent trend
over time and are completely unpredictable. However, most commodity prices have either a
downward or an upward secular trend (as well as narrow or broad variations from this trend) - a
downward trend in the case of certain old commodities which are gradually being replaced by
substitutes (e.g., cinchona, sisal), or an upward trend in the case of 'luxury' commodities or those
which can be produced in only a few suitable locations (e.g., cardamom, cloves). Regardless of their
trend, some commodity prices are subject to a regular and predictable cycle superimposed on such
trend (e.g., coffee, cacao, pepper) but the variations from this trend are usually not predictable. Due
to inflation, long-term costs normally tend to increase over time in nominal terms. But superimposed
on this fairly predictable cost stream, abnormal costs can also arise. Without notice, the activity might
be devastated by a cyclone or drought and have to be re-established; or routine normal input costs
might temporarily rise to high levels because of shortage due to, e.g., non-arrival of the supply ship.

The obvious impact of farm-system risk is to make management of the farm more difficult. The
outcome of a management decision under risk cannot be sure. Ex ante, outcomes may be good or bad.
Ex post, with hindsight, there are likely to be regrets about what might have been. All that the farmer
can do, given the imperfect information available, is to make that choice among the available risky
alternatives which most appeals to him or her given his or her preference for outcomes and degrees of
belief in their occurrence. Such a normative approach is based on subjective expected utility theory
as outlined in Section 11.10 below. For the farm manager, it implies the difficult task of specifying
subjective probability distributions of outcome possibilities relative to each choice alternative and
then choosing between these distributions.
A second impact of risk, especially for small farmers of Types 1 and 2, is that it generally makes
them cautious in their decision making. Farm-household survival demands that greater cognizance be
taken of possible adverse outcomes (i.e., downside risks) than of possible good outcomes. While
good outcomes (i.e., upside risks) are inherently attractive, for resource-poor small farmers this
attraction must be expected to be more than outweighed by concern for the possibly disastrous impact
of adverse outcomes. Downside risks challenge farm-system survival, particularly if a series of
adverse outcomes should occur. In the sense of Darwinian evolution with its concepts of natural
selection and adaptation to the environment over time, farm-system survival is enhanced by a prudent,
circumspect and cautious approach to risky decisions on the part of the farmer. It is not surprising,
therefore, that farmers (in common with most other decision makers) generally exhibit risk aversion
(rather than risk neutrality or risk preference) in their decision making. Given the opportunity, as
outlined in Section 11.10.1, they would be prepared to pay some risk premium in order to avoid risk
and achieve some sure or guaranteed outcome.
Finally, in terms of impact, there is no doubt that the agricultural sector and small farms in particular
are generally more susceptible to downside risk than other sectors of the economy. There are two
reasons for this. First, while just as subject as other sectors to risk from the economic, social, policy
and political environments, agriculture is, because of its time-dependent biological basis, far more
exposed to downside risks from a potentially hostile and uncontrollable Nature. Second, farmers (in
particular, resource-poor small farmers) generally do not have easy access to such formal
institutional avenues of risk mitigation as purchased insurance and futures markets. This is because of
agriculture's dispersed nature and largely atomistic small-scale production structure based on Type 1
(subsistence) and Type 2 (pan-commercial) small farms producing mainly food staples for home
consumption and having limited market orientation.

As already noted, farmers are cautious and circumspect in the face of downside risk. They exhibit
risk aversion, not neutrality or preference (Section 11.10.1 below). This is particularly so for
resource-poor small farmers whose livelihood can often be at stake from risk. Their basic risk-
management strategy of caution is manifested in a wide variety of operational strategies aimed at risk
mitigation (Anderson and Dillon 1992, Sections 2.2.1 and 3.3). As summarized in Table 11.1, for
Type 1 (subsistence) and Type 2 (part-commercial) small farmers, these strategies are most evident
in relation to production or yield risk - a not unexpected feature given the farmers' close dependence
on Nature. Fewer strategies are directly oriented to meeting price, marketing and financial risks.
Unlike larger commercial family farms and estates (i.e., farms of Type 5 and Type 6, respectively),
Type 1 and Type 2 farmers do not normally have available to them more formal market-based
institutional approaches to risk management such as a bank line of credit or overdraft facility, crop or
livestock insurance, forward pricing through price, futures or options contracts (Anderson and Dillon
1992, Sections 2.2.2 and 3.4; Calkins and DiPietre 1983, Ch. 9; Fleisher 1990, Ch. 7) or market
guarantees through vertical integration.
TABLE 11.1 - An Overview of Small Farmers' Risk-management Strategies
Functional orientation Strategy Type of risk reduced
Production Use of stable enterprises Yield, technology, policy

Use of tolerant cultivars
Diversification
in crops
in livestock
within seasons
across seasons
across space
Maintain flexibility
over time
in durable assets
Keep reserves seed
seed
fodder
Use risk-reducing inputs

Share-leasing
Assess new technology
Seek information
Marketing Spread sales over time Price
Arrange alternative outlets
Seek barter opportunities
Financial Maintain high equity ratio Financial, yield, price
Maintain credit worthiness
Maintain a cash reserve
Maintain fungible assets
Maintain social network
Off-farm employment
Not all small farmers, of course, use all the risk-management strategies listed in Table 11.1. Their use
varies by agro-ecological region. But whatever combination of risk strategies is used, they form an
integral part of the farming systems developed by small farmers over the centuries in particular
environments. Particularly in relation to production shocks, risk-management strategies used by small
farmers differ according to when they are applied (Matlon 1991). Major emphasis is on ex ante
approaches designed to insure the farm household against downside shocks before they occur. Once a
shock is apparent, interactive approaches involving the reallocation of resources are used to
minimize adverse impact. And after a shock has occurred, ex post approaches of a compensatory
nature are applied to ensure farm household survival. Table 11.2 illustrates, in a level-by-time
framework, the various approaches to risk management typically taken by small farmers in the semi-
arid tropics.
TABLE 11.2 - Example of Classification by Level and Time of Risk-management Strategies
used by Small Farmers in the Semi-arid Tropics
Time frame
Level
Ex ante Interactive Ex post
Plant Varietal selection for stress Replanting with earlier

resistance or tolerance maturing varieties
Animal Use of tolerant breeds and Sell to reduce stocking rate

types
Plot Early/staggered planting dates Change crops with replanting Grazing of failed
plots
Low hill density Change plant density through Late planting for
thinning or replanting forage production
High seeding rate
Intercropping
Run-off management
Delayed fertilizer application
Farm Diversified cropping Shirting crops between land Local non-farm work
types
Land-type diversification
Plot fragmentation
Household, Cereal stocks Off-farm work Cereal rationing

village, region
Livestock as fungible assets Livestock sales
Migration work
Social networks Food transfers by

gift or as aid
Non-farm work
Source: Matlon (1991).
In the drought-prone semi-arid tropics, diversification in crops, in varieties and in their toposequence
location plays a large part in ex ante risk management. Crop diversification involves cropping
different crops across dispersed plot areas as well as mixing crops in the same plots. Such
intercropping improves yield and income stability to the extent that crop mixtures yield better than
monocrops under stress conditions, reduce the incidence and buildup of pests and diseases, and
exhibit compensatory yield behaviour due to differences in crop physiology, phenology or
architecture. Varietal diversification in terms of days to maturity and variability in resistance or
tolerance to biotic stresses reduces the chance of total crop failure due to period-specific stresses.
Likewise, by diversifying plot locations across the soil toposequence, farmers are able to reduce
plot-yield covariation and thus reduce aggregate production variability.
Interactive approaches to semi-arid risk management, as listed in Table 11.2, rely on sequential
decision making as downside shocks unfold. At the start of each cropping season, farmers have
subjective degrees of belief about the onset, amount and duration of rains and the likelihood of pest
and disease occurrence. As the season unfolds, farmers revise these degrees of belief and
sequentially adjust their cropping pattern and cultivation practices as necessary.
Ex post compensatory strategies for risk management used by farmers usually relate more to the farm
household qua household than to farm production activities. Non-farm work, sale of fungible assets
such as cattle, rationing of cereal consumption and gifts of food from other households or relief
agencies are common approaches used by small farmers to maintain income and sustain household
consumption following crop failure.

Being rational, farmers typically make their risky decisions in a reasoned way. This, it may be
argued, implies that, on the basis of their experience, traditional knowledge and whatever other
information is available to them, farmers specify: (1) the alternative choices open to them; (2) the set
of uncertain outcomes associated with each of these alternative choices; and (3) their personal
subjective probability distribution for each of these sets of outcomes (reflecting for each farmer, his
or her degrees of belief in the possible occurrence of these uncertain outcomes). Personal judgement
is then exercised by the farmer to choose that alternative which, for him or her, has the most
preferred/attractive probability distribution of outcomes. Such a process of choice is generally
carried out by the farmer in an implicit or informal rather than an explicit or formal manner. This is
particularly so for Type 1 (subsistence) and Type 2 (part-commercial) small farmers. It is a little less
so for commercially oriented farmers of Types 3, 4 and 5 (independent specialized, dependent
specialized and large family farms, respectively) and of Type 6 (estates). These may sometimes use
more explicit formal procedures either themselves or through the services of professional advisers. In
particular, this is the case relative to long-term risky decisions in estate management.
As the above reference to professional farm management advisers implies, when such specialists
operate professionally in Mode 4 (i.e., in prescriptive mode, Section 2.1.8) in Field A (i.e., in
advising farmers on a participatory basis, Section 2.1.7), they have at their disposal a variety of
formal techniques to assist the farmer's evaluation of risky decisions and guide his or her risky
choices (Dillon and Hardaker 1993, Ch. 8; Hardaker, Huirne and Anderson 1997). Most relevant of
these formal approaches to risky choice are sensitivity analysis, stochastic budgeting, subjective
expected utility analysis (including certainty equivalent analysis and decision trees, and stochastic
dominance analysis), risk-oriented mathematical programming and Monte Carlo simulation. These all
somewhat overlap and interrelate. Except for sensitivity analysis, they can all be cast in an expected
utility framework based on the decision maker's personal preferences among outcomes and his or her
personal degrees of belief in their occurrence. Nonetheless, each is outlined separately in turn below.
The presentation is expository and non-algebraic. More comprehensive discussion of risky decision
making in agriculture is provided, in order of increasing analytical orientation, by Hardaker, Huirne
and Anderson (1997), Anderson, Dillon and Hardaker (1977), Rae (1994, Chs 9, 10 and 11), Dillon
and Anderson (1990, Ch. 7) and Robison and Barry (1987).

Conditional or parametric budgeting (Section 4.7) forms the basis of sensitivity analysis which aims
to assess how sensitive a decision's outcome may be to changes in those variables affecting the
decision (Belli 1996, Ch. 10). Sensitivity is quantified by testing the effect of variations in selected
cost and benefit variables on budgeted outcomes of the decision. These variations may be across the
particular range of values felt to be relevant for each variable or may be some arbitrary percentage
change above and below the value used in the base (non-parametric) budget. This base budget will
generally be based on either the expected (i.e., mean) or most likely (i.e., modal) values of the
variables entering the budget.
Tables 11.3, 4 and 5 provide an example of sensitivity analysis applied to assessing the financial
viability of small-scale sponge farming in Micronesia (Adams, Stevely and Sweat 1995). Table 11.3
shows the base budget for a 0.4 ha farm with a three-year harvest rotation, i.e., 0.4/3 = 0.133 ha are
harvested each year. Since in this particular case the base budget is meant to represent the likely
annual performance of the enterprise once established, the budget entries are most likely or modal
values as specified or agreed by the farmer. The budget indicates that with a survival rate of 90 per
cent from planting to harvest and a marketability of 80 per cent, the 0.4 ha farm would give a pre-tax
net annual return to the owner for his or her management, capital and risk of $784. The budget also
implies that, other things unchanged, the minimal survival rate from planting to harvest that would
allow total production costs, excluding opportunity costs, to be recovered is 42.8 per cent. This is the
break-even survival rate (in budget terms) or, to use the jargon of sensitivity analysis, the switching
value for this variable if the initial assumption is of a 90 per cent survival rate.
Table 11.4 shows the sensitivity of budgeted net return, production cost per sponge sold and break-
even survival rate to changes in four key variables. These four change-variables were chosen on a
priori grounds for their expected relevance to the budget calculations. Likewise the range of values
over which they are varied reflects subjective judgement of their possible range of variation. The
budgeted results of the sensitivity analysis show, e.g., that annual net return to the owner's
management, capital and risk may vary from -$416 to $4 984 in terms of possible market-price
variation, from $224 to $960 relative to survival rate, from $484 to $1 084 relative to marketability
percentage and from $676 to $1 166 relative to total variable cost. The results also indicate that the
switching value of survival rate between financial viability and non-viability is more sensitive to
market price than to the other variables.
The sensitivity analysis of Table 11.4 is based on one-at-a-time change in the variables. In the real
world, however, all the variables would be subject to simultaneous change from their expected
values. This more realistic situation is appraised in Table 11.5 in terms of the budgeted outcomes if
each of the variables were to simultaneously occur at their most favourable, their most likely and
their most unfavourable levels. This indicates that if the variables were to occur at their best possible
level, the owner's return to his or her management, capital and risk would be $6 413; at the other
extreme, if the variables took their worst possible level, the return would be a loss of $924.
TABLE 11.3 - Annual Enterprise Budget for a 0.4 Ha Sponge Farm with 0.133 Ha harvested
Annually
Revenuea $2400
Costs of production
Variable costs:
Oil and gas 200
Boat and motor maintenance 182
Marketing costs 60
Supplies 98
Total variable costs 540
Fixed costs:
Interestb 0
Depreciation 116
Total fixed costs 116
Total production costs 656
Net return before tax to owner for management, capital, labour and risk 1 744
Opportunity cost of owner's labour 960
Net return before tax to owner for management, capital and risk 784
Cost per sponge sold 0.273
Break-even survival rate (%) 42.8

a Based on a survival rate of 90 per cent and a marketability of harvested sponges of 80 per cent, yield of 2 400 units and a market
price of $ 1.00 per unit.
b Farmer equity of 100 per cent is assumed in the necessary capital investment of $650.
Source: Adams, Stevely and Sweat (1995).
It should be noted that only one alternative - investment in a 0.4 ha sponge farm - is explicitly
considered in the above example. The implied decision is between this investment and doing nothing,
i.e., maintaining the status quo of having, say, the money invested at a known rate of interest.
The above example well illustrates the pros and cons of sensitivity analysis. On the positive side, (1)
in a budget context, it is readily carried out using spreadsheet analysis on a personal computer; (2) it
gives an indication of the range of possible outcomes; (3) it indicates the relative sensitivity of
outcomes to the change-variables assessed; (4) it enables the determination of switching values for
key change-variables; and (5) it may pinpoint the need for additional information on some variables.
In general, however, these positive aspects are outweighed by the three major limitations of
sensitivity analysis: (1) it does not take into account the probability of occurrence of changes in the
variables; (2) nor does it take into account correlation among possible changes in the variables; and
(3) if, as often commonly used, the values of the change-variables are varied by standard (arbitrary)
percentages, e.g. ±10 or ±20 per cent, these changes may bear no relation to the likely variability of
the change-variables. The lack of account for the probability and correlation of change in the
variables is particularly damaging. Thus, e.g., Table 11.4 gives net return outcomes depending on
market price but no indication of their probability of occurrence which, of course, parallels the
(unspecified) probability distribution for market price. Likewise, the outcomes presented in Table
11.5 for simultaneous variation in the four change-variables should also be conditioned by their joint
probability of occurrence which parallels the joint probability of the four change-variables. The usual
practice of varying only one variable at a time, e.g., as in Table 11.4, is justified only if the change-
variables are uncorrelated; otherwise the change-variables should be considered jointly as in Table
11.5. In the context of the above example, survival rate and marketability percentage are probably
negatively correlated with market price. As a result, the sensitivities shown in Table 11.4 may be
exaggerated since they do not allow for counterbalancing variation in these three variables.
TABLE 11.4 - Summary of Sensitivity of Annual Net Returna, Cost per Sponge, and Break-even
Survival Rate to Changes in Four Key Variablesb
Variable Net Production cost per sponge Break-even survival

return sold rate
($) (cents) (%)
Market price per unit ($)
0.50 416 27.3 60.7
0.75 184 27.3 48.8
1.00 784 27.3 42.8

1.25 1 384 27.3 39.3
1.50 1 984 27.3 36.9
1.75 2 584 27.3 35.2
2.00 3 182 27.3 33.9
2.25 3 784 27.3 32.9
2.50 4 384 27.3 32.1
2.75 4 984 27.3 31.5
Survival rate (%)
75 224 35.7 42.8
80 408 32.7 42.8
85 584 29.7 42.8
90 784 27.3 42.8
95 960 25.5 42.8
Marketability (%)
70 484 31.2 45.4
75 634 29.2 44.0
80 784 27.3 42.8
85 934 25.7 41.8
90 1 084 24.7 40.8
Total variable cost (%

change)
-20 1 166 21.4 32.4
-10 838 25.0 41.4
0 784 27.3 42.8
+10 730 30.0 44.3
+20 676 32.0 45.8

a Net return to the owner's management, capital and risk.
b Values in bold represent results for the baseline set of assumptions.

TABLE 11.5 - Effect on Annual Net Return of Simultaneous Variation in the Four Key
Variables of Table 11.4
Variable Net returnb
Level of variable a Survival rate Marketability Unit price Total variable cost
(%) (%) ($) ($) ($)
At worst 75 70 0.50 648 -924
Most likely 90 80 1.00 540 784
At best 95 90 2.75 432 6413
a The worst and best extreme values from Table 11.4.

b Net return to the owner's management, capital and risk.

As its name implies, stochastic or risk budgeting is carried out by attaching probabilities of
occurrence to the possible values of the key variables in a budget, thereby generating the probability
distribution of possible budget outcomes (Dillon and Hardaker 1993, pp. 169-172; Hardaker, Huirne
and Anderson 1997, pp. 120-126). The probabilities used should be those of the decision maker, i.e.,
his or her personal or subjective probabilities based on experience, intuition and/or any other
available information. As an example, consider the budgets relative to survival rate presented in
Table 11.4. In this situation, the only variable assumed to be risky is survival rate. Accordingly, if the
farmer's personal probabilities for survival rate, denoted by P(si), are as shown in Table 11.6, these
also constitute the probability distribution for budgeted annual net return to the farmer's management,
capital and risk. This distribution has a modal (i.e., most likely) value of $784 and, as calculated in
Table 11.6, an expected or mean value of $639.
TABLE 11.6 - Stochastic Budget relative to Survival Rate for the Sponge-farm Investment of
Table 11.4
Survival rate Probability of survival Net Cumulative probability of net

(s) rate returna return
Central Interval (P(si)) (R) (P(R £ R*))

value
(%) (%) ($)
75 72.5 < s £ 0.10 224 0.10

77.5
80 77.5 < s £ 0.15 408 0.25

82.5
85 82.5 < s £ 0.25 584 0.50

87.5
90 87.5 < s £ 0.40 784 0.90

92.5
95 92.5 < s £ 0.10 960 1.00

97.5
Expected net return = S P(Ri)Ri = (0.10)224 + (0.15)408 + (0.25)584 + (0.40)784 + (0.10)960 =

$639.20
a Net return to the owner's management, capital and risk.
As shown respectively in Table 11.6 and Figure 11.2, the distribution of net return can also be listed
or graphed in the form of a cumulative probability distribution or cumulative distribution function
(CDF). Denoting net return by R, this cumulative distribution shows the probability that R will be less
than or equal to any nominated value R*, i.e., P(R<R*) and, ipso facto, the probability that R will
exceed any nominated value R* - which probability is necessarily equal to one minus the probability
that R will be less than or equal to R*, i.e., P(R>R*) = 1 - P(R £ R*). Thus, reading from the
cumulative distribution sketch of Figure 11.2, the probability that annual net return will be no more
than $500 (i.e., R* = 500) is P (R £ R* = 500) = 0.364. Conversely, the probability that annual net
return will exceed $500 is 1 - 0.364 = 0.636.
Just as with sensitivity analysis, stochastic budgeting with discrete probabilities is readily conducted
via spreadsheet analysis on a personal computer. More generally, specialist software add-ins such as
@Risk (Palisade Corporation 1997; Winston 1996) for use with spreadsheet software such as Excel
or Lotus 1-2-3 can be used. With @Risk, a wide variety of standard probability distributions such as
the Normal, Beta, Gamma, triangular or rectangular may be used, as well as specific distributions
specified by the user. An example of the use of @Risk for stochastic budgeting is provided by
Hardaker, Huirne and Anderson (1997, pp. 121-126).
FIGURE 11.2 - Cumulative Probability Distribution of Annual Net Return (R) for the Stochastic
Budget of Table 11.6
While the simple example of stochastic budgeting presented in Table 11.6 above is based on
enterprise budgeting, stochastic budgeting can be applied to any type of budget whether it be of a
whole-farm or partial, cash-flow or profit, single or multi-period nature.
Finally, it should be noted that stochastic budgeting is in fact a simple form of Monte Carlo simulation
analysis (Section 11.15), the only difference being that the latter often involves more elaborate
modelling of the decision alternative being simulated.

