SOKOINE UNIVERSITY OF AGRICULTURE

DEPARTMENT OF AGRICULTURAL ECONOMICS AND AGRIBUSINESS

PRODUCTION ECONOMICS
(AEA 201 AND HE 203)
(Second Year)

By

A.C. Isinika and E. M.M.Senkondo

1
 SOKOINE UNIVERSITY OF AGRICULTURE

DEPARTMENT OF AGRICULTURAL ECONOMICS AND AGRIBUSINESS

PRODUCTION ECONOMICS
(AEA 201 and HE 203)
(Second Year)

By

A. C. Isinika and E. M. M. Senkondo

2
1. INTRODUCTION

1.1 What is Economics?

Economics is a social science. It is defined as the study of how limited resources can best be
used to fulfill unlimited human wants. Whereas the wants and desires of human beings are
unlimited, the means or resources available for meeting these wants and desires are limited.
Economics is concerned with understanding the principles and developing rules that govern the
production and consumption of goods and services in an economy in order to make the best use
of available resources. Specifically, the study of economics deals with understanding and
modeling the behavior of individual consumers and producers as well as the aggregate of all
consumers and producers. The set of rules governing behavior are represented as economic
theories.

Economic Theory
Economic theory is a simplified representation of the complex real world. In economics, the
meaning of an economic theory is widely accepted to be the set of relationships governing the
behavior of individual producers or consumers or the set of relationships governing the overall
economy of a society or nation. Economic theory may also represent a hypothesis or proposition
about how something operates based on observation. Such hypothesis may be empirically
verified for their wider application over time and space. Economic models can be used to test
the validity of economic theory in time and space when answering the question “what would
happen if such and such a change took place? Economic models may be represented (1)
verbally, (2) graphically or (3) mathematically.

1.2 Some important concepts in Economics

Scarcity: is an important concept in economics because most of the goods and services are
scarce, including production inputs and consumption goods

• Note that there are human needs/desires on the one hand, most of which are insatiable, and
resources on the other hand, most of which are limited or scarce

3
• Resources, which are abundant in supply such as air, are free while scarce resources are
allocated based on the price mechanism or by rationing.

Allocation: Since many resources are scarce and the needs are unlimited, a mechanism must
exist to guide the use of resources. This entails choosing the best alternative which reflects the
real opportunity cost of the resource, i.e. that which is consistent with individual or social
objectives. For example a farmer must decide how much of one hectare of land to allocate to
maize and sorghum. Prices play an important role in the allocation of resources as well as goods
and services.

Goals (End or Objective): These represent the needs to be satisfied. Since individual needs or
desires appear to be unlimited, goals compete for scarce resources. Economics helps to choose
among competing alternatives by use of indicators. Examples of some indicators include prices
for consumers and producers, the profit motive for producers and utility in the case of
consumers. The goal is often to maximize or minimize these indicators accordingly.

Economic Problem
Thus for a problem to be economic, (1) scarcity and (2) alternatives must exist. For example if
there is scarcity but no alternatives, there is no choice to be made. On the other hand, if there are
alternatives but no scarcity (the goods or resources are free), the problem is not of economic
nature. In both examples allocation would not solve the problem.

1.3 Economic Modeling

Relationships in the real world are very complex. In order to understand relationships between
factors or causes and effects of economic relationships a number of assumptions are often made
so as to simplify the complex world. An economic model may be:
- Descriptive, (words) used since early periods of the professions (18th Century).
Most early text describing economic relationships, e.g. Adam Smith’s “The

4
 Wealth of Nations,” are entirely descriptive. Verbal description was limited in its
scope of addressing the question what if a certain change was introduced?
- graphic to represent relationships. An improvement over verbal representation,
and had more analytical capacity but limited in the number of variables or factors
that could be represented. In two dimension, limited to only two variable
- Mathematical. Capable of describing complex economic theories and analyses.
The advent of econometrics and Computers has made the application of
mathematical models even simpler and more popular.