The term utility is used here in the sense of satisfaction. Subjective expected utility theory is a logical
approach to risky decision making based on a few eminently reasonable axioms such as, e.g., if a is
preferred to b and b is preferred to c, then a will be preferred to c (Anderson, Dillon and Hardaker
1977, Ch. 4). These axioms imply: first, that a personal utility function U (such as the curve ABC in
Figure 11.3) can be ascribed to a decision maker such that an amount of utility U(Oi) can be
associated for him or her with the outcome Oi, second, that if Oi is preferred to Oi, then U(Oi)>U(Oj );
third, that if Oi is uncertain (i.e., not sure), then the decision maker can specify a subjective
probability for its occurrence corresponding to his or her degree of belief in its occurrence so long as
the set of outcomes {Oi} associated with a particular choice is mutually exclusive and exhaustive;
and, fourth, that the utility of the particular choice having the set of risky outcomes {Oi} is equal to the
expected utility of the set of outcomes, i.e.,
U({Oi}) = EU({Oi })
where E denotes statistical expectation so that, e.g., for the discrete case where Oi (i = 1,2,...n) has
probability P(Oi):
U({Oi}) = EU({Oi}) = U(O1)P(O1) + U(O2)P(O2) +...+ U(Oi)P(Oi) +...+ U(On)P(On) =
Subjective expected utility theory is thus remarkable in that it brings together and integrates the three
crucial elements of risky decision making: first, the decision maker's personal preferences about
possible outcomes; second, the decision maker's personal degrees of belief in the occurrence of
possible outcomes; and, third, through the use of his or her own personal preferences and
probabilities, the decision maker's personal responsibility and accountability for whatever decision
is taken.
Figure 11.3 illustrates the concept of a utility function and expected utility for the typical case of a
risk-averse decision maker. The utility function U is depicted by the curve ABC which, being
concave, has positive but diminishing slope to the right, indicating (1) that more is preferred to less
but, because of satiation, with diminishing marginal utility and (2) that the decision maker is risk
averse. Reading from the curve, it can be seen, e.g., that an outcome yielding $3 000 has a utility
value of 2.3.
While raw outcomes are here measured in money terms, any relevant measure such as, e.g., bags of
rice, may be used as appropriate. Too, it is important to note that the utility scale is always arbitrary,
analogous to a temperature scale so that any function U* will serve as well as the function U so long
as U* = aU + b with a > 0. Thus it makes no sense to say that one outcome, e.g., would give twice as
much utility as another. All that can be said is that one outcome exceeds another in utility, i.e., is
preferred to another. Nor, because of the personal nature of utility, can comparisons be made between
one person's utility values and those of another person. Also, the utility assessment of alternative farm
decisions must usually be on an aggregate whole-farm, project or enterprise basis, not on a technical
unit basis such as per ha or per tree or per head. Only if each alternative is of the same size in terms
of technical units is it appropriate to make utility comparisons on a technical unit basis. Likewise, for
utility analysis, financial outcomes should be measured as net returns or profit, not as gross margins
since maximizing U(TGR - TVC - TFC) is not the same as maximizing U(TGR - TVC) = U(TGM).
Note also that if an alternative involves n technical units, each giving a payoff of $X for a total payoff
of $nX, diminishing marginal utility implies that U(nX) < nU(X).
FIGURE 11.3 - Illustration of the Concept of a Utility Function and Expected Utility for a Risk-
averse Decision Maker
Suppose the decision maker whose utility function is depicted in Figure 11.3 has to choose between
two alternatives, one being a sure prospect offering a sure gain of $2 600 and the other a risky
prospect offering a 50:50 chance of either $1 000 or $5 000. From the graph of Figure 11.3, the sure
gain of $2 600 has a utility value of 2.12. The utility of the risky prospect ($1 000, $5 000; 0.5, 0.5)
is given by its expected utility. This is calculated as U ($1 000)(0.5) + U ($5 000)(0.5) which, from
Figure 11.3, is equal to (1.2)(0.5) + (2.8)(0.5) = 2. Since 2.12 > 2, the alternative offering a sure gain
of $2 600 would be preferred to the risky prospect. This is so even though the expected value of the
risky prospect's raw outcomes, i.e., its expected money value (EMV) of ($1 000)(0.5) + ($5 000)
(0.5) = $3 000, is greater than the sure prospect's money value of $2 600.
Risk attitude
Figure 11.3 also illustrates the concepts of risk aversion and certainty equivalence (Hardaker, Huirne
and Anderson 1997, Ch. 5). The decision maker's risk aversion is shown by the fact that, although the
risky prospect has an EMV of $3 000, because of its risky nature its utility value of two is less than
that of a sure $3 000 which, reading from the utility curve of Figure 11.4, has a utility value of 2.3. In
fact, corresponding to point B on the utility curve, the risky prospect's utility value of two is the same
as that of an outcome of $2 400. In other words, for this particular decision maker, the risky prospect
($ 1 000, $5 000; 0.5, 0.5) is equivalent in utility terms to a sure prospect of $2 400. This amount of
$2 400 is thus referred to as the certainty equivalent of the risky prospect. In general, subjective
expected utility theory implies that, for any risky choice faced by a decision maker, it will always be
possible to specify some sure prospect which has the same utility value as the risky prospect. In other
words, the decision maker would be indifferent between taking the risky choice and receiving the
sure prospect. This sure prospect is known as the certainty equivalent (CE) of the risky choice.
Moreover, since different decision makers have different utility functions, they will also legitimately
have different CEs for the same risky prospect.
The difference between the mean or expected value (EV) of a risky prospect and its CE (i.e., EMV -
CE in the case of a money prospect) is the decision maker's risk premium for the risky prospect. For
the example of Figure 11.3, the risk premium is given by the distance BD implying a risk premium of
$3 000 - $2 400 = $600: this decision maker would be prepared to give up 600 'expected $' in order
to avoid the risky prospect. Another decision maker could legitimately have a different risk premium
for the same risky prospect.
If the decision maker is risk neutral, his or her utility function is linear, thus implying a risk premium
of zero as would be the case, e.g. if the line ADC of Figure 11.3 were the decision maker's utility
function. Conversely to the case of risk aversion, risk preference implies (1) a convex utility curve,
i.e., one having positive but increasing slope to the right corresponding to increasing marginal utility,
and (2) a negative risk premium since with risk preference the decision maker would be prepared to
pay a premium to face the risk opportunity.
The above relationships between the shape of a decision maker's utility function, risk attitude, risk
premium and marginal utility are summarized in Table 11.7.
TABLE 11.7 - Relationships between the Shape of the Utility Function, Risk Attitude, Risk
Premium and Marginal Utility
Attribute
Shape of the utility function U(X) Risk attitude Risk premium Marginal utility
(EV - CE) (dU/dX>0)
Concave aversion positive diminishing

(EV > CE) (d2U/dX2 < 0)
Linear neutrality zero constant

(EV = CE) (d2U/dX2 = 0)
Convex preference negative increasing

(EV < CE) (d2U/dX2 > 0)
As noted in Section 11.5, most decision makers, particularly including small farmers, are risk averse.
Perhaps the only situation in which small farmers may behave in a risk preferring way is when their
household faces a food crisis. In such circumstances they may (be forced to) take decisions (such as
selling assets, borrowing funds, depleting resources) which place the household's future livelihood
from the farm in jeopardy. Generally, however, risk preference is sufficiently rare that it need not
concern the farm management analyst. Too, for many risky farm decisions where the sets of possible
outcomes are not vastly different in size or probability, it may often suffice to assume risk neutrality
and assess alternative choices in terms of their expected or mean outcomes.
A number of methods is available for the elicitation of a decision maker's utility function. These are
all based on his or her responses to a chain of questions involving choice between hypothetical risky
prospects. Details of the methods (along with much fuller discussion of the concept of a utility
function) are given by Anderson, Dillon and Hardaker (1977, Ch. 4) and Hardaker, Huirne and
Anderson (1997, Chs 3, 4 and 5).
In general, being based on mind experiments rather than real-world risky prospects, elicitation is not
always reliable. However, this is not a problem relative to small farmers. Except perhaps for case
studies involving representative small farms, it would not be justifiable to attempt to elicit a small
farmer's utility function. Decision guidance to such farmers on an individual basis will generally be
more efficacious if based on the certainty equivalent approach (Section 11.11). Only for larger
commercial farms and estates, and then only for significant risky-choice problems, is it likely that
direct elicitation of the manager's utility function to provide a basis for risky choice would be
warranted. More generally, risk appraisal pertinent to farmers or farming systems in different agro-
ecological zones can be based either (i) on an arbitrary assumption of a general risk-averse utility
function such as 'Everyman's Function' U = logeW where W is the risky outcome, or (ii) on methods
such as stochastic dominance analysis (Section 11.13) and risk-oriented mathematical programming
(Section 11.14), the software for which can enable sensitivity analysis to be carried out on a personal
computer relative to varying degrees of risk aversion and forms of the utility function. A particular
case for which such approaches are relevant is in the evaluation of trial data for new technology, e.g.,
for the evaluation of new crop varieties before their release to farmers.
Just as for the elicitation of small farmers' utility functions, the elicitation of small farmers' subjective
probability distributions for relevant risky events is generally not a practical possibility. There are
too many small farmers, each facing too many risky decisions - for each of which (unlike for his or
her utility function which remains the same for different decisions) each farmer will have different
sets of degrees of belief. Only for case studies of representative situations might small farmers'
probabilities be elicited. Conversely, probability elicitation could be a normal element of specialist
services used by the managers of larger commercial farms and estates in their appraisal of important
risky choices.
A farmer's degrees of belief in uncertain events are usually best elicited in the form of discrete
probability distributions using a visual display procedure. As illustrated by Figure 11.4, the
procedure is based on the decision maker allocating counters across the various possible event-
situations in accord with his or her degrees of belief in their occurrence. The elicited relative
frequency of the counters across the various events then corresponds to the decision maker's
subjective probability distribution for the events considered.
Based on the sponge-farm example of Section 11.8, Figure 11.4 shows the farmer's elicited frequency
distribution for the joint-event possibilities of survival rate percentage (si) and marketability
percentage (mj ). Dividing each of these frequencies by the total number of counters used (i.e., by 100
in this case) gives the farmer's subjective probability for each joint event (si and mj ). Thus the joint
event of a 90 per cent survival rate and marketability of 80 per cent has a joint probability of 15/100
= 0.15. Likewise, the row and column frequency totals give the farmer's marginal probability
distribution, respectively, for survival rate and marketability, i.e., for each of these considered
without regard to the other. Thus the probability of marketability being 70 per cent is 12/100 = 0.12.
As a practical matter, in allocating counters to complete a display such as depicted by Figure 11.4, it
will usually be easier if the marginal frequencies are allocated first. Of course, if only a single
variable is involved, only a one-way rather than a two-way display needs to be completed. If more
than two variables need to be considered jointly, a multi-stage display procedure needs to be used.
As an example, Figure 11.5 shows the second and final-stage display for the elicitation of the farmer's
joint probabilities P(si, mj and pk ) for survival rate, marketability and market price (pk ) in the
sponge-farm example. The first-stage display is that of Figure 11.4. In the second-stage display of
Figure 11.5, the 5 x 5 = 25 joint events of Figure 11.4 are listed out to give all their possibilities of
joint occurrence with five levels of market price, i.e., a total of 25 x 5 = 125 triple-event possibilities
(si, mj and pk ) involving the allocation of, for convenience, 500 counters. All such elicitations would
proceed on an iterative basis of changing allocations until the decision maker was content that the
display reflected his or her degrees of belief. From Figure 11.5, the elicited probability for the triple
event of, e.g., a survival rate of 85 per cent, marketability rate of 80 per cent and market price of
$1.50 is 8/500 = 0.016.
FIGURE 11.4 - Application of Visual-display Procedure for the Elicitation of the Farmer's
Subjective Probabilities for the Joint Occurrence of Survival Rate (si) and Marketability (mj)
Levelsa in the Sponge-farm Example of Table 11.4
Survival rate Marketability (mj)

(si) Total
(%)
(%) 70 75 80 85 90
75 · ···· ···· · 10
80 ·· ···· ····· ·· · 15
·
85 · · · · · · · · · · · · · · · · · · · · · · 25
···
90 · · · · · · · · · · · · · · · · · · · · · · 40
·········· ···
·····
95 · ··· ···· ·· 10
Total 12 26 37 18 7 100
a The specified levels are the mid-points of exhaustive exclusive intervals centred on the specified levels. Thus a level of, e.g., 80
refers to the range from >77.5 to £ 82.5.
Two features of the above visual-display elicitation procedure need to be noted. First, since
probabilities are to be derived, the set of events considered must be exhaustive (i.e., cover all
possibilities) and exclusive (i.e., not overlap). Second, particularly if joint events are being
considered, the number of intervals considered for each variable should be no more than three or
four, otherwise the display becomes too difficult to comprehend adequately.
FIGURE 11.5 - Second-stage Visual Display for the Elicitation of the Farmer's Subjective
Probabilities for the Joint Occurrence of Survival Rate (si), Marketability (mj) and Market
Price (pk) Levels in the Sponge-farm Example of Table 11.4
si mj pk Total
(%) (%)
0.50 1.00 1.50 2.00 2.50
75 70 · ·· · · 5
80 70 · ··· ··· ·· · 10
85 70 ··· ·········· ·· ·· 20
···
90 70 ··· ·········· ·· ·· 20
·· ·
95 70 ·· · · · 5
75 75 ···· ····· ···· ··· · 20
···
80 75 ···· ·········· ·· · 20
···
85 75 ··············· ··· · 25
····· ·
90 75 ··············· ····· ··· 50
···············
·· ·····
·····
95 75 ···· ····· ··· ·· 15
·
75 80 ··· ····· ···· ··· ·· 20
···
80 80 ··············· ··· · 30
···· ·····
··
85 80 ··············· ···· · 40
·········· ···
· ·····
·
90 80 · · · · · · · · · · · · · · · · · · · · · · · · · 75
··············· ····· ·
··············· ··
··· ·····
····
95 80 ·········· ···· ·· · 20
···
75 85 ·· ·· · 5
80 85 ··· ···· · · · 10
85 85 ···· ················ · 25
·· ·· ·
90 85 ··············· ····· · 40
·········· ·····
··· ·
95 85 · ····· ··· · 10
75 90 0
80 90 · ·· · · 5
85 90 ··· ····· ···· · ·· 15
90 90 ···· ····· ··· · · 15
·
95 90 0
Total out of 500 105 185 112 68 30 500

Of course, farmers' degrees of belief in uncertain events are not formed in a vacuum. As well as
involving intuition and feeling, they are based on whatever information may be available from
experience or from others. The incorporation of such information into probability judgements and
possible psychological pitfalls in probability assessment are outlined by Anderson, Dillon and
Hardaker (1977, Ch. 2) and Hardaker, Huirne and Anderson (1997, Chs 3 and 4). Fuller treatment of
probability elicitation is provided by Morgan and Henrion (1992, Chs 6 and 7).
Judgemental fractile method
As noted in relation to stochastic budgeting (Section 11.9), specialist add-in software such as @Risk
(Winston 1996) is available for use with spreadsheet software to fit a wide range of probability
distributions to risky decision variables. This is particularly useful for the fitting of continuous
distributions and gives the analyst the opportunity of visually checking the goodness of fit of
alternative distributions. The data to which such distributions are fitted may come from visual display
procedures as outlined above or be elicited by the judgemental fractile method (Anderson, Dillon
and Hardaker 1977, pp. 23-25) or be generated by simulation modelling (Section 11.15).
The judgemental fractile method leads directly to the decision maker's cumulative distribution
function (CDF) for the risky variable of concern. The farmer would first be asked for the f0.0 and f1.0
fractiles, i.e., the upper and lower range of the variable. Next, the f0.5 fractile value (or median) such
that it is equally likely the uncertain variable will be above or below this value is ascertained. Next,
elicit f0.25 such that it is equally likely the uncertain variable is less than f0.25 and between f0.25 and
f0.5 Similarly, find f0.75; then - continuing to halve the fractile intervals - elicit f0.125, f0.375, f0.625 and
f0.875. A CDF can then be fitted by hand or by personal computer to the set of elicited fractile values.
Thus, if the judgemental fractile method had been used, the CDF of Figure 11.2 could have been
sketched from elicited fractile values of, say, f0.0 = $0, f0.125 = $260, f0.25 = $410, f0.5 = $585, f0.75 =
$700, f0.875 = $765 and f1.0 = $960.
Rectangular distribution
The simplest, albeit the roughest, approach to probability elicitation that an analyst can use is to
assume that the decision maker's degrees of belief can be specified as either a rectangular distribution
or as a triangular distribution. For a discrete variable having n possible values, the rectangular
distribution gives each value a probability of 1/n. For a continuous variable with range (a, b), the
distribution is rectangular in shape with a uniform probability of 1/(b - a). The rectangular
distribution is thus appropriate if all possible values of the risky variable are assumed to be equally
likely. This, it may be argued, is the case when the decision maker feels totally uncertain about the
likelihood of possible outcomes; equal ignorance about each may then be translated into an
assumption of equal probability of occurrence. However, this begs the question of how the range of
the distribution can be set.
Triangular distribution
The triangular distribution is a continuous distribution whose specification requires elicitation of
only three values of the risky variable - its lowest, highest and most likely values denoted,
respectively, by a, b and m. The formula for the triangular distribution for an uncertain variable X is:
f(X) = 2(X - a)/(b - a)(m - a), X £ m
f(X) = 2(b - X)/(b - a)(b - m), X > m
and the formula for its CDF is:

F(X) = (X - a) 2/(b - a)(m - a), X £ m
F(X) = 1 - [(b - X) 2/(b - a)(b - m)], X > m.
The mean E(X) and variance V(X) of the triangular distribution can be found as:
E(X) = (a + m + b)/3
V(X) = [(b - a) 2 + (m - a)(m - b)]/18
Figure 11.6 gives the triangular distribution and its CDF for the net return data of Table 11.5. While
this is a single-variable distribution for possible decision outcomes, it is based on simultaneous
variation in four variables influencing outcomes. For convenience, the distributions of Figure 11.6 are
in units of a thousand dollars. To fit the probability distribution, the probability of m = 0.784 is
calculated as 0.273 using the above equation for f(X) with a = -0.924 and b = 6.413. To fit the
cumulative distribution, F(X) is calculated for a number of X values, e.g., for X = 0, F(X) = 0.068; for
X = 3, F(X) = 0.718.
Suppose a farmer's utility function for money gains and losses is approximately represented by U(X)
= 2.05X - 0.01X2 for X£ 80, where X is in units of $10. The farmer has to decide whether to spend
more on fertilizer for his or her two ha of crop than last season's $40/ha. The profit from fertilizer use
depends on the type of season that occurs. The farmer's subjective probabilities for the type of season
are given in the payoff matrix of Table 11.8 along with the possible dollar profits for each alternative
action and state (i.e., type of season). Since the farmer's utility function is non-linear and exhibits
diminishing marginal utility (i.e., d2U/dX2<0), the monetary payoffs of Table 11.8 are specified on a
total crop basis of two ha. Next, using the farmer's utility function (and remembering that X is in units
of $10), the utility of each of these total crop payoffs is calculated to give the utility payoff matrix of
Table 11.9. The expected utility of each alternative decision is then calculated using the farmer's
subjective probabilities. The optimal choice is shown to be to spend $80/ha - this action has the
highest expected utility of 8.512. In contrast, in expected monetary terms, the action of spending
$120/ha has the highest EMV of $76 (and would be the optimal choice if the farmer were risk neutral
and thus had a linear utility function). Substituting the utility value of the optimal risky act into the
utility function and solving for X gives the CE of the optimal risky act as $42, i.e., the farmer would
be indifferent between a sure prospect of $42 profit and the risky prospect of outcomes from spending
$80/ha on fertilizer for his or her two ha of crop. This CE of $42 compares with the EMV of $64
from spending $80/ha on the two ha of crop. The farmer's risk premium in this instance is thus $(64 -
42) = $22. He or she would be prepared to forgo 22 'expected dollars of profit' to avoid undertaking
the risky prospect.
FIGURE 11.6 - Triangular Probability Distribution f(X) and Cumulative Distribution Function
F(X) for the Net Return Data of Table 11.5
TABLE 11.8 - Monetary Payoff Matrix for a Farmer's Fertilizer Decision Problem together
with EMV and CE of Each Risky Prospect
Alternative actions
Type of season Farmer's subjective probability Spend Spend Spend Spend

$40/ha $80/ha $120/ha $160/ha
($ profit or loss)
Poor 0.1 -160 -240 -320 -400
Fair 0.2 -40 -160 -240 -320
Good 0.5 40 80 120 160
Excellent 0.2 240 400 480 480
EMV 44 64 76a 72
CE 37 42b 40 26
a Indicating optimal choice if farmer were risk neutral.
b Indicating optimal choice for farmer with utility function U(X) = 2.05X - 0.01X2, X £ 80, for X in
units of $10 gain or loss.
TABLE 11.9 - Utility Payoff Matrix for the Farmer's Fertilizer Decision Problem of Table 11.8
Alternative actions
Spend Spend Spend Spend

Type of season Farmer's subjective probability
$40/ha $80/ha $120/ha $160/ha
(Utility payoff)
Poor 0.1 -3.536 -5.496 -7.584 -9.800
Fair 0.2 -1.672 -7.072 -10.992 -15.168
Good 0.5 4.020 7.880 11.580 15.120
Excellent 0.2 8.688 13.200 15.072 15.072
EV 7.500 8.512 8.076 5.224

Comparison of the certainty equivalents of alternative risky choices is probably the most practical
guide available for small farmers in their risky decision making (Makeham, Halter and Dillon 1988).
For many, it perhaps approaches the implicit intuitive way in which they make their risky choices. In
essence, the CE approach involves:
(1) specifying the set of possible alternative decisions/choices/acts;
(2) specifying the set of possible outcomes associated with each possible decision;
(3) specifying the subjective probability distribution of each decision's set of possible outcomes;
(4) specifying the CE of each decision's probability distribution of possible outcomes; and
(5) choosing that alternative whose probability distribution has the largest CE.
Such specification and comparison of each alternative's CE may be carried out informally by
introspection or more formally using a payoff matrix as in Table 11.8 or a decision-tree format as
outlined in Section 11.12 below. Either way, the CE approach has much to recommend it: first, it is
reasonable to expect that for any risky prospect he or she may face, the decision maker can specify
some equivalent sure prospect or CE; second, the CE approach is logical in that it recognizes and
integrates the decision maker's personal probabilities and personal preferences; and, third, it has the
attraction of being based on what might be called guided intuition, i.e., intuition guided by a
recognition of the importance of personal probability and personal preference in optimal risky
choice. Note that, just as with utility appraisal, the CE approach generally needs to be applied on an
aggregate rather than a technical unit basis.
The CE approach can be illustrated by reference to the fertilizer decision problem of Table 11.8.
Having either explicitly or implicitly specified the payoffs and probabilities associated with each of
the four risky prospects, the farmer would establish introspectively his or her CE for each risky
prospect and choose that action with the largest CE. Note that, as implied by the definition of a CE,
the farmer's ranking of the risky prospects based on certainty equivalence as shown in Table 11.8
agrees with the utility-based ranking of Table 11.9.

The four essential features of any risky decision problem - i.e., available acts, uncertain events,
uncertain payoffs and their subjective probabilities of occurrence - can be
specified/displayed/modelled in the form of a decision tree or decision flow diagram. A decision
tree displays the choices available, the uncertain events affecting each alternative, the possible
outcomes or payoffs from each alternative and their probability of occurrence. Sequences of possible
decisions and uncertain events are sketched in time-sequence from left to right as successive branches
respectively emanating from decision nodes and event nodes on a tree graph. Working backwards
from terminal outcomes (i.e., from right to left), certainty equivalents are compared across
alternatives to select a sequence of decisions which maximizes the decision maker's expected
satisfaction (Anderson, Dillon and Hardaker 1977, pp. 124-130; Hardaker, Huirne and Anderson
1997, pp. 107-119). As an example, Figure 11.7 shows the risky fertilizer decision problem of Table
11.8 displayed as a decision tree, first in its entirety (tree A) and then in an equivalent folded-back
form (tree B) based on the farmer's CE for each risky choice. Again the optimal choice is indicated as
spending $80 per ha on fertilizer.
As the above example indicates, a decision tree is a means of operationalizing the certainty
equivalent approach (Dillon and Hardaker 1993, pp. 245-252). The decision tree (1) forces the
decision maker to explicitly consider alternative actions and possible events influencing their
outcomes, (2) enables the decision maker to express a complex problem situation as a sequence of
decisions, (3) helps the decision maker to quantify the decision process, and (4) facilitates optimal
choice based on the comparison of certainty equivalents. Again, as with utility analysis, outcomes
should be specified on an aggregate basis and the risky events considered should be exhaustive and
mutually exclusive so that their probabilities of occurrence sum to one.
FIGURE 11.7 - Risky Decision Problem of Table 11.8 expressed as a Decision Tree
Figure 11.8 shows the decision tree for a more realistic example in which a farmer has to decide
whether or not to plant a new rice variety which promises to increase yield but only if fertilizer is
applied. The farmer estimates there is a 40 per cent chance that fertilizer will not be available.
Growing period is 90 days for the new variety and 120 days for the old one. With the new variety,
double cropping becomes feasible. However, double cropping will give a good yield only if the
monsoon is not early. Should the monsoon be early, the farmer risks losing not only the second crop
but also its planting costs. The farmer believes there is a 50 per cent chance the monsoon will be
favourable, i.e., not early. Two decisions - old or new variety? and single or double crop? - and two
chance events - fertilizer availability and timing of the monsoon - are involved in this decision
problem. As shown in Figure 11.8, it can be depicted as a two-stage decision tree to be solved using
certainty equivalence. Note that double cropping and timing of the monsoon are irrelevant if the old
variety is planted. The possible outcomes for each possible combination of choices are based on
budgeting of each possibility.
Analysis of the decision tree of Figure 11.8 proceeds by determining the CE of each terminal risky
prospect and then working backwards through the tree. Thus at decision node 4 there are two
prospects: the risky prospect ($1 200, $2 000; 0.5, 0.5) and the sure prospect of $1 300. Suppose the
farmer has a CE of $1 500 for the risky prospect. This implies that, if at decision node 4, her or his
preferred choice would be to double crop. Likewise, suppose the farmer's CE for the risky prospect
at event node 7 is $850. This implies that, if at decision node 5, her or his preferred choice would be
to single crop with a sure prospect of $900. The decision tree can thus be folded back to the single-
stage tree of Figure 11.9 involving choice between just two risky prospects, i.e., ($1 500, $900; 0.6,
0.4) vs ($1 000, $950; 0.6, 0.4), for which suppose the farmer has CEs respectively of $1 150 and
$970. His or her optimal choice at decision node 1 is thus to plant the new variety corresponding to a
CE of $1 150. Having chosen the new variety and finding fertilizer to be available, the farmer should
choose to double crop at decision node 4 of Figure 11.8. But if fertilizer turns out not to be available,
he or she should choose to single crop at decision node 5.
The use of decision trees is not without significant difficulty. Attempts at full depiction of risky
decision problems may often lead to trees that are a bushy mess and too difficult to comprehend.
Artistry is needed to reduce the tree to the essentials of the decision problem. In particular,
continuous variables need to be replaced by discrete approximations. Nonetheless, construction of a
decision tree can be a valuable exercise. It forces the farm manager to specify all the relevant acts,
events, payoffs and probabilities, as well as the sequence in which acts and events occur. Doing this,
even without solving the decision tree, can greatly enhance the farmer's understanding of the risks he
or she faces.
Specification and appraisal of decision trees can also be facilitated by the use of such personal
computer software as DATAä (TreeAge Software Inc. 1996) and PrecisionTreeä (Palisade
Corporation 1996).

Also known as risk efficiency analysis and stochastic efficiency analysis, stochastic dominance is a
means of comparing alternative risky choices directly on the basis of their outcomes' cumulative
probability distributions (Dillon and Anderson 1990, pp. 146-153; Dillon and Hardaker 1993, pp.
241-253; Hardaker, Huirne and Anderson 1997, pp. 145-152). In essence, it is another form of
subjective expected utility analysis but one that is not so discriminating between alternative choices
as are direct utility comparison, certainty equivalence and decision tree analysis. As in these other
utility-based approaches, stochastic dominance analysis should usually be conducted on the basis of
aggregate outcomes, not on a technical unit basis. Stochastic dominance analysis does not necessarily
lead to the one best choice. Rather, by comparison of their outcome distributions, stochastic
dominance analysis sorts possible risky choices into two groups: first, those that should not be taken
because they are dominated by or are less preferred/inferior to a second group which is not
dominated. In making this sort into two groups, stochastic dominance analysis relies on general rather
than specific features of the decision maker's utility function.
FIGURE 11.8 - Example of a Decision Tree for a Risky Decision Problem
The principles of stochastic dominance are illustrated in Figure 11.10. This shows the cumulative
probability distributions or cumulative distribution functions (CDFs) for the outcomes of five
alternative decisions: a sure prospect A, a risky prospect B with four discrete outcomes and three
risky prospects C, D and E with continuous outcomes. The first general feature of utility used in
stochastic dominance analysis is that 'more is preferred to less', i.e., the utility function U(X) slopes
upward to the right. Considering the CDFs in Figure 11.10, this implies that the possible decisions A,
B and C are clearly less preferred to (or dominated by) the alternative decisions D and E. The CDFs
for D and E lie everywhere to the right of those for A, B and C so that, at any level of (cumulative)
probability, D and E offer better possible outcomes than A, B or C. Decisions/acts/choices A, B and
C should therefore not be taken. Rational risky choice lies only between D and E, neither of which
dominates the other because their CDFs intersect.
FIGURE 11.9 - Folding Back of the Decision Tree of Figure 11.8 on the Basis of Certainty
Equivalence
To discriminate further between possible choices, stochastic dominance analysis requires the
additional assumption that the decision maker is risk averse. Consider the CDFs for the two
undominated decisions D and E in Figure 11.10. Because their CDFs intersect, neither of these
decisions can be said to dominate the other on the basis of more being preferred to less. However,
because the area a is greater than the area b , decision E would always be preferred to decision D by
a risk-averse decision maker. The general rule is that if, moving from left to right, the cumulative
difference in area between two CDFs that cross is always positive, then the rightmost CDF at
probability zero will always be the preferred choice if the decision maker is risk averse. This rule
means that a CDF with a lower tail lying to the left of another distribution can never dominate it. A
number of more advanced techniques of stochastic dominance analysis is also available. These
achieve greater discriminatory power among alternative decisions by placing bounds on the decision
maker's degree of risk aversion (Dillon and Anderson 1990, pp. 150-154; Hardaker, Huirne and
Anderson 1997, pp. 149-153).
FIGURE 11.10 - Stochastic Dominance Analysis of Five Alternative Decisions A, B, C, D and E
The attractive features of stochastic dominance analysis are (1) that it does not require knowledge of
the decision maker's utility function, (2) that it is based on direct comparisons between full
probability distributions of outcomes and (3) that it can be carried out on a personal computer with
the use of such add-in specialist software as @Risk (Palisade Corporation 1997). The disadvantages
are (1) that it requires knowledge of the decision maker's subjective probability distributions for the
possible outcomes of all the relevant alternative decisions, (2) that it necessitates pairwise
comparisons of risky alternatives, the potential number of comparisons rising exponentially with the
number of alternatives, thereby quickly making the assessment burdensome unless use is made of such
software as that of Goh et al. (1989); and (3) that it generally does not lead to the one best alternative
but rather to some set of undominated alternatives.
Relative to resource-poor small farmers, stochastic dominance analysis is, except for possible use on
a case-study basis, not directly useful on an individual basis. As with the other techniques of risky
decision analysis, there are too many small farmers (each with their unique personal preferences and
degrees of belief) to make general use of the technique either feasible or practical. However, some
applications oriented to groups of farmers may be worthwhile. Thus stochastic dominance analysis
may be used to evaluate new technology, e.g., in plant breeding programs to assess possible new
varieties aimed at particular farming systems or agro-ecological zones. Such analysis would need to
be based on probabilities developed by the analyst through discussion with relevant experts.

The various approaches to risky choice outlined above have all been more oriented to partial rather
than whole-farm decision making. This does not fit well with the fact that a farm is a system. Partial
approaches are likely to miss system-wide implications arising from the interrelatedness of the
system. As shown in Chapter 9, mathematical programming provides a means of casting farm planning
and management in a whole-farm system context.
A variety of mathematical programming approaches suited to whole-farm planning under risk has
been developed (Hazell and Norton 1986, Ch. 5; Hardaker, Pandey and Patten 1991; Hardaker,
Huirne and Anderson 1997, Ch. 9). These may be divided into two classes: first and by far the
simpler, those oriented to single-stage risky decision situations (as, e.g., depicted in Figure 11.7)
with uncertainty only present in activity net returns or gross margin (i.e., in the Z values of Table 9.9);
and, second, those oriented to multi-stage risky decision situations with consequent uncertainty about
future input-output coefficients and resource constraints. The latter refer to situations involving more
than a single act-event sequence so that initial choices (plans) can be revisited (adjusted) by later
decisions as relevant uncertain events unfold over time (as, e.g., in Figure 11.8). Such situations are
referred to as having embedded risk and are modelled by what are referred to as stochastic
programming models, single-stage or non-embedded risk situations are modelled by risk
programming models (Hardaker, Huirne and Anderson 1997, Ch.9; Rae 1994, pp. 338-348).
Simplified programming and linear programming, as outlined in Chapter 9, are usually based on
expected (i.e., mean) rather than sure activity net returns or gross margins. Ideally, the plans they
generate would suit risk-neutral farmers but not necessarily the risk-averse majority of farmers. A
variety of approaches has been developed to overcome this deficiency. First was quadratic risk
programming, a form of non-linear mathematical programming, based on the strong assumption that
utility is maximized in terms of the mean and variance of the probability distribution of total net
revenue. Such problems are now readily solved using software such as GAMS (Brooke, Kendrick
and Meeraus 1992) which incorporates the powerful non-linear programming package MINOS.
Earlier, lack of non-linear software led to the development of linear approximations to quadratic
programming such as MOTAD, Target MOTAD and mean-Gini programming (Hardaker, Huirne and
Anderson 1997, Ch. 9). Today, with the general availability of such software as GAMS, these earlier
approaches to risk programming have been superseded by non-linear mathematical programming
models which (1) directly maximize expected utility given knowledge of the decision maker's utility
function or (2) maximize expected utility for specified degrees of risk aversion (Patten, Hardaker and
Pannell 1988; Hardaker, Huirne and Anderson 1997, Ch. 9). The latter method, known as utility-
efficient programming, is obviously more feasible than the former. Both, however, necessitate
knowledge of the probability distributions of total net returns from alternative allocations of farm-
system resources. Again, as for the other utility-based approaches to risky choice discussed earlier in
this chapter, these utility-based risk programming methods are hardly applicable, except on a case-
study basis, to the mass of small farmers. They could be of use to larger commercial farmers able to
buy such special planning services.
Little needs to be said about stochastic programming. As noted earlier, it is aimed at risky decision
making where resources initially have to be allocated at the first stage of multistage act-event
sequences, i.e., when embedded risk is present. Stochastic programming suffers from 'the curse of
dimensionality' - the complexity of the decision model increases exponentially with the number of
stages and the number of possible outcomes at each stage. With more than three stages and just a few
possible outcomes at each stage, solution is difficult (Hardaker, Huirne and Anderson 1997, pp. 196-
203). So, while the risky decision problems at which stochastic programming is aimed are real and
important, in practical terms their complexity as yet defies analytical solution. The small-farm analyst
can forget stochastic programming as a practical guide to real-world farm planning.

Certainty equivalence is probably the most useful formal approach to such risky annual farm
decisions as choice of crop variety, planting date, fertilizer rate, pest control measures etc. For longer
term investment decisions, Monte Carlo simulation is probably the most useful formal approach. It is
applicable to any level of the farm system but is especially appropriate as a guide to (1) the optimal
size/level/investment in expensive capital items (e.g., dams, cocoa dryers, irrigation systems) and (2)
the likely profitability of important single activities, especially on large mono-crop farms and estates.
To illustrate its use, Monte Carlo simulation is applied in Section 11.15.2 to appraise the merit of a
farmer's possible investment in banana production.
The broad purpose of simulation is to gain 'experience' in the operation of an activity. There are
broadly two ways of acquiring such experience. The first is to actually carry out the investment and
then sit back and watch what happens to it over the next 10, 20... years as it is subjected to the
occurrence of risky weather and other events. Usually, because of cost and the required long time
period, this is impractical.
The second way of gaining experience is to build a model which mimics the essential elements of the
activity. This is then subjected to the same array of risky events as the real activity would face.
Information is thus obtained about the likely performance of the simulated activity. Such a model is a
simulator. The use of simulators is now so widespread that little discussion is needed: airplane-pilot
training simulators; model cars that train adults in driving skills and children in road safety; mock
control panels on which technicians learn how to operate a power station - these are now
commonplace. Their agricultural equivalents are activity and whole-farm budgets when these are
used for ex ante planning purposes; and for more complex activities, extensions of such budgets in the
form of mathematical programming and simulation models.
What all simulators have in common is that they provide knowledge at relatively little cost - how to
react to an explosion in a power station without actually blowing it up; whether bananas or coconuts
would be the best investment without actually planting them. The second thing they have in common is
their ability to compress the acquisition of experience into very short periods of time - e.g., in the
case of banana growing, only the ten days or so needed to construct and run the model rather than the
15 to 20 or so years of real time required if a field trial were to be conducted.
Simulation models are of two broad kinds: those that ignore risk and those that attempt to incorporate
it. The former are concerned with the imitation of some prescribed event or event series that is
assumed to have sure consequences. The cardamom investment of Section 10.12 is of this kind: the
'events' are the specified crop production operations and inputs of Table 10.5, and the 'consequences'
are the activity net returns of Table 10.7.
The second kind of simulation model is concerned with evaluating the consequences of risky events.
Perhaps the simplest example of simulation of this second kind is a farmer's (perhaps introspective)
consideration of past rainfall patterns to assess the likely performance of a future crop. For obvious
reasons, formal models of this second kind are referred to as stochastic or Monte Carlo models. Such
models incorporate probabilities for the occurrence of relevant risky events. By repeated running
(i.e., operation) of the model, with each run based on its own sampling from the incorporated
probabilities, a sample of simulated outcomes is generated (Anderson, Dillon and Hardaker 1977,
pp. 267-272; Hardaker, Huirne and Anderson 1997, pp. 45-46, 226-228; Rae 1994, pp. 300-303).
Given a sufficient number of runs, an estimate of the probability distribution of possible outcomes
(measured on an aggregate rather than on a technical unit basis) can be obtained. This probability
distribution can then be compared with those of other alternatives as discussed above in relation to
risky choice based on expected utility, certainty equivalence or stochastic dominance analysis.
There are seven general steps in constructing and operating a stochastic simulation model:
1. Define the objectives. This will determine the nature of required output and the form in which it will be most useful.
2. Initiate a flowchart sketch. This is a working document showing the interrelationships in the
model and is subject to ongoing revision as the model is developed and used.
3. Provide the input. This consists of numerical raw material on which the model is to operate. In the
following example, input consists of:
- a base cost/return budget (Table 11.10).
- a set of weather records and associated economic events (Table 11.11).
- a table of subjective probabilities of occurrence of relevant economic events (Table 11.12).
- an identification code applied to the various economic events (Table 11.13).
- a table of random numbers.
4. Formulate management and operating rules. These rules or conventions specify (a) the general
conditions under which the model is to operate (e.g., the number of simulated periods or years for
which it is to generate output), and (b) the way in which the model is to behave in all relevant
circumstances.
5. Trial and calibration. Verification of the model necessitates its calibration against some
independent data series of relevance. Too, results from the first few runs of a model may not be of
acceptable quality. If so, the model will have to be adjusted and refined (primarily through adjusting
the operating rules).
6. Operation. Having been calibrated, the model is run repeatedly according to the revised rules and
its output results are accepted. The number of runs of the model is determined by considerations of
cost and of the precision required in the estimation of the probability distributions of relevant
variables.
7. Interpret and extend the output. The results of step 6 might be of direct use in themselves, but
often they require further subjective interpretation, elaboration, explanation if they are to be
understood by output consumers.
Step 2 above - development of a flowchart - deserves a little further comment. While deciding on the
broad configuration of the simulation model (i.e., what output is required? what input is available?
what operating rules are necessary?), a parallel step is to develop the interrelationships among these
model components by preparing a flowchart. This serves two purposes. Initially it is a working
document, a scratch sheet on which specific model components are specified, arranged, re-arranged
in logical order. When finalized (if indeed it ever can be regarded as 'final', because nearly all
models can usually be improved upon and refined), the flowchart becomes the basis on which a
computer program can be written (Section 11.15.3). Use of computers in simulation is, however, by
no means mandatory; simple problems can be adequately analysed with a hand calculator.
Background
In following sections a simulation model of a banana production activity is constructed and operated
for the purpose of deciding whether a farmer should invest in this type of activity. It is assumed the
farm is located in an area for which long-term weather records are available. In this area bananas are
a generally profitable investment but, because of tropical storms which periodically reduce yields
and destroy or damage plantations, the activity is a high-risk one. In any given year there can be wide
variations between expected yield, cost and profit levels for a plantation of given age and those
results which actually occur because of the effect of storms. Meteorological records of past storms
are available. These are used as the basis for stochastic simulation of the results of establishing and
operating a banana activity under typical conditions in the area. First, however, it is useful to note the
several different economic effects which storms can have on long-term crops.
From the viewpoint of storm damage, multi-year/perennial crops fall into three groups. The first
group consists of crops such as banana which might suffer a yield-effect involving partial or complete
loss of current yield, depending on storm intensity. The first cost-effect will be in the form of costs of
cleaning up the plantation. A second possible cost-effect might also arise in the need to completely
re-establish/replant a plantation if it has been completely devastated. However, the need to replant
also depends on the age of the plantation at the time of storm occurrence: a storm of given intensity
might cause rehabilitation costs in a young crop or a different level of re-establishment costs on a
planting which is nearing the end of its useful life.
The second group consists of the larger more massive long-life species - mango, jak, coconut, lychee
etc. These might suffer partial or complete loss of current output plus complete or partial destruction
of the trees themselves. Also, considering the large volume of vegetative material which might have
to be removed from the field, the cost of replanting crops of this group after a hurricane might well
exceed the cost of their initial establishment if this had been on clean ground.
In the third group, consisting primarily of vine crops grown on artificial or live trellis supports
(pepper, vanilla, passionfruit etc.), the first effect again might be loss of current yield. The second
effect, complete plant destruction requiring replanting, will seldom occur. But a third effect - trellis
destruction - is common and can be economically more serious than the effect of the storm on the
plants themselves. This group also includes ratoon crops such as sugarcane. Again there might be loss
of current yield, but the main effect will be in the form of delayed harvesting or increased harvesting
costs. If the crop has recently been harvested, however, there will often be no yield - or cost-effect at
all, regardless of storm intensity.
In addition to their direct yield- or cost-effect on crops, storms might also have indirect cost-effects
through damage to the environment or general farm infrastructure. In following sections only those
direct effects noted for the first group of crops are considered.
Activity time profile and base budget
As previously with the cardamom investment of Section 10.12, the first step in activity evaluation by
simulation is to define the time profile of the activity. Bananas are an activity of intermediate life-
span. Previous bush fallow is slashed down and left to decay in place, planting points are cleared and
corms are planted. In Year 2 the stand is maintained but no crop taken. Yield begins in Year 3 and
continues for a total of four to six years (here four years is assumed). The plants are then slashed
down and the land allowed to revert to bush fallow. Meanwhile, another planting is coming into
production on a second block. Thus to produce one ha of crop, about two ha are usually operated on
average, one under crop and the second under fallow. On a given farm, each individual intermediate-
life planting is one phase of a long-term (continuing, indefinite) banana activity.
Here the activity is evaluated over three six-year crop cycles. These could occur over a maximum of
3 x 6 = 18 years; but if one or more cycles are prematurely terminated because of storm damage, the
total three-cycle period will be less than 18 years.
The time profile of the activity provides the necessary framework within which to construct the base
budget of Table 11.10. These data are self-explanatory. It is necessary only to note that the specified
yields and costs are those that would apply to the activity in the absence of storms. (Prices are
assumed constant.) Subsequent operations of the simulation model discussed below are concerned
largely with making adjustments to these base yield and cost data to allow for the random effects of
storms of varying levels of intensity.
TABLE 11.10 - Base Budget for Simulation Analysis of Banana Investment (per Ha Basis)
Year of crop cycle