Example: The relationship between the price of a product (PQ) and the quantity demanded
(Q) might be represented as follows:
(a) Verbal: As the price increases, the quantity demanded decreases
(b) Mathematical: Qi = f(Pi)
(b) Graphic:

PQ1

PQ2

Q1 Q2

Figure 1.2 A demand model

1.4 Types of Economics

Consumption versus Production economics

Economics involves making choices. A person who faces a limited income (and no one does
not) must choose to buy those items that make him or her feel most satisfied, subject to an

5
income limitation or constraint. Choice is central to consumption economics. A person is said to
derive utility (satisfaction) from consuming the item. The basic economic problem involves
maximization of the utility, subject to the constraint imposed by the availability of income.

The producer (firm owners) of goods and services that are desired by consumers is also faced by
a set of choices to be made. The producer also attempts to maximize utility. To maximize
utility, the producer is motivated by a desire to make more money, in order to fulfill unlimited
wants. Although the producer may have other goals, he/she frequently strives to maximize profit
as a means of achieving utility or satisfaction. Profits are subsequently used to purchase other
goods and services that provide utility or satisfaction to the producer of firm owner. Other
producers may however strive to maximize other goals, such as the amount of land owned, the
size of their enterprise, the type of machinery on their farm (the latest), nonetheless, profit is
often an overriding goal among most farmers.

Microeconomics versus Macroeconomics

Traditionally economics is categorized into Micro-economics and Macro-economics.

Microeconomics is concerned with the behavior of individual decision making units, ie

individual consumers as they allocate income or individual firm managers or producers (such as
a farmer) as they allocate their resources consistent with their goals.

Macroeconomics deals with the big picture i.e. issues that confront the entire economy. These
include inflation, unemployment, interest rates, and exchange rates (sometimes called the macro-
prices) and money supply. Macroeconomics is also concerned with the role of government
policy in determining fundamental questions that must be answered by any society. These
include: (1) What to produce, (2) How much to produce and (3) How should available goods and
services be allocated.

Although micro and macroeconomics are considered separate branches of economics they are
actually closely related. The macro-economy is made of individual consumers and producers.

6
 Their individual actions are reflected in the aggregate depending on the prevailing government
policies.

Static versus Dynamic Economics

Statiic economics can be thought of as one or more snapshot of events taking place in an
economy. The economic relationship represented by a graph for supply (SS) and demand (D1D1)
reflects a static relationship at a point in time. An event or shock such as an increase in income
that leads to increased demand would shift the demand curve to the right. A second demand
curve (D2D2) would accordingly be drawn to reflect this change. This tool of analysis is called
comparative statics, an intermediate between static and dynamic economics. Comparative statics
is often used in economics.

D2

P
D1
S
P2

P1 D2 (New income)

D1 (Old income)
S

Q1 Q2

Figure 1.2 Comparative Statics

With comparative statics, the underlying cause of the income change that leads to a shift in
demand is not considered, nor is the time interval between the two demand schedules (timeless).

7
In static and comparative statics, future expectations are single valued Comparative statics is
used in both micro and macroeconomic analysis.

Dynamic economics on the other hand can be thought of as a moving picture of the economy.
Time is important in dynamic economic analysis. Dynamic economics attempts to show the
process by which an individual consumer, firm or economy moves from one equilibrium to
another. For example, dynamic economic analysis may attempt to show the path of machinery
investments made by a farmer over a twenty-year period. Or, changes in the quantity of a
commodity demanded one day, one week or one month from the time of the price change.

Risk and uncertainty are an important part of dynamic analysis. Often future outcomes are not
known with certainty. Thus they are multiple rather than single valued. For example, applying
100 Kg of Urea to one hectare of maize may increase yield by 15 bags, 10 bags or 5 bags
depending on other factors such as the amount of rainfall, the rate of weeding etc. When the
probability of an outcome is known, we are dealing with risk but when the probability is not
known we have an uncertain situation.

Positive versus Normative Economics

Positive economics deals with describing the functioning of economic units (what is) in terms of
characteristics and the nature of relationships for households, firms, markets and economic
systems. Under positive economics the specification of goals or policies is left to decision-
makers. There is no prescription of what is good or bad. Disagreements may be resolved by
gathering more facts.

Normative economics on the other hand is prescriptive. The analyst says what ought to be. This
means there is value judgement. Economic policy analysis falls under normative economics.
Positive economics serves as a basis for making decisions in normative economics. For example
positive economics may be able to show that a 5% tax increase on the rich will raise 10 billion
shillings without significantly reducing their consumption demand. Normative economics may
go further to recommend that this money will be best utilized if it is allocation to public

8
education and public health because this will mostly benefit the poor who cannot afford to go to
private schools and hospitals..

Economics versus Agricultural Economics

Agricultural economics is an applied science dealing with how people choose to use technical
knowledge and scarce productive resources (land, labor and management) to produce food and
fiber and distribute them for consumption to various members of the society over time.
Agricultural economics uses scientific methods to find answers to problems in agriculture and
agribusiness. An agricultural economist must be a good economist in the sense that he/she must
have good mastery of economic theory. However, an agricultural economist is an economist
with specialization in agriculture. Knowledge in agriculture (e.g. agronomy, animal husbandry,
forestry, mathematics, statistics, business and other social sciences) is necessary if an agricultural
economist is to portray relationships accurately using models of some component of the
agricultural sector.

Agricultural Production Economics

Role of economic theory in production
Agricultural production economics is concerned primarily with economic theory as it relates to
the producer of agricultural commodities. Some major concerns in agricultural production
economics include the following:

Goals and objectives of a farm manager: Often assumed to be maximizing profit. However,
individual farmers may have their own unique goals. Nonetheless, most economic models used
for representing the behavior of farm managers assumes that they are striving to maximize
profits. Or at least striving to maximize revenue, subject to some resources constraints.

Choice (What to produce): A farm manager faces an array of options with regard to what to
produce, given available labor, farm tools and equipment. The manager must decide how much
of each particular commodity to produce.

9
Allocation of resources among outputs: Once the decision on what to produce has been
made, the farm manager must decide how available resources are to be allocated among various
outputs. For example, which crops should be planted on which farm plots? How much of labor
and machinery should be allocated to each farm plot?

Risk and Uncertainty: Models of production economics frequently assume that the
manager knows with certainty the applicable production function ( for example the yield that
would result for a crop if a particular amount of fertilizer was applied) and the prices of both for
inputs to be purchased and output to be sold. However in agriculture, the assumption of perfact
knowledge with respect to the production function is never met. The weather, an important
factor in agricultural production cannot be predicted with certainty, Animals and plants farmers
must make decisions with less than perfect knowledge with regard to the amount they will obtain
at the time of harvest and the price at which they will sell their produce.

Thus agricultural production economics deals with:

• How individual farms react to given changes in the economy
• How Agricultural economists should advice farmers on how to react to such changes
For example, if the demand for an agricultural commodity changes (due to change in income
or taste) how will farm prodction and farm income be affected?
• How changes in supply or prices of agricultural inputs affct farm operations?

1.5 Assumptions of Pure or Perfect Competition

Agriculture is often considered the closest approximation of the perfectly competitive model.
The basic assumptions of the model are reviewed below:

A large number of buyers and sellersexist in the industry: as such, buyer and sellers are price
takers in the output and input markets respectively. Although this assumption is valid to a large
extent, it does not always hold true. For example, fertilizer in Tanzania is distributed by only a
handful of dealers, the main export crops such as cotton, coffee, tobacco and cashew-nuts are
bought by a limited number of buyer. In the case of tobacco, there has been evidence of

10
collusion among buyers to set producer prices, thereby violating the underlying assumption that
no single buyer or seller is large enough to influence the price.

Farmers are price takers and not price setters: This implies that farmers can sell as much as
they want at the market price without their supply to the market affecting the market price. This
means the level of production by individual farmers is so small that their additional supply to the
market does not influence the price. This assumption may however be violated in the case of
monopolistic producers/sellers of certain products such as fertilizer, agricultural chemicals or
farm implements.