Measure
1 2 3 4 5 6
Yield (cases) 0 0 150 250 200 100
Price ($) 2 2 2 2 2 2
Gross return ($) 0 0 300 500 400 200
Total cost ($) 160 80 60 60 40 20
Net return ($) -160 -80 240 440 360 180
Storm incidence and effects

The base-budget data of Table 11.10 must now be adjusted according to possible storm damage in
any crop-cycle year. The first step is to quantify these future possible storm events. The best guide to
what to expect of future weather is what has happened in the past complemented by any acceptable
forecast information that may be available. This does not mean that the past sequence of weather
events and intervals between them is expected to be exactly duplicated in the future. Rather, it is to
say that if three hurricanes have occurred in the past 100 years, it is a reasonable subjective
judgement to expect three more in the course of the next 100 years. The particular future years in
which they will occur cannot be forecast, nor their intensity, nor the sequence in which future storms
of varying intensity - hurricanes, gales, storms, 60-knot winds etc. - will arrive.
Storm records for the area extend back to 1909. These are shown in the first two columns of Table
11.11 for the 87-year period to 1995. On the assumption that this past pattern of weather events will
occur again (but only in some unknown random order), it is now necessary to quantify the economic
effects of storms of various intensity levels. As tabulated in Table 11.11, in order of decreasing
strength there are five types of storm: hurricanes (or tropical cyclones or typhoons), severe gales,
gales, moderate gales and storms. At this point, the objective is to determine the effect of weather
events of each of these intensity levels on the economics of banana production.
In the present example, no recorded information on the relation between storm intensity and damage is
available. Bananas in the area are a smallholder crop and growers do not keep production records for
more than a few years. So several operations towards defining some working relationship between
storm intensity and economic loss are necessary. First one would try to establish some informal
correlation. Although formal production records are not kept, the memories of some individual
growers, older family members and the villages collectively extend back many years. Armed with the
weather record of Table 11.11, the analyst would go into the villages and ask two kinds of questions
about the effect of storms on banana yields and costs: 'What did happen ... (to past crops as a result of
past individual storms?)'; and 'What do you think would happen ... (if a hurricane, severe gale, gale
etc. were to hit the plantations today?)'.
After enough farmers in enough villages - say around 50 farmers in all - had provided their subjective
recollections and estimates, a schedule relating storms of the five intensity levels to their respective
yield- and cost-effect on bananas could be constructed. Ideally this would be in percentage terms:
e.g., a past hurricane or severe gale would have caused a decrease in yield (and gross return) of 50
per cent (relative to what it otherwise would have been as per Table 11.10) and a cost increase of
100 per cent of the initial establishment cost of $160 per ha (Table 11.10); a moderate gale would
have reduced crop income by 30 per cent and increased cost by 60 per cent of initial establishment
costs etc.
These percentage effects of past storms of several intensity levels are tabulated in the two rightside
columns of Table 11.11. However, when they are applied to future storms the magnitude of effect is
also partly determined by the place of that storm in the storm sequence (whether a strong gale follows
a hurricane etc.) and by the age of the plantation at the time of storm occurrence. The various
situations are provided for below in the formulation of model operating rules.
TABLE 11.11 - Incidence and Effect of Major Weather Events on Banana Crops in the Study
Area, 1909 to 1995
Weather record Effect

a b
Year Type of event Yield loss Plant damage
(%) (%)
1909 Hurricane 50 100
1911 Severe gale 50 100
1914 Moderate gale 30 60
1918 Gale 40 80
1927 Storm 20 40
1932 Storm 20 40
1933 Gale 40 80
1935 Storm 20 40
1955 Gale 40 80
1964 Gale 40 80
1966 Gale 40 80
1967 Gale 40 80
1971 Gale 40 80
a Measured as percentage of base-budget yield (Table 11.10). Translates directly to loss in gross revenue.
b Translates into a cost increase of this percentage of initial plantation establishment cost of $160 per
ha.
Probability of storm losses and their Monte Carlo sampling
Having developed the relationship between storms of the five intensity levels and their economic
consequences, the next step is to arrange these latter data as a probability distribution. This is done in
Table 11.12. Storms which result in a 50 per cent yield loss and a 100 per cent damage level are
given an event identification number of I; those which result in 40 per cent yield loss and 80 per cent
damage are given an event identification number of II etc. (This identification system refers to the
level of economic effect, not necessarily to the level of storm intensity because, as noted above, the
level of storm effect will depend partly on time of occurrence of the storm in relation to the banana-
age cycle (see Operating Rules below).)
TABLE 11.12 - Probability of Storm Losses in Banana Production in the Study Area
Type of storm
Aspect
I II III IV V
Yield loss (%) 50 40 30 20 0
Plant damage (%) 100 80 60 40 0
Frequency in 87 years 15 7 6 3 56
Probability of occurrence 0.17 0.08 0.07 0.03 0.65
Based on Table 11.11, the data of Table 11.12 indicate that economic-loss levels of 100 and 50 per
cent (damage costs and fruit loss, respectively) occurred in 15 of the 87 years on record. They have
occurred with a relative frequency of 15/87 = 0.17 and can be expected to occur again in any future
year with a probability of 0.17. The probabilities of occurrence of the other categories of
progressively less serious losses/costs are 0.08, 0.07 and 0.03, and there is a 0.65 probability that in
any future year no economic losses of either kind will occur. Though based solely on historical data,
these probabilities are subjective because they are accepted by the decision maker, i.e., he or she
takes them as his or her degrees of belief in storm occurrence.
Having Table 11.12, the next step is to construct Table 11.13 which assigns an identification code to
each of the five possible economic-loss events I to V of Table 11.12, based on the probability of
occurrence of each respective event. Since an event of type I is expected to occur in 17 per cent of
future years, it is assigned the set of 17 code numbers 0 to 16 within the set of numbers 0 to 99; an
event of type II with a probability of occurrence of 0.08 is given the set of eight code numbers 17 to
24 etc. as shown in Table 11.13.
The last piece of input needed in model construction is a set of random numbers (as found in many
introductory statistical texts or as generated by computer) containing values over at least the same
range as the values chosen for the event identification code, i.e., 0 to 99 in the present instance. Part
of such a random-number set might be the randomly chosen sequence:
46 93 38 59 16
4 39 74 69 34
67 27 31 19 98
41 37 9 21 81
40 87 76 8 72
TABLE 11.13 - Assignment of Event Identification Code Numbers to Economic-loss Events in

Banana Investment Simulation Exercise
Economic-loss event Code numbers
Yield loss Plant damage

Event
(%) (%)
I 50 100 0 to 16
II 40 80 17 to 24
III 30 60 25 to 31
IV 20 40 32 to 34
V 0 0 35 to 99
A run of such random numbers is used to generate the risky event sequences to which the model will
be subjected. Thus, referring to the identification code numbers specified in Table 11.13, the number
46 indicates the occurrence of an economic-loss event of type V, 93 indicates the same, likewise for
38 and 59, while 16 indicates an economic-loss event of type I. This procedure is referred to as
Monte Carlo sampling. It is the mechanism for making the simulation stochastic (Dillon and Hardaker
1993, Section 5.5).
This concludes the third step in model construction, preparation of input, as listed in Section 11.15.1.
The next step is the formulation of operating rules.
Operating rules for the simulation model
The second part of a simulation model consists of operating rules or conventions which govern the
behaviour of the model under all relevant circumstances (here under all possible relevant weather
conditions). The rules formulated for the banana example are listed and discussed briefly below.
Rules 1 to 3 are general assumptions relating to how the model is to be managed, e.g., the time-period
over which it is to be run. Rules 4 to 10 are technical conventions which, as noted earlier, have been
developed largely on the basis of what smallholder growers in the area collectively think would
happen to yields and costs as a result of possible storm events occurring over a crop cycle, supported
by consideration of the effects of past storms.
Rule Evaluation period. The banana activity will be evaluated over three cycles. In the absence of
1 storms this implies a total period of 18 years.
Rule Cycle length. Each normal cropping cycle will be over six years; however, if or when severe
2 100 per cent or 80 per cent plant damage occurs in Year 5 of any cycle, that cycle will be
terminated at that point. (Storm severity levels are discussed in Rules 8, 9 and 10 below and,
for historical storms, were listed in Table 11.11).
Rule Each cycle after the first will start in the year following the end or termination of the previous
3 cycle. Thus Year 1 of the second cycle will occur in Year 7 of the activity, assuming the first
cycle has run its normal six-year course. (This will result in production 'gaps' between cycles
but farmers have other income sources such as coconuts, jak etc. The assumption simplifies
discussion. The alternative which many farmers do in fact adopt is to operate three rather than
two fields of bananas.)
Rule The economic effects of all storms of whatever intensity will be in terms of effects on returns
4 and costs as given in Table 11.12.
Rule In the absence of storms the annual net return of the activity will be as shown in the base budget
5 of Table 11.10.
Rule Plant damage caused by a storm will be repaired in the year of storm occurrence. The cost of
6 this will be proportional to the severity of the storm and based on the cost of establishing the
plantation in Cycle Year 1 ($160). Thus a storm causing 100 per cent damage will increase
activity costs in the year of occurrence by $160, a 20 per cent damage storm by $32 etc.
Rule The costs of storm damage repair will be in addition to the normal operating costs of the
7 relevant year as listed in the base budget of Table 11.10, i.e., total costs in any year will
consist of normal operating/maintenance costs plus damage repair costs (if any).
Rule A severe storm causing 50 per cent or 40 per cent yield loss in any year will result also in 25
8 per cent or 20 per cent yield loss, respectively, in the following year if it is storm free; less
severe storms will cause yield loss only in the year of storm occurrence.
Rule If a storm causes 100 per cent or 80 per cent plant damage in Cycle Year 5, that field will be
9 abandoned and a new planting will be commenced (in a second field) in the following year.
This was noted in Rule 2 above.
Rule If a storm of any intensity occurs in Year 1 of any cycle, it will have no economic
10 consequences. (Newly planted banana palms protected in deep planting holes are storm-
immune.)
At this point these operating rules are tentative. The model would now be subjected to a few trial
runs. Depending on the acceptability of results, the rules might be retained, adjusted or perhaps other
rules added. Discussion of this calibration step is here deferred to a later section.
Operation of the model
The model is 'run' by taking the gross return/total cost/net return lines of the base budget of Table
11.10 and successively adjusting these activity 'results' according to the occurrence of storm-caused
economic-loss events (Table 11.12). The occurrence of these events is determined by (successive)
selections from a set of random numbers, each of which corresponds to the identification code of one
of the five possible risky events (Table 11.13).
Since the evaluation is to be over three consecutive cycles of crop, it is convenient to refer to model
operation over these three cycles as one 'run'. The results of the first run (of three crop cycles) are
tabulated in the worksheet of Table 11.14. The starting point for Run 1 is provided by the base
budget's gross return/total cost/net return data of Table 11.10. 'Activity Year' refers to the point in
time occupied by a production year, starting with the first year of the first cycle and continuing to the
last year of the last cycle. (Some cycles might not complete their normal six-year term.) 'Cycle Year'
refers to the production year within each successive crop cycle.
TABLE 11.14 - Worksheet for a Run of the Banana Monte Carlo Simulation Model (per Ha
Basis)
Cycle Year: 1 2 3 4 5 6
Base budget:
Gross return ($) 0 0 300 500 400 200
Cost ($) 160 80 60 60 40 20
Net return ($) -160 -80 240 440 360 180

Run1/Crop Cycle 1/Field 1
Activity year: 1 2 3 4 5 6
Random number: (46) (93) (38) (59) (16) Stop
Economic loss:
Gross return (%) 0 0 0 0 -50
Cost (%) 0 0 0 0 0
Simulated outcome:
Gross return ($) 0 0 300 500 200
Cost ($) 160 80 60 60 40
Net return ($) -160 -80 240 440 160
Activity year: 6 7 8 9 10 11
Random number: (4) (39) (74) (69) (34) (67)
Economic loss:
Gross return (%) 0 0 0 0 -20 0
Cost (%) 0 0 0 0 +40 0
Simulated outcome:
Gross return ($) 0 0 300 500 320 200
Cost ($) 160 80 60 60 104 20
Net return ($) -160 -80 240 440 216 180
Activity year: 12 13 14 15 16 17
Random number: (27) (31) (19) (98) (41) (37)
Economic loss:
Gross return (%) 0 -30 -40 -20 0 0

Cost (%) 0 +60 +80 0 0 0
Simulated outcome:
Gross return ($) 0 0 180 400 400 200
Cost ($) 160 176 188 60 40 20
Net return ($) -160 -176 -8 340 360 180

As a starting point, the gross return/total cost/net return lines of the base budget of Table 11.10 are
reproduced at the top of the worksheet of Table 11.14. The next step is to select a random number for
Year 1. It is 46. From Table 11.13 this number falls within the event identification code range 35 to
99, signifying the occurrence of event V, i.e., a year with no storm. (In any case, even had a severe
storm occurred in Year 1, it would have been ignored according to Rule 10.) Since there is no storm
effect in Year 1, the simulated outcome is the same as that of the base budget, i.e., a net return of -
$160. Analysis then moves on to Year 2.
A second random number is selected, 93, again signifying a storm-free year. This is repeated in Years
3 and 4; but in Year 5 the random number 16 brings with it a hurricane or strong gale. From Table
11.12 this event will decrease current gross return (through fruit loss) by 50 per cent and also, by
Rule 6, increase cost by 100 per cent of initial plantation establishment cost, $160. However,
according to Rule 9, since the field is nearing the end of its useful life, it would instead be
abandoned. It will therefore not incur clean-up costs. The only economic loss due to this hurricane
will be loss of 50 per cent of the yield/gross return. A new field will be started the following year,
i.e., in Activity Year 6 (Rule 2).
Continuing to Year 1 of the new crop cycle (Activity Year 6), the random number of four again
signals a major storm; but according to Rule 10 this causes no loss. The following Cycle Years 2, 3,
4 are storm free. Then in Cycle Year 5 the random number 34 signifies an economic-loss event of
type IV causing a gross return decrease of 20 per cent and a cost increase of 40 per cent. Since this is
a minor rather than a major storm (Rule 9), the field will be cleaned up rather than abandoned. Gross
return this year will be $400 - 0.2($400) = $320, while total cost will be the normal annual current
cost of $40 plus 40 per cent of the initial establishment cost of $160, i.e., a total of $104. Net return is
thus $216. The following year, Cycle Year 6, is storm-free (random number 67); and there is no
carry-over damage from the storm of the previous year (Rule 8) because that would have been
incurred only by a type I or II storm.
Continuing, Year 1 of the next (third) cycle, although suffering a type III storm, incurs no economic
loss by virtue of Rule 10. However, a similar storm in Cycle Year 2 does cause an increase in cost.
In Cycle Year 3 a type II storm occurs which has direct effects on returns and costs. However,
according to Rule 8, it also has the secondary effect of reducing yield/gross return in the following
year, which would otherwise have been $500, by 20 per cent, i.e., to $400. Except for this effect in
Cycle Year 4, the rest of the crop cycle is storm-free. This concludes Run 1 of the Monte Carlo
simulation.
Economic evaluation of Run I
The simulated results of operating the activity over three cycles for a total of 17 years constitute an
irregular stream of annual net returns (some of which are negative). For purposes of comparison with
other possible activities/investments, it is convenient to convert this stream to a present value (PV) as
per Section 10.7. As calculated in Table 11.14, the stream of net returns from Run 1 of the model is
$-160, -80, 240, 440..., 360, 180 occurring in sequence over Activity Years 1 to 17. Assuming an
annual discount rate of ten per cent, this stream has a PV of $773. On an amortized annual basis, this
is equivalent to an annuity of $96 per year over the three-cycle or 17-year evaluation period (Section
10.10).
The particular result of Run 1 is in a sense accidental: it occurred as a result of the 'accident' of
selecting the particular sequence of random numbers 46, 93, 38... 37 from the table of random
numbers. If the table of random numbers had been entered at any other point, a quite different series
would have been selected. This would have generated a different weather experience and thus a
different final result. In other words, operating the model for only a single run does not give sufficient
experience to claim with reasonable confidence that net returns from this banana activity over three
cycles would have a PV of $773. More experience is needed.
The necessary additional experience would be gained simply by running the model repeatedly until
sufficient estimates of the activity's PV of net returns were available to give a reasonable estimate of
its probability distribution. With the aid of a computer (which can do nothing that an analyst can't do
but does it faster), any number of runs can be made - dozens or hundreds as need be - and any amount
of simulated banana growing experiences can be accumulated. For illustrative purposes, only 20 runs
of the model are used here. This is probably about the minimum number of runs on which to attempt
judgement of the banana activity. Fifty to 100 runs would be far better.
Economic evaluation of all runs
As noted in Section 11.15.1, the final step in simulation modelling consists of interpreting the results.
With 20 runs of the banana model, the results in terms of the PV of each run (assuming a discount rate
of ten per cent) might be as follows: Run 1: $773; Run 2: $516; Run 3: $810; Run 4: $498;... Run 20:
$382. Having these results, there are various approaches that might be taken to assess them from the
decision-making perspective of whether to invest in bananas or in some other alternative (for which
analogous information would presumably be available).
One obvious possibility is to take the average of this set of PVs, $529, and use this - after conversion
from its technical unit to an aggregate basis - as the basis of comparison. However, use of this single
value might imply a degree of certainty which - considering the nature of the data and the many
operating assumptions which have been made - is not warranted. Alternatively, the mean might be
paired with the standard deviation of the set of PVs to provide a better basis for the comparison of
risky alternatives.
However, considering the risk aspect, a better approach might be to consider the returns from banana
growing in terms of reliability rather than in terms of averages. To this end the simulated results can
be arranged as a reverse cumulative probability distribution (i.e., reverse CDF) as shown in Figure
11.11. This is obtained by graphing the estimated PVs from the 20 simulation runs against their
respective reverse cumulative probabilities. These latter are developed in the worksheet of Table
11.15 based on the sparse-data rule that, given m observations on a continuous random variable,
when these are arrayed in ascending order of size, the n-th observation is a reasonable estimate of the
n/(m+1) fractile (Anderson, Dillon and Hardaker 1977, pp. 42-44). Given these fractiles the CDF of
PV (i.e., F(PV)) can be specified and the reverse CDF values 1 - F(PV) calculated as in Table 11.15.
A free-hand curve of the relationship between the simulated PV values and their associated reverse
cumulative probabilities can then be sketched as in Figure 11.11. This is an income-reliability
schedule. It shows, for any level of PV, the reliability of getting this or a higher level of return from
three six-year cycles of banana production, e.g., from Figure 11.11, a return of $600 has a reliability
of about 30 per cent. Conversely, reading up from the reliability level axis indicates the minimum
level of return which can be expected at any particular level of reliability. It is apparent that high
returns are associated with low reliability. Thus a PV of $820 or greater would be achieved only
about five per cent of the time. The other 95 per cent would result in a return of less than $820.
TABLE 11.15 - Estimated CDF and Reliability (Reverse CDF) of PV of Net Returns from
Monte Carlo Simulation of Banana Production (per Ha Basis for Three Six-year Cycles)
Run no. Cumulative probabilityb Reliability

PVa
(n) F(PV) [1-F(PV)]100
$ %
1 308 0.048 95.2
2 331 0.095 90.5
3 353 0.143 85.7
4 382 0.191 80.9
5 420 0.238 76.2
6 435 0.286 71.4
7 467 0.333 66.7
8 472 0.381 61.9
9 477 0.429 57.1
10 498 0.476 52.4
11 516 0.524 47.6
12 558 0.571 42.9
13 563 0.619 38.1
14 572 0.667 33.3
15 621 0.714 28.6

16 645 0.762 23.8
17 676 0.809 19.1
18 702 0.857 14.3
19 773 0.905 9.5
20 810 0.952 4.8
a Present value (at an annual discount rate of ten per cent) of the stream of annual net returns generated by each run of the
simulation model. Listed in ascending order of size.
b Calculated as n/(m + 1) for m = 20.
An income-reliability schedule such as that of Figure 11.11 might be used as follows. The farmer or
analyst would specify the level of reliability he or she demanded of the activity before undertaking to
invest in it. This might be 80 per cent. He or she would then use the curve to read off the minimum PV
of net returns per ha to be expected with this level of reliability. At 80 per cent this is about $400. If
this particular return-reliability combination is satisfactory in the eyes of the farmer he or she would
go ahead with the banana activity; if it is not, then he or she would seek some alternative investment.
FIGURE 11.11 - Estimated Reliability of the PV of Net Returns from Banana Production
Such a return vs reliability approach is obviously practical. It is also attractive in its direct
recognition of trade-offs between income and risk. Nonetheless, it has the disadvantages (1) of
involving an arbitrary level of reliability and (2) of not assessing the probability distribution of
outcomes in its entirety. A better approach would be to use one of those based on subjective expected
utility. In particular, the decision maker might determine his or her CE for the probability distribution
of the PV of aggregate net returns from investing in bananas and compare this CE with those of the
alternative choices. Alternatively, stochastic dominance analysis as per Section 11.13 might be used
to guide choice - though, unlike the CE approach, this would not necessarily determine the optimal
choice.
Stochastic dominance analysis of the PVs of the simulated net return streams of alternative long-term
risky investments is greatly facilitated by such personal computer software as DATA (TreeAge
Software Inc. 1996) and @Risk (Palisade 1997). With a sufficient number of runs (say 250 or so) of
the simulation model, such software can be used to generate a cumulative frequency distribution of the
outcomes from each alternative. Again these outcomes should be measured on an aggregate rather than
technical unit basis. Examples are provided by Hardaker, Huirne and Anderson (1997, pp. 226-228),
Rae (1994, pp. 301-303) and Winston (1996). Such computer-generated CDFs can then be compared
by stochastic dominance analysis as illustrated relative to Figure 11.10, or by certainty equivalence,
or by direct utility evaluation if the decision maker's utility function is known. However, as noted
with the other formal approaches to risky farm decisions, such approaches to the assessment of the
outcomes from stochastic simulation are probably only relevant to investment appraisal on large
farms and estates. Except perhaps for representative-farm analysis and case studies, they are not
justified for small farms.
Model calibration
An important step in model construction is calibration or verification to ensure that results can be
used with a reasonable degree of confidence. Calibration is achieved by comparing the output of trial
runs (e.g., the simulated net return data of Table 11.14) with some independent data series and, if
necessary, making adjustments to the operating rules so as to improve the model's performance as a
mimic of reality. Either the final output of the model or some intermediate output may be used as the
basis for calibration.
Calibration of some models is easy. Models designed to simulate the operation of a farm irrigation
supply dam in which the uncertain parameters are rainfall/runoff/inflow to the dam and irrigation
outflow from the dam can often be accurately calibrated, in regard to inflow, by reference to the
behaviour of a local municipal reservoir for which good records are kept. On the other hand,
calibration of outflow (determined by rainfall and crop requirements) will usually be more difficult.
For the banana model, several calibration possibilities might exist. Thus, although the model relates
to a particular area, it might be possible to find a farm or research station either nearby or elsewhere
which has a similar weather regime and for which crop production records are available. A second
possibility might be to base calibration on the relative behaviour of some different crop which had
been subject to the same weather events as the subject banana crop. Thus there might be coconut
estates in the area which have production/marketing records extending back the required number of
years. To use such records for calibration it would only be necessary to establish some linkage, in
terms of the effects of hurricanes, gales etc., between coconuts and bananas.
The third possibility could be to seek some other relevant data series for the same crop. For the
banana model, such a series might be, e.g., the annual total of banana exports in cases from the region.
Suppose such a data series is available for the 32 years 1964 to 1995. Because this is a series of
physical quantities, the model's intermediate output in terms of percentage fruit loss due to storms -
lines 1 and 2 of Table 11.12 - provides a suitable calibration factor (and the final output of net returns
is, for this purpose, ignored). Also, for calibration purposes the model is assessed using the actual
weather-event sequence over the period 1964 to 1995 for which the calibration series, exports, is
available (rather than using randomly generated weather events as in Table 11.14). There should - it
is hoped - be some reasonable correlation between the physical yield-effects of storms at farm level
and the calibration series (exports) at regional level. (However, because the simulated outputs refer
to production on an individual farm while the calibration series refers to exports from all farms in
different localities in the region - and all localities are not affected to the same extent by a given
storm - a very close correlation could not be expected.) In Figure 11.12, the level of exports in storm
years is plotted against the simulated farm-yield decline in those same years. The graph indicates, as
hoped, that there is a strong correlation between the level of exports and storm-induced decreases in
simulated farm yield. The stochastic simulation model, at least in these terms, may be accepted as
well verified. Had the analysis of Figure 11.12 not proved satisfactory, some adjustments either to the
model's operating rules or to the storm effects of Table 11.12 would be necessary.
In summary, calibration is an essential part of simulation model building. It proceeds by comparing a
sample of results from the model with some independent data series relating to the events being
simulated. Trial-and-error adjustments are then made to the model's assumptions and operating rules
until the model is believed to be generating reliable output. In the final analysis, this inevitably
involves subjective judgement.
As noted previously, preparation of a flowchart is always a useful early step in model construction
and in more complex problems an essential one. A flowchart for the banana example would be
developed along the lines begun in Figure 11.13. This serves two main purposes. First, problem
specification and analysis are always facilitated by a flowchart. Even the attempt to construct one
will compel the analyst to think more carefully about what initial input data are needed, what
operating rules are to be applied to the data, and the internal interdependence between output from
one part of the model and its use as input to another. Second, a flowchart greatly facilitates
communication about the model both with other experts and with its intended beneficiaries. Thus,
while it is not necessary that the farm management analyst be able to write a computer program, it is
essential that he or she be able to communicate with the programmer. The best way to facilitate this is
via a flowchart. It is also now possible, using such software packages as Stella II (High Performance
Systems Inc. 1997), to develop Monte Carlo simulation model flowcharts dynamically on screen.
Whether the model should be computerized or the calculations done by hand depends on several
cost/benefit factors in addition to the obvious factors of problem complexity and computer
availability. The banana model could probably be constructed and run the required number of times
using a hand calculator in about a week. It would take about twice as long to obtain results if a
computer program has to be written. Therefore, if the model is to be used only once and then
discarded, or applied to only one farmer or a small group of farmers, it would hardly justify
computerization.
Essentially, whether computerization is justified will depend on the size of the market for model
output or the number of potential users. This can often be increased 'horizontally' by so generalizing
the model that its output is applicable to a maximum number of farmers. Simply put, a government
agriculture department is more likely to fund a model which is applicable to all banana growers in an
area than one intended to benefit only one or a few farmers; and a regional development agency is
more likely to fund a general model which is applicable across its region than one which is
applicable only to part of the region. Likewise, a model which is sufficiently general to handle a
number of crops subject to the same weather events, e.g., bananas, coffee, coconuts etc., would justify
computerization more than would a crop-specific model.
The number of potential users may also be increased by 'vertical' generalization: banks, wholesalers,
export shipping companies etc. are all firms whose operations will be subject in greater or lesser
degree to the same weather events as affect farmers. Therefore, when feasible, it is desirable to
construct a model which will meet the needs both of farmers and of these various farm-related
agencies.
FIGURE 11.12 - Calibration of Banana Simulation Model by Plot of Simulated Farm-yield
Decreases against Regional Exports of Bananas, 1964 to 1995
A final point is that, although computer use is increasingly widespread (particularly in risky decision
analysis and simulation using such software packages as DATA, @Risk and Stella II), there still
remains an aura of sanctity surrounding computer-generated output. This often obscures the fact that
such output is only as reliable as is the quality of input data and the logic of the operating rules by
which this input has been processed. If there are deficiencies in the input or illogicalities in the rules,
these will not be removed by using a computer any more than they would be by ciphering with one's
fingers instead of with clam shells. The risk of accepting fiction for fact merely because results are
obtained with a computer is greatest in those models (including most simulation models) which
contain a high degree of subjectivity in their composition. It is particularly great where there exists no
really reliable independent data series against which model operation can be calibrated.
FIGURE 11.13 - Initial Elements of a Flowchart for the Monte Carlo Simulation of Banana
Production
In addition to its use in planning long-term crops, Monte Carlo simulation has many other possible
applications in farm-system planning and management. Some of these are briefly noted below.
Application to production activities
Hay-making. This activity usually requires a minimum period of five to six days of hot, fine, low-
humidity weather. If these conditions do not prevail after the alfalfa or berseem has been cut, the crop
(or at least the current harvest if it is a semi-permanent species) will be lost. Typically, there will be
only some periods of the year when the right weather conditions will or will not exist with certainty.
There will be transitional periods when successful hay-making is a matter of chance. Simulation of a
hay-making activity based on daily weather records would determine the number of successful cuts
that can be anticipated over the year or, alternatively, the reliability that may be attached to obtaining
five, six, ... successful cuts, and thus the probable income from this activity.
Crop systems. A common planning problem in monsoon lands is to determine (a) the feasibility of
continuous paddy under given 'natural' conditions; or (b) the ways and extent by which these
conditions must be modified (e.g., by artificial grain-drying) in order for continuous cropping to
become possible; or (c) that particular crop sequence or rotation which is both feasible and
economically optimal. Here the common physical constraint is usually rainfall or irrigation water
supply. The farming system, income, employment - indeed the whole village economy - all depend on
the arrival date, intensity and duration of the annual monsoon. These are highly random events. In
long-settled areas the relative feasibility and comparative economics of various traditional
cropping/farm systems will usually have been determined long ago from centuries of experience. But
when new crops/varieties become available and in new settlement areas where experience is lacking,
economic information can often be generated by weather-based simulation.
Technology evaluation. It is a truism that not all 'improved' technologies are really helpful when
applied at village level. Many of them, whether of the biological or mechanical kind, promise much
but often deliver little. These 'empty boxes' can often be identified ex ante by testing them as
components of simulated systems. E.g., tractors often appear to be the best means of achieving
production intensification, or of ensuring that one critical crop can be produced within some limited
time period. But if the weather and soil moisture conditions which tractors require are carefully
specified, it might be found (by simulation) that these necessary conditions occur only infrequently,
and gains in the speed of performance of operations are more than offset by time spent in waiting for
the necessary conditions to occur. Slow-moving buffalo might in fact be the fastest method of land
preparation when their use is simulated within the context of a specific weather-soils-economic
environment.
Farm development through intensification. Production intensification is generally sought by
maximizing the number of crops produced on a given land area in a given time, by selecting crop
species/varieties which have relatively short growing periods and by selecting production
technologies which minimize the interval between successive crops/phases. The general planning
problem is illustrated by the flowchart of Figure 11.14, the core of which shows the time taken for
each operation in producing one cycle of a relay-planted paddy crop using alternative technologies
under a range of operating assumptions. The average time required for each operation (days of time,
not number of labour days) is shown and those operations that are weather-dependent are marked by
an asterisk. Actual values of the latter would be generated by running the model for simulated series
of weather events (rainfall amount, duration, humidity level etc.). (Each operation/technology would,
of course, have associated inputs/costs and outputs/returns; these are not shown.) Because of the
number of possible technologies at each weather-dependent production stage, there are literally
thousands of routes through the model; computer simulation is the only feasible way of evaluating
them.
Application to resource-generating activities
Discussion so far has concerned the simulation of production activities. As suggested by the
following examples, there is also much scope for the use of simulation techniques in the evaluation of
resource-generating activities within farm systems.
Farm dams and village water-supply ponds. Here the usual problem is to determine the optimal
size/capacity/cost of these structures. Operationally, the procedure consists of reconciling by
simulation some set of dam-inflow factors (rainfall, runoff) with some set of dam-outflow factors
(seepage loss, evaporation loss, irrigation pumping) in a way which maximizes benefits relative to
costs. This particular problem is usually handled by successively specifying several dam
sizes/capacities, applying water from these alternative storage situations to alternative possible crops
or crop sequences (which will generally have varying water requirements), and then extending such
application successively to some increasing size of irrigated area. As an example, five possible sizes
of dam might be specified, along with five possible crops or crop rotations to be grown on five
possible total areas of irrigated land:
Dam capacity (10 000 m3): 1 2 3 4 5
Crop sequence*: PPP PPM PMM PMS MSS
Irrigated area (ha): 5 10 15 20 25

* P paddy; M maize; S soybean.
The 5 x 5 x 5 = 125 combinations of dam size/crop/irrigated area, each of which represents one
possible activity, would then be evaluated by simulation in which, say, 20 years of historical daily
rainfall records would provide the basic inputs for determining both inflow into dam storage and crop
irrigation requirements (i.e., dam outflow). When this is done, some of the combinations would
possibly result in 'running out of water' in most or all of these 20 years, while other combinations
would result in non-use, i.e., wastage of most of the stored water most of the time. The optimal
dimension of this dam-building/resource-generating activity at, e.g., whatever level of reliability the
analyst chooses to specify, would be determined by simulation.
On-stream water pumping. Here the resource-generating activity consists of installing a pump on a
stream which has a random seasonal flow. The problem is to select the optimal size of pump. No dam
is involved; water is pumped directly from stream to field. Necessary pump size will obviously vary
inversely with the time (days) that the stream is expected to flow. Optimal pump size can most
conveniently be found by simulating the operation of a range of pumps of various capacities and cost
in relation to the water requirements of a range of different crops and/or crop areas.
FIGURE 11.14 - Flowchart for Monte Carlo Simulation Model of Alternative Rice-based Farm
Systems
Field and channel design for spate-flood irrigation. In some dry regions where it is not uncommon
for the total annual rainfall to occur in the form of one or a few intense storms of short duration, e.g.,
as in parts of Yemen, much irrigation is still done by controlled spate or wild-flood methods. Runoff
rushing down the wadi is diverted into the impounded fields and a farmer's annual grain crop might
be grown entirely on the soil moisture from the spate-flood from a single storm. There are three main
technical planning problems on these farms. These are to determine (1) the optimal size/capacity of
the wadi-to-field delivery channel, (2) the size of the individual fields to be irrigated, and (3) the
height/size of the polder embankments. If the fields are too small or the embankments too high, the
excessive depth of impounded spate water will delay planting. If they are too large, this will result in
insufficient soil moisture on which to grow the crop. Where the necessary rainfall and/or stream-flow
data are available, they would permit simulation of wild-flood systems to determine these technical
design parameters.

By their long-term nature, farm investment decisions are risky. This is true whether the decision
relates to the whole farm (e.g., buying or selling a farm) or to some particular aspect of the farm's
service matrix or inventory of capital items (e.g., investing in draught oxen), the outcomes of such
decisions consist of a risky sequence of net returns over the period of the investment. These net
returns may be in financial terms or in non-financial economic terms (i.e., in kind) such as, e.g., days
of draught power, which may or may not be converted to financial terms on the basis of opportunity
cost.
As noted previously, except on a case-study or representative-farm basis, it is not to be expected that
formal approaches to risky long-term decisions will ever be worthwhile for Type 1 (i.e., subsistence)
or Type 2 (i.e., semi-subsistence or part-commercial) farms. Formal approaches, as outlined in
Sections 11.8 to 15, may be pertinent to larger farms and particularly to estates. Indeed, the
archetypal long-term investment decision is that faced by an estate manager in deciding whether or
not to replace an existing perennial tree crop with a new crop. Such decision problems are usually
framed in the context of having payoffs constituting a stream of annual financial net returns which are
converted to a PV by application of a discount rate which is assumed to be invariant over the life of
the investment. A risk-free example is provided by the cardamom investment of Section 10.12 and a
risky example by the banana investment of Section 11.15.2.
The time- and risk-oriented planning procedures of Chapter 10 and this chapter can undoubtedly
provide some guidance to long-term farm investment decisions. However, these procedures ignore a
number of problems (Anderson, Dillon and Hardaker 1977, Ch. 8; Hardaker, Huirne and Anderson
1997, Ch. 10; Robison and Barry 1996, Chs 11 and 23). First, even though it is subjective, the
relevant discount rate is unlikely to be invariant over time. It is most likely to be uncertain and to
fluctuate year by year into the future as it is influenced by changes in the market rate of interest and by
inflation. Stochastic simulation models could incorporate such uncertainty. Second, an essential
element of risky long-term investments is that their payoffs consist of not just a single probability
distribution of outcomes but of a series of (perhaps serially correlated) probability distributions of
outcomes extending (e.g., year by year) over the life of the investment. In comparing alternative
investments, choice thus involves not choice between single probability distributions but between
streams of probability distributions. Such choice is obviously a more difficult task. Usually it is
assumed away by taking a sequence of particular payoffs and discounting this stream to a PV.
Repetition of this process using Monte Carlo sampling can then be used to generate a probability
distribution of PVs (as in the banana example of Section 11.15.2). This raises a third problem. The
PV of a stream of net returns does not give any information about the sequential pattern of the stream.
Two streams of net returns may have the same PV but markedly different sequential patterns, e.g., the
following two net return streams:
Year: 1 234 5
Stream 1 ($): 0 0 0 0 20 000
Stream 2 ($): 11 800 0 0 0 1 000