The products are homogenous: This implies that the products produced by all firms in the
industry are identical. As such there is no need for advertising since there is nothing to
distinguish the product of one firm from another. This assumption holds true for many
agricultural products. For instance, a consumer is unlikely to distinguish between maize from
two different farmers. Nevertheless, some producers may attempt to differentiate their products
in order to fetch premium prices. For example, at Morogoro market, peas from Matombo sell at
a higher price than similar looking peas from Iringa.

There is free entry and exit: This implies mobility of resources in and out of farming. The free
mobility assumption may be easily met under small holder peasant agriculture but it becomes
increasingly difficulty to transfer resources out of agriculture as the level of specialization and
capitalization increases. A farmer who just acquired a tractor for maize production may find it
difficult to dispose it off his/her within a short time if market conditions change such that maize
production is no longer profitable. Other artificial restraints to free entry may also exist. These
include government policy. For example, the pan territorial price policy for inputs and outputs
which existed in Tanzania during the 1970 and 1980s is a case in point. This policy in effect
subsidized agricultural production in remote areas thereby interfering with free entry.

Perfect certainty: By this assumption, it is supposed that all variables of concern to the
producer are known with certainty. Some economists distinguish between pure and perfect
competition. They argue that pure competition can exist even if all variables are not known with

11
certainty to the producer and consumer. However, perfect competition will exist only if the
producer knows not only the prices for which output will be sold but also the prices for inputs.
Moreover, with perfect competition, the consumer has complete knowledge with respect to
prices. Most important, with perfect competition, the producer is assumed to have complete
knowledge of the production process or the function that transform inputs or resources into
output or commodities. Since nature varies from year to year, this assumption is violated
continuously in agriculture

Nevertheless, the assumption of pure of perfect competition is often assumed in economic

analysis involving the agricultural industry because, despite its weakness, the perfectly
competitive model of economic behavior comes close to representing farming than other
available models such as monopoly, monopolistic competition or oligopoly. In using the purely
competitive model during analysis, modifications are made depending on the specific condition
of the particular market or situation being analyzed.

1.6 Discussion Questions

1. Discuss the role of microeconomics versus macroeconomics in agricultural economics.

Does microeconomics have a greater impact than macroeconomics on the farm manager?

2. If pure competition is not an adequate representation of the economic model that underlies
farming in Tanzania, why do the assumptions of pure competition continue to be important
to Agricultural economists?

3. The real world is dynamic, why do agricultural economists continue to rely so much on
comparative statics as a tool of analysis

4. Is agricultural production economics positive or normative?

5. To become an agricultural economist, is it more important to know agriculture or to know

economic theory?

12
2. PRODUCTION FUNCTIONS
2.1 Important Concepts

What is a production function?

A production function represents a technical relationship between inputs (resources or factors of

production) and output. For a particular technology, a production function describes how inputs
are transformed into products. For example the rate at which fertilizer that is applied into the soil
combines with moisture and sunlight and gets transformed into maize or beans. A production
function may be represented in mathematical, graphic or tabular form as illustrated below.
Symbolically the production function may be written as:

(2.1) Y = f(x)

Where Y = output (The dependent variable)

X = input for x ≥ 0 (The independent variable or variable input)

In the case of more than one independent variable inputs the production function may be written
as:

(2.2) Y = f(x1, x2, x3,…………,xn)

The exact functional form of the production function depends on the technology. Equation 2.3

gives a specific example of a production function with one variable input or dependent variable.

(2.3) Y = 0.75x + 0.0042X2 – 0.000023x3

In a production function, each value of Y is assigned to at least one value of x and two or more
values of x can relate to the same value of Y, but the converse is not true. No value of x can
relate to two different values of Y. This is a cardinal rule of mathematical functions.

13
 Table 2.1 represents a production function in tabular form:

Quantity of Nitrogen Yield of Maize

(Kg/ha) Kg/ha)
0 6,276
45 9,415
90 13,181
135 14,436
180 15,441
225 16,068
265 15,566

Note: This is a hypothetical example

Y
Bags of maize

Bagsof Fertilizer X

Figure 2.1 Graphic representation of a production function

Production functions may be determined through

(a) experiments as in the case of fertilizer rates
(b) by studying farm records involving a homogenous group of farmers
(c) using a combination of experiments and available farm record

Researchers, for purposes of economic analysis normally estimate production functions.