have the same PV of $12 418 at a discount rate of ten per cent. Clearly, depending on their
circumstances and the degree of imperfection in the capital market, different farmers would have
different preferences between these two streams. PV is thus no sure basis for risky choice. It should
be used with appropriate caution.
To conclude, it must be said that, while a variety of formal approaches to handling risky farm
decisions have been developed, they are no panacea. Often their cost in terms of needed information
and analysis will not be worthwhile. This is particularly so for small farmers. And when worthwhile,
as may be the case for large commercial farms and estates, they can only provide guidance rather than
sure direction. In sum, given the endemic nature of risk in farming and the complexity of farm
decisions, a farm manager's risky choices must inevitably rely in part, or even in major part, on his or
her intuition and subjective judgement.
11.17 REFERENCES
Adams, C., J.M. Stevely and D. Sweat (1995). 'Economic Feasibility of Small-scale Sponge
Fanning in Pohnpei, Federated States of Micronesia', Journal of the World Aquaculture Society
26(2): 132-142.
Anderson, J.R. and J.L. Dillon (1992). Risk Analysis in Dryland Farming Systems, FAO Farm
Systems Management Series No. 2, Food and Agriculture Organization of the United Nations, Rome.
Anderson, J.R., J.L. Dillon and J.B. Hardaker (1977). Agricultural Decision Analysis, Iowa State
University Press, Ames.
Beal, D.J. (1996). 'Emerging Issues in Risk Management in Farm Firms', Review of Marketing and
Agricultural Economics 64(3): 336-347.
Belli, P. (ed) (1996). Handbook on Economic Analysis of Investment Operations, OPR and LLC,
World Bank, Washington. D.C.
Brooke, A., D. Kendrick and A. Meeraus (1992). General Algebraic Modelling System: A User's
Guide, GAMS Release 2.25, World Bank, Washington, D.C.
Calkins, P.H. and D.D. DiPietre (1983). Farm Business Management: Successful Decisions in a
Changing Environment, Macmillan, New York.
Fleisher, B. (1990). Agricultural Risk Management, Lynne Rienner Publishers, Boulder.
Goh, S. et al. (1989). 'A Generalized Stochastic Dominance Program for the IBM PC', Southern
Journal of Agricultural Economics 21: 175-182.
Hardaker, J.B., R.B.M. Huirne and J.R. Anderson (1997). Coping with Risk in Agriculture, CAB
Hardaker, J.B., S. Pandey and L.H. Patten (1991). 'Farm Planning under Uncertainty: A Review of
Alternative Programming Models', Review of Marketing and Agricultural Economics 59: 9-22.
High Performance Systems Inc. (1997). Stella II, High Performance Systems Inc., Hanover.
Makeham, J.P., A.N. Halter and J.L. Dillon (1988). Best-bet Farm Decisions, Professional Farm
Management Guidebook No. 6, ABRI, University of New England, Armidale.
Matlon, P.J. (1991). 'Farmer Risk Management Strategies', in D. Holden, P. Hazell and A. Pritchard
(eds), Risk in Agriculture: Proceedings of the Tenth Agriculture Sector Symposium, World Bank,
Washington, D.C., Ch. 5.
Morgan, M.G. and M. Henrion (1992). Uncertainty: A Guide to Dealing with Uncertainty in
Quantitative Risk and Policy Analysis, Cambridge University Press.
Palisade Corporation (1996). Precision Tree: Powerful Decision Analysis for Spreadsheets,
Palisade Corporation, Newfield.
Palisade Corporation (1997). @Risk: Advanced Risk Analysis for Spreadsheets, Palisade
Corporation, Newfield.
Patten, L.H., J.B. Hardaker and D.J. Pannell (1988). 'Utility-efficient Programming for Whole-
farm Planning', Australian Journal of Agricultural Economics 32: 88-97.
Robison, L.J. and P.J. Barry (1987). The Competitive Firm's Response to Risk, Macmillan, New
York.
Robison, L.J. and P.J. Barry (1996). Present Value Models and Investment Analysis, The
Academic Page, Northport.
TreeAge Software Inc. (1996). DATA (v3.0), TreeAge Software Inc., Boston.
Winston, W.L. (1996). Simulation Modelling using @Risk, Duxbury Press, Belmont.
APPENDIX: MANAGEMENT, FARM
MANAGEMENT AND FARM SYSTEMS
6. REFERENCES
This appendix gives the authors' perspective on farm management in the context of management per se
and of farm systems. It outlines, first, the concept of management; second, the distinctive features of
agriculture that make farm management different from management in other fields, particularly
industrial management; third and fourth, the major characteristics of farm management theory and of
its practice; and, fifth, how government may help to make the management of small resource-poor
farms more effective. The presentation is conceptual rather than empirical and its orientation is to
small farms that are inevitably becoming less subsistence and more commercially oriented as
development occurs.

Management is nothing new. It is something we do every time we choose between alternatives as we

attempt to make the most of our lives. Women and men have been necessarily acting as managers ever
since the origin of Homo sapiens.
It is not surprising, given the necessity for food, to find many references to farm management in the
writings that have survived from ancient times. Best known in this regard are the Roman treatises De
Re Rustica of Cato (234 to 149 B.C.) and Rerum Rusticam Libritres of Varro (116 to 28 B.C.)
(Anon. 1913; Bradley 1725). Our favourite, however, is the Greek Hesiod who lived in the ninth
century B.C. In his Works and Days (Evelyn-White 1914) he gives us what to our knowledge are the
first recorded principles of farm management. Few would dispute the good sense of much maxims as
"Do not put your work off till tomorrow and the day after, for a sluggish worker does not fill his barn,
nor one who puts off his work" or "Avoid shady seats and sleeping until dawn in the harvest season ...
Be busy then, and bring home your fruits, getting up early to make your livelihood sure" or "Propitiate
[the gods] with libations ... so you may buy another's holding and not another yours." Some, however,
might query Hesiod's advice to "Do not let a flaunting woman coax and cozen and deceive you; she is
after your barn. The man who trusts womankind trusts deceivers."
As well exposited by George (1968), since the time of the ancients there has been a continuing stream
of writings on management in the context of, in particular, trade, business, warfare, politics and
statesmanship. Perhaps best known from the medieval period are two books oriented to management
in civil society and politics: Thomas More's (1478 to 1535) Utopia and Machiavelli's (1469 to
1527) The Prince. Through the eighteenth and nineteenth centuries, much of the work of the classical
economists such as Adam Smith, James Mill, John Stuart Mill and William Jevons is sprinkled with
considerations of the role, function and importance of management. Other notable contributors in this
period were Carl von Clausewitz, a Prussian general who wrote on military management; Charles
Dupin, a French engineer who emphasized worker welfare and the need for integrity and
accountability in management; and Charles Babbage, an Englishman and pioneer of computing, whose
work was seminal to the development of the scientific approach to the study of management.
Spurred on by the industrial revolution and the emergence of large-scale organizations (such as
railways and multi-branch banks), what is now known as the scientific management movement began
in about 1870. Many were involved but, today, Frederick W. Taylor - a U.S. engineer - is credited as
being the founder and driving force. His interest was in management in an industrial context, where,
in order to maximize output with a given level of effort, he argued that the scientific method had to be
applied to job determination, worker selection, creation of a good work environment and
specification of output standards so as to determine properly the task for each worker (Taylor 1911).
If the worker then achieved excellence in output, he or she was to be rewarded. If standard output
was not achieved, the worker was to be penalized.
Relative to the management of small farms, two points must be made about Taylor's contribution.
First, it provided the impetus not only for changes in the practice of industrial management but also,
since early in the twentieth century and particularly since World War II, for the scientific study of
management. This area of study has burgeoned in recent decades, particularly in industry but also in
the field of farm management. A vast literature and a myriad of texts have been generated, particularly
in industrial management. University programs in management and business administration are now
commonplace and most agricultural faculties, as well as specialist colleges, offer units in farm
management. Second, Taylor's contribution was in the context of industrial production with its
associated large number of employees whose impact on organizational performance was crucial. The
same is true of most of the industrial studies of management made since Taylor's initial promulgation
of scientific management. For this reason, much of the modem conceptualization of management with
its emphasis on the industrial context and a large workforce does not fit well within the context of
agricultural production. This is particularly so for small-farm systems using only family labour.
Nonetheless, despite the above strictures, there is much in the modem conceptualization of
management that is pertinent in a generic sense to farm management.
To elaborate the general conceptualization of management, consider the following representative
sampling of definitions:
"[Management aims to] accomplish group purposes with the least expenditure of material or human resources" (Koontz 1969,
p.415).
"[Management aims to ensure] that the organization serves its mission in an effective way, and also
that it serve the needs of those who control or otherwise have power over the organization (such as
its owners, government agencies, unions, pressure groups)" (Mintzberg 1983, p.13).
"Management is the art of getting things done through other people" (Hellriegel and Slocum 1986,
p.8).
"Management is the effective and efficient integration and coordination of resources to achieve
desired objectives" (Hitt, Middlemist and Mathis 1989, p. 13).
"Management is the process of working with and through others to achieve organizational objectives
in a changing environment. Central to this process is the effective and efficient use of limited
resources" (Kreitner 1989, p.9).
While not irrelevant to farm management, definitions such as the above clearly do not specifically fit
farming situations. However, before presenting a definition that does fit farming, major facets of the
modem concept of management (as presented by authors such as those of the definitions cited above)
should be noted.
First, successful management is seen as entailing both efficiency in resource use and effectiveness in
the achievement of goals.
Second, management is seen as a universal process involving a set of management functions. These
functions are variously elaborated either simply as planning, organizing and controlling or in more
detail as, e.g., planning, decision making, organizing, staffing, communicating, motivating,
leading, monitoring and controlling.
Third, management is seen as involving a number of roles. These are of three types: interpersonal
roles as figurehead, leader and liaison link; informational roles as monitor, disseminator and
spokesperson; and decisional roles as entrepreneur, disturbance handler, resource allocator and
negotiator.
Fourth, relative to its stakeholders, the management of an organization is seen as having three sets of
duties (Mintzberg 1983, p. 13). First of these is direct supervision and running of the organization.
Second is the management of the organization's boundary conditions - its relations with the various
environments surrounding it. The third set of duties relates to the development of the organization's
strategy which may be viewed as a mediating force between the organization and its environment. As
Mintzberg (1983, p.14) puts it: "In managing the boundary conditions of the organization, managers
develop an understanding of its environment; and in carrying out the duties of direct supervision, they
seek to tailor strategy to its strengths and to its needs, trying to maintain a pace of change that is
responsive to the environment without being disruptive to the organization." Toffler (1980, pp. 235-8)
has argued that the major changing environments to which today's management must adapt are the
physical environment in relation to resource depletion and pollution; the social environment in
relation to socially oriented special-interest groups; the information and communication
environment with its exponential growth in technology, coverage and accessibility; the political
environment with its increasingly better-informed constituency demanding transparency and
accountability; and the moral environment with increasing public concern for high ethical standards
in management.
Fifth, good management is seen as dependent on the manager possessing technical, interpersonal,
conceptual and diagnostic skills (Hellriegel and Slocum 1986, pp. 22-5). Technical skills are those
that a manager needs to perform specialized tasks within the organization. Interpersonal skills are
those needed to communicate and work well with other people. Conceptual skills relate to a
manager's ability to think in the abstract, to see relationships and opportunities, to plan, and to take a
global perspective of the organization and its environment. Diagnostic skills are those used to define
and understand situations, to recognize problems and see avenues to their solution.
Sixth, successful management is seen as the product of three essential elements, viz., ability,
motivation to manage and opportunity (Kreitner 1989, p.22). Since each of these elements is
essential, the relationship between them must be multiplicative.
Seventh, four major approaches to management have been delineated (Hellriegel and Slocum 1986,
pp. 25-9; Kast and Rosenzweig 1974; Koontz 1976). The traditional approach emphasizes order,
stability and routine procedures for carrying out the planning, organizing, leading and control
functions. This approach perhaps corresponds best to the traditional approach to farm management.
The behavioural approach emphasizes knowledge about how people behave and why they act as they
do. It assumes numerous employees and groups, and is thus less relevant to the management of small
farms. The systems approach recognizes interrelatedness within the organization and views it as a
purposeful (rather than purely mechanistic) system that needs to be considered in a holistic rather than
reductionist way. Such an approach is clearly relevant to farm management as has been recognized in
recent decades with the development of farm-systems theory and analysis. Lastly, the contingency
approach recognizes that there is no best way of managing for all situations. Different situations are
seen as requiring different management approaches - sometimes traditional, sometimes behavioural
and sometimes system-oriented. Such an approach clearly has merit in farm management.
Eighth, management is seen as probably being best described as a mixture of science, practice and
art. Successful management, as measured by efficient goal achievement, can not be guaranteed by the
scientific principles of management that are currently available. Insofar as much of management is
routine, it is a practice. Not least, successful management is often the outcome of good luck in
combination with the artistry of a creative use of knowledge, opportunity and intuition.
Ninth, quantitative methods of analysis can often be highly relevant to some management functions.
This is particularly so for planning resource allocation and decision making. However, whether or
not quantitative analysis is cost effective will generally depend upon the data required, the
complexity of the analysis, the size of the organization or scale of the enterprise and the degree to
which its performance is under the manager's control. Simple budgeting may thus be the only cost-
effective approach for small enterprises (including small farms) while large enterprises may make
full use of all the latest quantitative aids to decision making.
Tenth and finally in this overview of the concept of management, the following question must be
asked. Is managerial ability innate and thus dependent on genetics or is it something that can be
developed through education and training? Though the relevant genes have not yet been located, there
seems to be no doubt that people do differ in their innate managerial ability. At the same time, the
plethora of management education programs available today attest to the view that management ability
can be learnt.
Against the above general management background, we can now consider management in the specific
context of agricultural production. To begin, consider the following representative sampling of
twentieth-century definitions of farm management as proposed in relation to commercial family farms.
"[Farm management is concerned with] how can the individual farmer so organise the factors of production - land, labour and
capital - on his farm, so adapt practice to his particular environment, and so dispose of his product, as to yield him the largest net
return, while still maintaining the integrity of his land and equipment" (Butterfield 1910, p.3).
"Farm management, as the subdivision of economics which considers the allocation of limited
resources within the individual farm, is a science of choice and decision making" (Heady and Jensen
1954, p.6).
"Farm management is concerned with the organisation and deployment of the resources put into a farm
business - the land, the capital, the labour and that item of over-riding importance, the ability and
skills of the individual farmer" (Dexter and Barber 1960, p.13).
"[Farm management] is concerned with the organization of resources, with planning their use, both
within and between enterprises, and with the control of plans both during their implementation and
afterwards" (Barnard and Nix 1973, p.18).
"Farm management is concerned with the decisions which affect the profitability of the farm business"
(Castle, Becker and Nelson 1987, p.3).
"Farm management can be thought of as being a decision-making process. It is a continual process ...
The decisions are concerned with allocating the limited resources of land, labor, and capital among
alternative and competing uses. This allocation process forces the manager to identify goals to guide
and direct the decision making" (Kay and Edwards 1994, p.9).
There are clearly both similarities and marginal differences among these definitions. It is interesting
that the earliest definition from 1910 is little different from the most recent of 1994. The common
thread through most of them is decision making about the allocation of resources. Though all the
definitions are by respected authorities, we find them inadequate. To our mind, none of them captures
all the essential elements of farm management. It is also notable that, reflecting their family-farm
orientation, they do not emphasize labour or people any more than other resources. This is in contrast
to the general management definitions cited earlier.
Our preference, if we want to go beyond saying "farm management is what farmers do", is for what -
compared with the above definitions - is a more succinct but far more comprehensive definition. This
is of farm management as the process by which resources and situations are manipulated over time
by the manager of the farm system in trying, with less than full information, to achieve his or her
goals (Dillon 1980). In contrast to the cited definitions, this statement either recognizes or better
emphasizes: first, that farm management is not farm management research, teaching or consulting;
second, the dynamic nature of the farm system and its environment; third, the fact that the farm
manager deals not just with resources but also faces the challenge of situations to be met and
opportunities to be seized; fourth, the active role of manipulation as distinct from the more passive
role of merely organizing and controlling; fifth, the uncertainty and consequent risk present in the farm
manager's decisions, thereby implying attempted (rather than sure) achievement of objectives based
on personal preference and subjective judgement; sixth, by referring to goals rather than profit, the
reality of non-profit goals is recognized; further, in the family-farm context, the nonsense of
endeavouring to differentiate between the farm as a business and as an economic entity is done away
with - the latter subsumes the former and, in our judgement, must be the context in which farm
management operates. Too, not only does this definition better capture the challenge and excitement of
farm management but, with the deletion of the word 'farm', it serves as an excellent definition of
management in general.

2.2 Time Effects
From a management perspective, agriculture is quite distinctive. This distinctiveness primarily

relates to the time-dependent biological nature of agricultural production and to the risks this
necessarily entails. As well, there are other differences related to cost structure and, in the context of
small farms, to the markets faced, their lack of political influence, and the virtually inseparable
relationship between the farm system and the farm household.
Industrial production, being independent of the natural environment, is mechanical. In contrast, the
biological nature of agricultural production causes it to be strongly influenced by the natural
environment. In consequence, agriculture has its own innate rhythms and significant elements of
agricultural production are not under the farmer's control. Species suitability and growth potential,
climate, disease and pest occurrence, yield, livestock freewill and gestation period are a few
examples. Over time, repetition of the same production process carried out in the same fashion on the
same land can give dramatically different results. Nor can production be easily speeded up (or
slowed down) in response to favorable (or unfavorable) prices. Indeed, because plant and animal
biology are not thoroughly understood, they are still the subject of much research. This can benefit
farmers but adds to the burden of the farm manager's role if he or she is to keep abreast of relevant
research results.
2.2 Time Effects
Unlike most other production enterprises, because of its biological nature, agriculture is inextricably
time dependent (Dillon and Anderson 1990, Ch. 6). Compared with industrial production, agricultural
production both takes more time and is generally time-related in terms of the seasons. As well, the
form of agricultural production, i.e., the species to be grown, is generally dictated by climatic
conditions in terms of precipitation, solar radiation etc. Thus, e.g., while grain production requires
some three to six months of time and the correct climate and seasons, factory production of a car
requires less than a day and is independent of such seasonal elements as rainfall, temperature, day
length etc.
From a management perspective, there are four influences of time on agricultural production that
make it distinctive. First, there is what can be called the flexibility effect. By its nature, time gives the
farm manager flexibility in the form of options that would not be available if production were
instantaneous. Time gives the farmer the opportunity to sequence input injections and/or output
harvests in varying ways. Thus the farmer has to decide not only the amount of fertilizer, irrigation
water, livestock feed etc. to be used, but also the pattern of its use over the production period. Too,
within the timespan of any particular activity, decisions about its operation (including marketing of
the product, if relevant) can be revised on a sequential contingency basis relative to market or
climatic conditions. Obviously, such time-induced flexibility has both advantages and disadvantages
for the farm manager. On the one hand it provides the opportunity to correct or improve decisions
over time but, on the other hand, it greatly increases the possible volume and complexity of decisions.
Second, in contrast to industrial production, in many agricultural production activities, time may be a
variable input under the manager's control, i.e., a decision variable to be manipulated by the farmer
on the basis of its opportunity cost. This is particularly the case for livestock-fattening and
aquaculture activities, as well as for some annual crops, and also relative to the replacement decision
for perennial crops and woodlots. Such time dependence of output, of course, always exists in
agriculture. Often, however, as for most grain crops, it can be ignored because the input of time
required, once a particular cultivar has been chosen, is fixed and/or not under the farmer's control.
Third, and again in contrast to industry, time may have an indirect effect on agricultural production
through the influence of carryover effects. These may be positive as, e.g., when there is a carryover
of nutrients from fertilizer application or green manure crops in prior periods. Conversely, they may
be negative as in the case of carryover of the eggs of pests or the spores of disease. The extent of such
carryover is a function of the time elapsed (e.g., a year or two for green manure but up to 30 years in
the case of anthrax spores) and may influence the farmer's current decisions on product choice and
input use.
Fourth, time - in conjunction with the biological nature and climatic dependence of agricultural
production - makes farm management a far more risky endeavour than industrial management. In the
case of small farms, this is compounded both by the close association between the farm and the
household and, if they are market oriented, by their situation as price-takers in both input and output
markets. By its nature, time introduces uncertainty into farmers' decision making. Because of the
complex processes of change, evolution and growth that accompany the flux of time in agriculture and
its market and other environments, farmers cannot be sure of the future either in terms of (i) the
outcomes of today's decisions or (ii) the opportunities (including new products and technologies) that
may become available or (iii) the developments that may occur in their personal, family, social,
political, legal, natural or economic environments. Given the information available to the farm
manager, the best he or she can do is to make those decisions which are expected to best achieve the
desired goals. There can be no guarantee that these risky decisions will, with hindsight, turn out to
have been the best that could have been made. Only perfect foresight can guarantee no ex post regrets
about what might have been.
There are four resource-portfolio effects that distinguish farm from industrial management. First,
fixed costs in farming are often proportionately higher than in industry. This is due to the importance
of land in agricultural production and, on family farms, to the fixed cost of family labour. Because of
the more risky nature of agriculture, it is generally more difficult to cover the fixed costs of
production in farming than in industry. In consequence, more than industry, agriculture exhibits boom
or bust characteristics.
Second, relatively higher investment in fixed resources such as land and improvements also causes
greater asset-fixity effects in agriculture. This fixed investment reduces the opportunity for farm
managers, as entrepreneurs, to move into or out of or within agriculture - as may also be the case for
small farmers in situations where tenure is on some traditional basis and there is no land market.
Third, far more than in industry, farm management has a custodial or guardianship responsibility for
the natural resources under its control. This is particularly so for land and water which play a more
dominant role in agricultural production than in industrial production. Increasingly, for the benefit of
present and future generations, society is demanding that farmers manage their land and water
resources so as to enhance or at least sustain their quality rather than to deplete or pollute them.
Fourth, because of the importance of land as a fixed input in agricultural production, the farm
manager, more so than his or her counterpart in industry, confronts the law of diminishing returns. In
consequence, as product and input prices change (as they do more so in agriculture than in industry),
the farm manager has an increased volume of decisions to make about how much of each variable
input to use and the consequent expected level of output. Concomitantly, the fixity of the land resource
and its high contribution to farm costs undoubtedly make it more difficult for management to achieve
economies of scale in agriculture than in industry.
Management differences between agriculture and industry are exacerbated in the case of small
commercial family farms. First, the managers of small farms suffer from a lack of market power.
They are generally price-takers in both their input and output markets. Often in their input markets
they face administered prices. In output markets they generally face few buyers and have to bear the
burden of product perishability and the tyranny of distance from the market.
Second, the close relationship between the farm system and the family on most small commercial
farms implies that the farm is not purely a business. The farm and its household constitute an
economic entity whose goals are inevitably going to place a heavy emphasis on profit and long-term
commercial viability but will also involve non-profit elements. Some degree of conflict between farm
and family goals is inevitable. The special difficulties for management of working with family
members in a business cannot be overstated.
Third, compared with industry, the factors of production on small family farms are often not
separately supplied: the manager will generally also be a worker, the farm's transport will also be the
family transport, and the family home will be the manager's office. This can lead to difficulties in
allocation - e.g., to what extent should the farmer spend his or her time as manager or as labourer?
Too, on family farms, labour is much less specialized than in industry. The farm operator has to be
able to do everything from managing the farm to repairing its infrastructure, keeping its books and
ploughing the fields.
Fourth, it is a fact of life that the managers of most small businesses, not just farms, have not had the
opportunity of formal education and training to develop their managerial skills and capability. Nor,
unlike managers in large-scale industry, have they been purposefully selected on the basis of their
managerial ability. It is therefore to be expected that the general level of management in small
agricultural enterprises will not be as good as that in larger industrial enterprises. Unless government
ensures otherwise, this problem may be compounded by the fact that small farmers are generally
disadvantaged in terms of access to information and credit. They also usually lack influence in the
setting of agricultural policies and research agendas.

In the context of small farms, the current theory of farm management is best expressed as the
integrated consideration of two complementary theoretical frameworks derived from modem
management theory (Kast and Rosenzweig 1974). The first of these frameworks is farm-system
theory with its conceptualization of the farm as a purposeful system. This theory constitutes a way of
looking at the farm; it provides a checklist of aspects of the farm that should be the concern of
management. The second theoretical framework is the overlay of farm system theory by what might be
generally called the theory of management by objectives. This involves the managerial functions of
planning, organizing and controlling the operation of the farm system over time through the use of
economic and other principles and administrative procedures. Management by objectives
corresponds to the analytical and decision-making activities necessarily undertaken by the farmer in
his or her role as manager.
Management by objectives with the aim of maximizing financial profit subject to relevant constraints
has been the traditional theoretical approach to commercial farm management (Case and Williams
1957; Jensen 1977; Nix 1979). In contrast, farm-system theory has come to formal prominence only
over recent decades. However, the theory of farm management by objectives has always to some
degree had implicit within it the recognition of the farm as a purposeful system.
Both theories could be used as frameworks for the descriptive or positive analysis of a farm's
management and performance. Generally, however, they are couched and used as normative theories,
i.e., with the aim of providing guidance as to what the farm manager should do if system goals are to
be achieved as well as possible. Discussion here is from such a normative perspective.
A system is an organized unitary whole consisting of a set of interrelated elements, components or
subsystems, each of which is related directly or indirectly to every other element, component or
subsystem in the system. The basis of systems theory is that, by their nature, systems can only be
properly considered qua systems.
With the advent of systems thinking in the 1950s, the traditional approaches to understanding
phenomena of 'reductionism' (i.e., reducing phenomena to their basic parts and studying these parts in
isolation) and 'mechanism' (i.e., that phenomena can be fully explained by mechanical cause-effect
relations) have been seen as inadequate for understanding and manipulating systems (Ackoff, 1973).
Rather, the systems approach is based on 'expansionism' or the view that a system is more than the
mere sum of its parts. That is, the system's behaviour is not fully deducible from the behaviour of its
parts considered in isolation. The systems approach also recognizes that, as well as mechanistic
effects, the operation of purposeful systems with their goal-setting and goal-seeking behaviour will be
influenced by teleological (i.e. purpose-driven) effects arising from choice and free-will. System
performance must therefore be judged not simply in terms of how each part works separately, but also
in terms of how the parts fit together and relate to each other, and in terms of how the system relates
to its environment and to other systems in that environment.
Applied to purposive man-made systems such as farms, the systems approach highlights two features
- subjectivity and purpose - that are eschewed by traditional science in its quest for understanding.
First, a farm system involves not merely the interactions of physical forces but also free-will choices
under uncertainty and contests of will arising from the purposiveness of behaviour of animate
elements (i.e., people and livestock) in the system. Farm systems thus exhibit non-determinism and
capriciousness of behaviour that, before the event, can be captured only by subjective judgement and
not by objective fact. Second, the understanding (and management) of a purposive system such as a
farm necessitates a teleological or means-end approach of (i) recognizing the system's goals and (ii)
assessing alternative strategies as to how those goals might best be achieved given whatever may be
the initial state of the system and its constraining conditions. Use of the systems approach to
understand and better manage purposeful systems such as farms, however, is no easy task (Dillon,
1992). Difficulties lie in (a) the free-will and non-deterministic attributes of the farm system; (b) the
complexity of purpose and operation that the farm system may possess; and (c) the uncertain dynamic
character of the farm system and its environment.
The Farm System
Farm-system theory views a farm as a unique goal-setting (i.e., purposeful) open stochastic
dynamic artificial (i.e., man-made) system having a major aim of generating income (in cash or kind)
for its stakeholders through agricultural production (Dillon 1992). Generally, a farm system is
complex rather than simple. Its purposefulness is ensured by its human and social involvement which
enables the system to vary its goals under a given environment. The uniqueness of any particular farm
system is guaranteed by its location, history, resources and human elements. In consequence, while
the general principles of farm-system management are appropriate for any particular farm, their
application must generally lead to different decisions on each particular farm. The openness of the
farm system is obvious from its interaction with its environment. The stochastic nature of the farm
system is guaranteed both by the free-choice capacity of its human (and, if present, animal) elements
and by the stochastic nature of the environment with which it interacts. Necessarily, a farm system is
also dynamic by virtue of its purposefulness, openness and stochasticity which ensure that the system
changes over time. Too, any farm system is a mixture of abstract and concrete elements or
subsystems. The concrete elements are associated with the physical activities and processes that
occur on the farm. The abstract elements relate to the managerial and social aspects of the farm.
Farm-system theory sees the farm as having a variety of boundaries or interfaces, i.e., contact points
with other systems or environments. Clearly, the physical farm boundary is relevant to the system's
land management and is easily specified. Other relevant boundaries will be specified by farm-system
interfaces with input suppliers, credit agencies, the local community, government agencies etc.
Intra-system resources available to the farm system may be categorized as either physical or non-
physical. By their nature, physical resources (such as land, water, buildings, machinery, recreation
facilities etc.) are easily specified. Non-physical resources are more difficult to list. They include the
knowledge and skills possessed by management, labour and the farm system's social organization. As
well, the set of social relationships and the formal and informal organizational structures in which
these exist constitute a non-physical resource, as do such attributes as the farm's equity ratio, credit
worthiness and the like.
Subsystems of the Farm System
From the perspective of farm-system theory, any farm system - being a purposeful man-made
organization - is seen as involving five major subsystems (Dillon 1992; Kast and Rosenzweig 1974,
pp. 111-3). These are:
(1) The technical subsystem whereby resources, technology, knowledge and opportunity are used to produce agricultural
products. The technical subsystem will itself involve a number of production subsystems such as, e.g., the cropping system or the
wheat system or the irrigation system. The precise nature of these subsystems on any particular farm will be determined by
geographic, social, economic and technical constraints. Their nature constitutes the subject matter of most farm management
handbooks and manuals. Historically, their interrelationships as subsystems of the farm system have not been adequately
recognized. This led to the Post World War II call for farm management analysis to be on a 'whole-farm basis' which, though an
improvement, is usually still inadequate in failing to appreciate sufficiently the farm's non-technical subsystems.
(2) The organizational structural subsystem which corresponds to the formal structure of authority,
communication, job descriptions, responsibilities and task allocation within the farm system. On
small farms this structural subsystem will not be elaborate.
(3) The informal structural subsystem which exists in any farm system involving two or more
persons. The greater the number of persons involved in the farm's social organization (the farm
family, workers and their families, neighbours, school teachers etc.), the more complex the informal
structural subsystem will be. It is abstract and consists of individuals' motivations and behaviour,
informal relationships, feelings of status, power and influence. It also reflects the mixture and
interaction of personalities, attitudes, expectations and aspirations of people in the farm system.
Regardless of how the formal organizational structure is defined, psychological and social interaction
between individuals and groups in the farm system will occur outside the formal structure. Even on
small family farms, the 'grape vine' is pervasive - though less so than in large farm systems.
(4) The goals and values subsystem which relates to those goals and values held by the farm system
as a purposeful organization. In general, the goals and values subsystem will not correspond exactly
to the overall set of (often conflicting) personal goals and values held by individuals in the farm
system. The farm's goals and values subsystem will, however, be influenced by the goals and values
of individual members since significant disharmony would be destructive to the farm system. The
goals and values subsystem will also be influenced by the general sociocultural and political
environment which determines the role individual farm systems are expected to play in society and
the requirements they must meet in order to be allowed to function.
(5) The managerial subsystem which relates to the entire farm system through the farm manager's
activity of setting goals, developing long- and short-term plans, specifying organizational structure,
deciding on enterprises, choosing technology, allocating resources, seizing opportunities, establishing
control processes, harmonizing relations between all the subsystems and with the various
environmental suprasystems relevant to the farm etc. - in short, managing the farm system as a
purposeful entity. Like the informal structural subsystem and the goals and values subsystem, the
managerial subsystem is abstract. It is not concrete in the same way that we can specify the formal
organizational structure and write job specifications, or write a manual of technology and a calendar
of operations corresponding to the technical subsystem. Components of the managerial subsystem will
generally be the persons involved in management (perhaps only one for a small farm system), their
knowledge, experience, judgements and preferences, and the information systems used in exercising
management.
A generalized schematic representation of the farm system in terms of the above five subsystems is
shown in Figure A.1. The central role of management is clearly emphasized. On large farms all five
subsystems would be significant. On small farms involving few people, the organizational structure
and informal social subsystems are unlikely to be very important. Nonetheless, all five subsystems
are relevant regardless of farm size; any appraisal of a farm system by its manager or by outsiders
should pay due regard to these five subsystems.
Integrating Processes
The five subsystems - technical, structural, informal, goals and values, and managerial - making up
the farm system may be viewed as constituting a differentiation of the overall system. They are the
building blocks. The cement which binds these subsystems together or facilitates their interaction and
interdependence to form a purposeful whole is a set of integrating variables or processes. These are
the role process, leadership, decision making, information flow and control mechanisms.
FIGURE A.1 - Generalized Schematic Representation of the Farm System
For the farm system to function effectively, its management must pay close attention to these
integrating processes. It must provide leadership, carry out effective decision making, ensure that
there is a flow of requisite information both for management purposes and to keep others informed,
and implement control mechanisms to guide and direct performance. Management must also generally
ensure an intra-system environment such that the role performance (i.e., expected behaviour) of
persons in the farm system assists rather than hinders system performance.
The Farm's Environmental Suprasystem
The environmental suprasystem is everything external to the individual farm system. The influence of
the suprasystem environment on the farm system is much greater than vice versa. One more or less
farm system would make virtually no difference to the environment - but it could be profoundly
important to those involved with the particular farm system.
The farm system's environmental suprasystem involves a number of significant suprasystems of a
general nature (Dillon 1992). These are the cultural, technological, educational, political, legal,
demographic, sociological, climatic and economic suprasystems. The state, dynamics, direction of
change and interrelations of these environmental subsystems are significant for the individual farm
system as a general influence of varying import over time.
The historical background, ideologies, values and norms of the enveloping society constitute the
cultural suprasystem. This influences views on leadership patterns, authority, rationalism and
interpersonal relations within the farm system.
The technological suprasystem sets the level of scientific and technological development potentially
available to the farm system. Closely related and complementary is the educational suprasystem
which determines the general level of literacy and numeracy in the environment, and the availability
of specialized services and skills.
The general political climate and organization of the enveloping society in terms of political parties,
concentration of power and system of government reflects the political suprasystem. Over time, the
individual farm system needs to stay in harmony with the political system.
The legal suprasystem is specified by constitutional considerations, the nature of the legal system, the
jurisdiction of government units and specific laws relating to commercial organization, land use,
technology, taxation, labour, health, natural resources, pollution etc. Such laws and regulations, as
well as more diffuse legal influences, constrain the operation of every farm system.
The demographic suprasystem determines the flow of human resources available to the society in
terms of number, regional distribution, age and sex. It also influences the pattern of market demand.
The class structure, social mobility, definition of social roles and mores, and nature of social
organization specify the sociological suprasystem. These sociological influences from the
environment will broadly influence farm system goals and organization, and will determine the social
setting within which the farm system operates.
For nearly all farm systems, the climatic suprasystem is very important through its close physical
interaction with the farm's technical system. At the macro level, global climatic effects can also be
significant through their effect on markets and world trade. Significant uncertainty exists about the
occurrence, magnitude and impact of global climate change.
Lastly and often most important as a general influence, the economic suprasystem sets the economic
framework both macro (capitalist or socialist, open or regulated etc.) and micro (taxation, commodity
policies etc.) enveloping the farm system.
As well as the above general environmental influences, any particular farm system may face other
significant environmental influences that are specific to it and must be allowed for by management.
Farm-system Goals
Goals of the farm system can be seen as arising from three sources. These are: (i) goals of the farm
system itself as a purposive entity (e.g., family sustenance or profit maximization); (ii) social goals
(such as resource conservation) imposed or wished on the farm system by the society in which it
functions; and (iii) personal goals (such as an outdoor life) held by participants in the farm system. A
major task of the farm's management is to harmonize these different goals so as to best achieve farm-
system goals. Theoretically, the goals chosen are seen as reflecting need or aspiration circumscribed
by values or normative views of what is good or desirable. These values will be a composite of
those held by farm-system participants and by the members of its enveloping society (Dillon 1992;
Kast and Rosenzweig 1974, pp. 155-6).
As a practical matter, it is reasonable to hypothesize or assume that small commercial farms have a
number of major or strategic goals. These are likely to include survival, being as profitable as
possible, growth of the farm business and conservation of the farm's natural resource base.
Obviously, such overarching goals are interdependent. They overlap to some degree and have
tradeoff possibilities. Survival, profit and growth are interlinked, with the generation of sufficient
profit being a necessary requirement for survival and growth. Conservation of the natural resource
base, like the need to meet health, labour, safety and other relevant legal (and sociocultural)
requirements and norms, is best viewed as a constraint. Hence there is logic in the traditional
orientation of farm management theory to profit maximization under relevant resource and other
constraints. Such an approach also has the advantage of indicating, in terms of forgone profit, the
opportunity cost of pursuing non-profit goals. It has the disadvantage, however, of not allowing for
the risk inherent in the stochastic nature of the farm system.
To allow for risk considerations, probably the best approach is to view the farm manager as
possessing a multiattribute expected utility function specified in terms of the selected strategic goals,
i.e., a function specifying the amount of satisfaction obtainable from the farm system through any
combination of goal achievements. Being rational, the manager is assumed to attempt to maximize this
utility function, subject to the constraints faced, on the basis of his or her personal judgements and
preferences about the risks being faced (Anderson, Dillon and Hardaker 1977). This is probably the
best technical description of what farm managers actually attempt to do - not in a formal quantitative
way but introspectively by means of inductive reasoning from past experience and intuition on the
basis of whatever information is available to them. We see this as occurring in much the same way
that a football player, without any knowledge of physics or differential equations, knows intuitively
how to play the ball. Of course, to the extent that small commercial farms are also family farms, such
judgements are further complicated by family goals and interactions between the farm and its
household.
Both in theory and in practice the nitty-gritty of management of the farm system is provided through
management by objectives. This is done via the application of the sequential management functions
of (as most simply categorized) planning, organizing and controlling (Barnard and Nix 1973) across
the management fields of production, marketing, finance and workforce. Being aimed at the optimal
achievement of farm-system goals, the functions of planning, organizing and controlling have a
normative orientation. Farm-system goals will be set by the owner(s) or stakeholder(s) of the system.
For small farms this will be the farm family or its head, who will often be the farm manager. The
actual strategic goals likely to be chosen - survival, profit, growth and resource conservation - have
been discussed above. Such strategic goals lie at the top of a means-end schema in which the
achievement of lower-level objectives or sub-goals is an end serving as a means to the attainment of
higher-level objectives. Hence, management by objectives relates not just to final objectives or
strategic goals but to the whole means-end hierarchy of objectives or goals. Necessarily, therefore,
the never-ending cyclical sequential process of planning, organizing and controlling is, in theory,
applicable to all levels of the hierarchy of objectives. However, the extent to which this should be so
is a decision depending on its benefit-cost ratio at each level. Logic implies that the more important
(or the higher) the objective, the more attention management should pay to its planning, organization
and control.
The applicability of management by objectives to various levels of the farm system's goal hierarchy
necessarily implies its applicability across different time spans, e.g., to short-, medium- and long-
term objectives. The longer the term, however, the greater the uncertainty, the greater the need for
management by objectives and the more difficult its successful application.
Planning
Over time, but conveniently year by year, the farm system manager has to decide on his or her
answers to three basic questions: (1) What to produce? (2) How to produce? (3) How much to
produce? The answers to these questions constitute the farm's annual production plan. This will
generally be complemented by other plans, relating, e.g., to farm or personal development. Ideally,
the current annual production plan should constitute the first year of a rolling or annually revised
multi-year farm plan extending into the future to the point where the net gain from further planning is
zero.
Planning decisions about what, how and how much to produce, just like any other non-lumpy
decisions, should be based on what we call the basic rule of production economics. This is that any
activity should be undertaken up to and no further than the point where the marginal benefit (i.e., the
benefit from undertaking one more unit) of the activity is just balanced by its marginal opportunity
cost. In theory, in a world with no uncertainty and with full information about all relevant production
relationships, prices and discount rates, optimal annual farm plans could be easily developed (given
adequate computer capacity) using the principles of production economics in a multi-year
mathematical programming framework specified to accommodate the farm's relevant goals and
constraints (Rae 1994). These constraints, theoretically, could include not only resource constraints
but also technical constraints to ensure that the farm plan meets, if the manager wishes, such criteria
as income stability over time, diversity in production, flexibility in product disposal, intra-year
dispersion of income, resource sustainability and environmental compatibility.
When uncertainty is present, as is always the case for real-world farm planning, the situation becomes
exponentially more complex. Nonetheless, in theory, analytical planning is still feasible so long as
information is available about the farm manager's (i) utility function and thus his or her risk
preferences, (ii) subjective judgements about the probability distributions of the relevant uncertain
outcomes and (iii) time-preference discount rate. With uncertainty present, choice of plan is a matter
of subjective judgement. Rather than choice between a variety of options with sure outcomes, the farm
manager faces a set of probability distributions for outcomes corresponding one-to-one to the
available set of alternative plans. These probability distributions are specified subjectively by the
farm manager, i.e., they express his or her personal degree of belief in the chance of various outcomes
occurring. Choice should then be made, we believe, on the basis of what is known as subjective
expected utility theory (Hardaker, Huirne and Anderson 1977).
The presence of time and its corollary, uncertainty, also adds another dimension to farm planning.
This is that annual plans may also be revised within years. As well, a contingency approach to
planning is possible. This implies that not just one but a number of plans may be specified
encompassing what is to be done if various contingencies (such as drought) occur during the planning
period.
In theory, data used in planning should relate to the period for which plans are being made. Since this
is in the future, such data is generally uncertain or taken as unknown. Instead, data from the past, e.g.,
from farm records of the recent past, are often used. This is seen as providing a link between past
control (i.e., monitoring and recording) and future planning. However, naive use of such past data
implies the unlikely assumption that the future will be like the past.
To sum up, at both the level of the farm system and of its technical subsystems, planning and decision
making for modern commercial farms have been the subject of much theoretical analysis in recent
decades. Stimulated by developments in management science, operations research and computer
technology, an array of mathematical programming, simulation and decision theoretic approaches
have been proposed and investigated. Their real-world value is discussed in Section 4 below.
Organizing
Compared with planning, organizing is an administrative rather than an analytical or decision-making
process. The aim of organizing is to ensure that the farm system's plan is implemented. The plan
specifies what, how and how much is expected to be produced. Ensuring that this is carried out
requires that necessary tasks are assigned and coordinated, that necessary inputs and arrangements are
organized on time, and that appropriate authority is delegated. Two comments can be made about
organizing in the context of small commercial farm systems. First, the managers of such systems vary
greatly in their ability to organize. Second, the capacity to organize is enhanced by having a written
farm plan complemented by a technical manual covering the calendar of operations for the products to
be produced.
Controlling
In the theory of management by objectives, control is seen as a process of monitoring plan
implementation and, as necessary, making adjustments to the plan or its implementation. The
information, both experience and data, gathered through control in prior production cycles is seen as
contributing to planning for future cycles.
Control may be exercised in more or less elaborate ways. It entails the collection of relevant
feedback information, the analysis of such data and, as need be, the taking of corrective action
(Barnard and Nix 1973). The collection of control data, i.e., the keeping of farm records, has long
been advocated by farm management experts and advisers. Many farm recording systems have been
promulgated and, these days, software systems are available. These may be more or less detailed but,
at minimum, should include major physical and financial information. Analysis of such data is based
on its concordance with or deviation from previous performance, expected performance or proposed
standard performance. In the 1950s and 1960s, but much less so today, comparative analysis of farm
records across groups of similar farms was popular as a guide to management. While helpful to the
diagnosis of problems, it is now generally recognized that comparative analysis should be used with
caution. Sometimes, deficiencies in plan implementation can be corrected as they are detected. More
often, given the farm system's time-dependent nature, adjustments will need to be made to the plan and
such corrective action incorporated in future plans.

The use made either directly or indirectly by small farmers of farm management theory is an
empirical question whose answer is sure to change somewhat over time. The question is also likely to
be complicated by the fact that small-farm systems differ greatly between countries in their historical
backgrounds, constraints and environmental suprasystems, and that farm managers differ in their
education and training.
The view of the farm as a purposeful open stochastic dynamic system with its various subsystems and
suprasystems is a conceptual device. Beyond emphasizing that the farm system should be recognized,
understood and managed as a purpose-driven system rather than as just a set of disjointed parts, the
systems approach in itself provides no analytical tools aimed at assisting farm management. These
come from the application of management by objectives to achieve system goals. Recognizing this,
we believe there is no significant gap between farm-system theory and its practice by farmers. In our
judgement, farmers generally have a very keen awareness of their farm as an open stochastic
purposeful dynamic system, of its various subsystems and of its environmental suprasystems. Farmers
may not have the jargon of systems theory but they do have its insights - and had them long before
farm-system theorists.
Ever since the advent of scientific concern for farm management early in this century, there has been a
large gap between the theory of farm management in terms of management by objectives as espoused
by 'experts' (i.e., academics, government agents and other advisory professionals) and its practice by
farmers. Today, probably only on a minority of large corporate farms in developed countries is
anything approaching full use being made of the array of analytical (including simulation) procedures
developed by farm management theorists. Across the spectrum of small farms, the use of even simple
analytical farm management techniques is minimal. Despite the long, sustained and widespread
advocacy of farm record systems by experts in many countries, few small farmers have used them in a
thoroughgoing way for purposes of planning and control. This is not to say that small farmers do not
carry out planning and control. Rather, they exercise these functions in informal rather than formal
ways. Likewise, so far as we are aware, none of the commercial attempts that have been made to
provide small farmers with whole-farm planning services based on mathematical (usually linear)
programming have been successful - all have eventually had to close down for want of demand. Nor
have attempts to persuade farmers to use even the simplest of formal procedures (such as decision
trees and certainty equivalents) to handle risky decisions been successful. However, some simple
concepts from farm management theory have proved popular with a number of small commercial
farmers. The most important examples are gross margins budgeting and sensitivity analysis for
planning. In part, this acceptance has probably been the result of banks and other formal sources of
credit requiring budgeted plans to justify the granting of loans.
The dichotomy between theory and practice is apparent in farm management texts. Those aimed at an
academic audience, e.g., Rae (1994), are heavily analytical and full of economic theory. Those aimed
at a farmer audience, e.g., Giles and Stansfield (1995), are far more practical and eschew any formal
presentation of economic theory. Texts such as Makeham and Malcolm (1993) and Barnard and Nix
(1973) which provide a reasonable bridge between farm management theory (especially production
economics) and practice are rare.
Why is there such a large gap between farm management theory and practice? The temptation for
theorists is to say farmers are ignorant and irrational. This temptation must be rejected. We know of
no evidence to suggest that, within their frame of reference, farmers are irrational. Nor that they are
so ignorant as to reject gainful opportunities. A more substantive explanation of the gap is possibly
that the suggested techniques of planning and decision making are too difficult and costly for farmers
to apply. Too much data is needed and the data is too difficult and time-consuming to collect so that
the costs of analysis exceed the benefits.
Another possibility is that the analytical products of farm management theory have not been promoted
sufficiently well. Given the efforts of government agencies around the world, this appears to be most
unlikely. There is also another possibility. This is that the majority of small commercial farmers may
judge there to be so much uncertainty in their environment that they cannot exercise sufficient control
to warrant the application of sophisticated analytical aids.
Of course, not all small farmers have rejected proposed aids to their managerial decision making.
Indeed, with modernization, pressures are such as to enhance the adoption of such aids. Worldwide,
the business of farming is coming under pressure from the influence of deregulation, open markets and
globalization combined with secular decline in agriculture's terms of trade. Significant changes
affecting the agricultural system have occurred in most countries in recent years. These changes
include: a paradigm shift from production-driven to market-driven agriculture; less government
interference; diminishing international trade barriers and increasing international competition
combined with rapid transmission of world market volatility; increasing concentration of food
processing among transnational and multinational companies requiring consistent quality product;
shorter market chains and increased vertical integration; consumer demand for a wide product choice
with year-round availability, consistency, convenience and safety; rapid changes in technology,
particularly in plant and animal genetics, precision farming, communication and information
management; greater integration of chemical, seed and machinery suppliers, and greater emphasis on
patents and property rights; increased concern for biodiversity, the environment and personal health;
ongoing reduction in the number of farms and a steady increase in farm size; a greater involvement of
wives in family farming; and a decline in the 'image' of agriculture.
These changes, we believe, will increase the pressure on small farmers to seek greater efficiencies
and be more competitive, as will external pressure from their credit sources. In seeking greater
efficiency, we do not expect small-farm managers to adopt sophisticated aids. Rather, we would see
them as: (1) becoming knowledgeable about their enterprise gross margins; (2) making greater use of
data and information from off-farm sources; and (3) keeping minimal farm records (or having access
to minimal standard data) sufficient to develop a basic annual farm plan using gross margins. We also
see this broader adoption of the use of gross margins and simple farm planning as being catalysed
over time by the use of personal computers and spreadsheet analysis together with the greater
involvement of wives in the office of the family farm. Only now is it being recognized that farm
women constitute a significant potential source of human capital in agricultural development.
These days, personal computers are readily available. Many farm management theorists seem to be
mesmerized by their possible potential and to see them as a catalytic vehicle for the adoption by
small farmers of sophisticated aids to management. Much specialist software has already been
developed in more developed countries for particular purposes such as least-cost feed mixing, pest
management, irrigation scheduling, livestock genetic evaluation, satellite-guided robot ploughing etc.
Such software is available for use on-farm or through specialist service agencies. In addition, a great
array of more general software is available commercially for on-farm financial recording and
analysis and also for risk analysis. This is a good thing. However, for the reasons already given, it
cannot be expected that the computer revolution will lead to rapid general adoption by small farmers
of management aids beyond budgeting and planning based on spreadsheet analysis using gross
margins and sensitivity analysis.

There is no magic wand that can be waved to make the general level of management on small farms
more effective overnight. Such improvement should be realistically viewed as an ongoing long-term
task involving incremental gains over time. These gains will accrue as the constraints on managerial
performance are lessened. Nor in an unregulated open economic system should it be expected that
small farms can escape the pressures of volatility in international commodity markets, ups and downs
in income, secular decline in agriculture's terms of trade, and the need to be internationally
competitive through ongoing technological improvement and the capturing of size economies. Today,
there is much more to the small commercial farm than the nineteenth century description - given by
S.C. Warren in his discussion of a paper by Schultz (1939) - as:
'A little land well tilled,
a little barn well filled
and a little wife well willed.'
Solution of the problem of small farm viability lies mainly in small farmers' hands. Those who do not
take the right steps to remain competitive, or who suffer bad luck, can expect to lose out under a
process of Darwinian selection. Government, however, through its influence on the environment of
small-farm systems, can have a significant indirect role in assisting small farmers to be effective and
successful managers. This indirect role is over and above any safety-net role that government may
play through social welfare policies aimed at assisting non-viable small farmers to leave agriculture.
Beyond ensuring that policies do not reflect urban bias, there are four broad opportunities open to
government to help enhance the quality of small-farm management. First, small-farm managers can be
helped by the provision of adequate public infrastructure of both a physical and institutional nature.
Public roads and other facilities such as ports which facilitate market access are probably the most
important physical infrastructure. Institutionally, the need is for government to ensure a framework of
arrangements that is not detrimental to the provision of communication, credit, electricity, transport,
storage, market, research and other services to meet the needs of small-farm systems. These service
activities may involve a mix of private and public sponsorship with the latter based, as need be, on
the user-pays principle. Research, e.g., might be conducted by both public and private agencies with
public-sector research funded in part by output-based levies on producers. The government should
also ensure that, except to meet desired social and other objectives, institutional arrangements are not
such as to hinder operation of the land market.
Second, government needs to ensure that it has liaison linkages, preferably of a formal nature, with
small farmers through their organizations. Without such liaison linkages, small farmers will continue
to be unheard or ignored because of their lack of influence relative to large producers and other
sectors of the economy. This is particularly important in the area of publicly funded research.
Historically, the evidence is that unless affirmative action is taken to seek out and accommodate the
research needs of small farmers, agricultural research agendas are mainly set with an orientation to
the wishes of large farmers. This applies just as much to economic and marketing research as to
physical crop and livestock research.
Third, government should ensure the provision of appropriate and adequate educational
opportunities and facilities to meet the requirements of small-farm systems. This need for human
resource development has a number of dimensions, all of which are of a long-term rather than short-
term nature. A primary need is for basic literacy and numeracy in the farm-system community.
Managers, workers and others associated with small farms in this modem age need to be literate and
numerate if they are to make use of modern technology, interpret market signals and remain
competitive. Vocational training opportunities for existing and potential managers of small-farm
systems are also very important. These opportunities may take a variety of forms from short-course
arrangements of a few weeks to live-in college programs offering one-, two- or three-year courses.
Alternatively, they may involve correspondence programs extending over two or three years with
occasional one- or two-week residential schools. These courses could be based on application of the
management functions of planning, organizing and controlling across the fields of production,
marketing, finance and workforce with emphasis on gross margins, budgeting and sensitivity analysis
as tools of production planning. Whatever their particular form, such farm management-oriented
programs might be offered by tertiary institutions as an adjunct to their degree programs or through the
Ministry of Agriculture (perhaps as part of extension activities). Lastly in the educational arena, the
government needs to ensure that there is a sufficient output of agricultural graduates to service the
demands of small-farm agriculture for management information and professional advice whether
through research, extension, private consulting or other activities such as policy advice to
government.
Fourth, the general area of information is one in which government can do much to assist the
managers of small farms. Traditionally, significant information transfer in agriculture has occurred in
many countries through government-sponsored extension services linked to public research agencies.
Increasingly, such extension services have moved towards including a management as well as a
technical focus. There has also been a realization that relevant and timely information has increasing
value as the business of agriculture becomes more complex and volatile. The more situation-specific
and time-specific the information, the greater its competitive value. Conversely, generalized public
information is of decreasing value. Nonetheless, just as can be argued for agricultural research, a
case may be argued that a government extension service to small farmers is in the national interest.
Such extension should not ignore management information. It could usefully include annual tabulations
of enterprise gross-margin budgets on a regional basis, information on expected market needs and
outlook on a commodity basis, information on risk-spreading and risk-transfer strategies for handling
price risk, meteorological (including probability and forecast) information on rainfall and frost
occurrence, and both technical and economic (including risk) information on new technologies and
products.
6. REFERENCES
Ackoff, R.L. (1973). 'Science in the Systems Age: Beyond IE, OR and MS', Operations Research
21: 661-71.
Anderson, J.R., J.L. Dillon and J.B. Hardaker (1977). Agricultural Decision Analysis, Iowa State
University Press, Ames.
Anon. (1913). Roman Farm Management: The Treatises of Cato and Varro, MacMillan Co., New
York.
Barnard, C.S. and J.S. Nix (1973). Farm Planning and Control, Cambridge University Press.
Bradley, R. (1725). A Survey of the Ancient Husbandry and Gardening collected from Cato, Varro,
Columella, Virgil, and Others the Most Eminent Writers among the Greeks and Romans. Motte,
London.
Butterfield, K.L. (1910). 'Analysis of the Rural Problem' in Report of the First Annual Meeting,
American Farm Management Association, Ames.
Case, H.C.M. and D.B. Williams (1957). Fifty Years of Farm Management, University of Illinois
Press, Urbana.
Castle, E.N., M.H. Becker and A.G. Nelson (1987). Farm Business Management, 3rd edn,
Macmillan, New York.
Dexter, K. and D. Barber (1960). Farming for Profits, Penguin, Harmondsworth.
Dillon, J.L. (1980). 'The Definition of Farm Management', Journal of Agricultural Economics
31:257-8.
Dillon, J.L. (1992). The Farm as a Purposeful System, Revd edn. Miscellaneous Publication No. 10,
Department of Agricultural and Resource Economics, University of New England, Armidale.
Dillon, J.L. and J.R. Anderson (1990). Analysis of Response in Crop and Livestock Production,
3rd edn, Pergamon, Oxford.
Evelyn-White, H.G. (1914). Hesiod: The Homeric Hymns and Homerica, Heinemann, London.
George, C.S. (1968). History of Management Thought, Prentice-Hall, Englewood Cliffs.
Giles, T. and M. Stansfield (1995). The Farmer as Manager, 2nd edn, CAB International,
Wallingford.
Hardaker, J.B., R.B.M. Huirne and J.R. Anderson (1997). Coping with Risks in Agriculture, CAB
Heady, E.O. and H.R. Jensen (1954). Farm Management Economics, Prentice-Hall, Englewood
Cliffs.
Hellriegel, D. and J.W. Slocum (1986). Management, 4th edn, Addison-Wesley, Reading.
Hitt, M.A., R.D. Middlemist and R.L. Mathis (1989). Management: Concepts and Effective
Practice, 3rd edn, West Publishing Co., Saint Paul.
Jensen, H.R. (1977). 'Farm Management and Production Economics, 1946-70', A Survey of
Agricultural Economics Literature, University of Minnesota Press, Minneapolis, Vol. 1, pp. 1-89.
Kast, R.E. and J.E. Rosenzweig (eds) (1974). Organization and Management: A Systems
Approach, 2nd edn, McGraw-Hill Kogakusha, Tokyo.
Kay, R.D. and W.M. Edwards (1994). Farm Management, 3rd edn, McGraw-Hill, New York.
Koontz, H. (1969). 'A Model for Analyzing the Universality and Transferability of Management',
Academy of Management Journal 12:415-30.
Koontz, H. (1976). 'The Management Theory Jungle', in H. Koontz and C. O'Donnell (eds),
Management: A Book of Readings, McGraw-Hill, New York, Ch. 3.
Kreitner, R. (1989). Management, 4th edn, Houghton Mifflin, Boston.
Makeham, J.P. and L.R. Malcolm (1993). The Farming Game Now, Cambridge University Press.
Mintzberg, H. (1983). Structure in Fives: Designing Effective Organizations, Prentice-Hall,
Englewood Cliffs.
Nix, J. (1979). 'Farm Management: The State of the An (or Science)', Journal of Agricultural
Economics 30: 277-91.
Schultz, T.W. (1939). 'Theory of the Firm and Farm Management Research', Journal of Farm
Economics 21: 570-86.
Taylor, F.W. (1911). Principles of Scientific Management, Harper & Bros., New York.
Toffler, A. (1980). The Third Wave, Bantam, New York.
FAO FARM SYSTEMS MANAGEMENT
SERIES
1. Guidelines for the conduct of a training course in farming systems development, 1990 (E F S)
2. Risk analysis in dryland farming systems, 1992 (E)
3. The forest-garden farms of Kandy, Sri Lanka, 1992 (E)
4. Changes and development in Solo Valley farming systems, Indonesia, 1993 (E)
5. Institutionalization of a farming systems approach to development, 1992 (E)
6. Farm management research for small farmer development, 1993 (E F S)
7. Farming systems development and soil conservation, 1994 (E)
8. Farm and community information use for agricultural programmes and policies, 1994 (E)
9. Methods of policy analysis for agricultural programmes and policies - A guideline for policy
analysts, 1994 (E)
10. The farming systems approach to development and appropriate technology generation, 1995
(E)
11. L'évolution des systèmes de production agropastorale par rapport au développement rural
durable dans les pays d'Afrique soudano-sahélienne, 1996 (F)
12. El desarrollo de los sistemas de agricultura campesina en América Latina - Un análisis de la
influencia del contexto socioeconómico, 1996 (S)
13. Farm management for Asia: a systems approach, 1997 (E)
The FAO Technical Papers are available through the authorized FAO Sales Agents or directly from
Sales and Marketing Group, FAO, Viale delle Terme di Caracalla, 00100 Rome, Italy
E - English
F - French
S - Spanish
***
This publication breaks new ground in synthesizing orthodox farm management approaches and more
recent systems thinking in the context of analysing and planning the improvement of small family
farms. Starting with a new schema for classifying farm types and agricultural systems, it demonstrates
the application of a range of quantitative analytical methods for understanding and working within the
complex world of traditional small farm systems. These techniques are illustrated with practical
examples taken from various parts of Asia, but the approach and the methodology will be of interest
to all concerned with small farm development anywhere in the world.

Farm Management For Asia System

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Farm Management For Asia System

Uploaded by

Copyright:

Available Formats

Farm Management for Asia: a Systems

Approach. (FAO Farm Systems Management

ACKNOWLEDGEMENTS AND DEDICATION

An a future amount to be received or paid in n years' time; it is equivalent to an annuity

AB: allocation budgeting

ac ft: acre feet (of irrigation water)

AME: adult male equivalent, a standardized unit of labour measurement

APP: average physical product, equal to Y/X for Y = f(X)

AVP: average value product, equal to (APP)py

b: the upper-range or maximum value of a random variable

B: size of the available cash budget

B/C: ratio of benefits to costs

B-1: in AB and SP, the initial set of resource constraints

C: depending on context, cost or the coconut-to-copra conversion rate

CDF: cumulative distribution function

CE: a decision maker's certainty equivalent for a risky prospect

CV: coefficient of variation, calculated as 100 to express the standard deviation of a

CY: crop-cycle year of a perennial crop

d: the constant annual rate of depreciation under the declining-balance method

D: the annual depreciation cost as measured by the straight-line method

dU/dX: marginal utility of X for U = f(X)

dY/dX: marginal product of X in Y = f(X) or the first derivative of Y with respect to X

EMV: the expected money value of a risky prospect

EU: subjective expected utility of a risky prospect

EV: expected value of a risky prospect

FC: fixed cost

GAMS: General Algebraic Modelling System

GM: gross margin, calculated as gross return less variable cost

GR: gross return or gross revenue

HYV: high-yielding variety

i*: nominal annual rate of interest or rate of discount

K: in LP, the maximum level of an activity permitted by a particular constraint

L: the expected total years of useful life of a capital item

LP: linear programming

LU: labour unit, a standardized unit for measuring labour

m: the modal value of a random variable

-M: in LP, the notional high negative p value placed on Qi

mds: maunds, a volumetric measure used in Pakistan and India

MOTAD: Minimization of Total Absolute Deviations, a form of linear risk programming

MPP: marginal physical product, equal to dY/dX for Y = f(X)

mt: metric tonne(s)

MVP: marginal value product, equal to (MPP)py

n: depending on context, the number of observations on a variable, or the size of a

NPK: fertilizer containing nitrogen, phosphorous and potassium

NR: net return or net revenue, calculated as TGR minus TC

Nu: Ngultrum; Bhutanese unit of currency

OC: overhead costs, defined as fixed costs accrued on a whole-farm basis

Oi: the i-th possible outcome of a particular risky decision

pg : unit price of grain

pi: unit price of the i-th good

px : unit price of input factor X

py : unit price of product Y

P: quantity of phosphate fertilizer

pk: patok(s), a Javanese land unit equal to about 0.1 ha

q: number of times per year that compounding or discounting is to occur

R: depending on context, rainfall, or net return, or an annual payment amount that

Ri: income from the i-th activity

R*: any nominated value of net return R

RN: random number

Rp: Rupiah; Indonesian unit of currency

Rs: Rupee; Indian, Pakistani and Sri Lankan unit of currency

RTC: index of relative time-concentration of annual production or income of a product,

RTD: index of relative time-dispersion of production or income, calculated as 1 - RTC