Farmers normally don’t go about estimating production functions using the methods listed

14
above. Nevertheless, they know (be it intuitively) the technical relationship between inputs and
output in their farm through years of experience.

Fixed versus Variable Inputs and the Length of Run

When we represent the production function in terms of inputs and output:

(2.4) Y = f(x)

The model is actually oversimplified because it is assumed that all the other inputs that go into
the production process are held constant (ceteris paribus) except for the variable input of interest
during this analysis. For example in the case of variations in the quantity of nitrogen, we assume
that other inputs such as land, labor, pesticides moisture are held constant as the amount of
fertilizer is varied. This should symbolically be represented as:

(2.5) Y = f(x1, x2, x3, x4, ………….x5)

In equation 2.45x1 is the variable input while x2, x3……xn are fixed inputs, meaning that they are
held constant as x1 is varied (all other things being equal or ceteris paribus).

A variable input is an input that the farm manager can control or for which he or she can alter the
level of use. This implies that the farmer has sufficient time to adjust the amount of the input
being used.

A fixed input is usually defined as an input, which for some reason the farmer cannot change
during the production period. For example once the production season starts, the amount of land
available to the farmer can be considered as being fixed for that season. Thus, the farm manager
may be facing fixed inputs due to (1) time, (2) economic prudence such as in a situation where
increasing or decreasing the input would reduce profits (e.g. the amount of yeast in bread
making) or (3) a personal decision such as in a situation when the farmer is doing an experiment

15
to determine the milk production response when the ration is changed. The set of fixed inputs is
also referred to as the plant.

However, given enough time the amount of all inputs can be changed. Thus economists define
production situations as they relate to inputs in three phases:

The very short run - this is a length of time too short for the firm manager to change
any of the inputs. Thus in the very short run all inputs are fixed. An example is a production
season once it commences. The farmer will have assembled all the input and most of them
cannot be returned to the stores. As such they may be considered as being fixed.

The short run or intermediate run - a time period of such length that at least on of the
inputs is variable. As such they may be considered to be fixed.

The long run - a time period of such length that all inputs can be varied. For example
before plans are put into effect, all inputs are variable

These periods are rather arbitrary, defined to simplify economic analysis. In practice, farmers
are always faced with some fixed input as the production season proceeds. They are essentially
moving from one short run to another.

Important Assumptions Underlying Production Functions

Perfect Certainty - When production functions are given, representing the technical relationship
between inputs and outputs, it is assumed that the technology is perfect and farm managers make
their production plans under conditions of certainty. But in reality, there is a lot of risk and
uncertainty with regard to the efficiency of the production process (technology) as well as prices.
Agricultural production involves a lot of inputs that cannot be controlled. These include the
weather, government policies, events in other countries etc. As such, last year’s production is
only a guide to this year’s production decisions. This assumption of perfect certainty is made to
simplify the real world so as to facilitate analysis.

16
Yield
Good rainfall

Risk
Too much rain or
Uncertainty
Dry year

Fertilizer
Figure 2.3 Uncertainty due to Weather Variations

Level of technology: It is assumed that farmers use the most efficient technology available to
them, i.e. the process that gives the most output for the given level of inputs. If farmers don’t
use the most efficient technology available, it is possibly because they are not aware of it or they
cannot afford it. For example farmers continue to use the hand hoe because they cannot afford or
they do not have access to more efficient tools of primary and secondary tillage.

Length of time: It is assumed that the farmer manager is operating in one of the time
periods in relation to inputs used as discussed above:

The very short run - when all inputs are fixed

The short run - when at least one input is variable
The long run - when all inputs are variable.

17
2.2 Total Physical Product, Average Physical Product and Marginal Physical Product

As previously discussed, production functions provide a technical relationship between inputs

and outputs, in the next section we will attempt to develop some general principles that apply to
production functions in general. To do this the classical production function will be used since it
displays all the characteristics necessary for the purpose and it is commonly used for describing
production relationships in agriculture. Equation (2.6) represents an example of the classical
production functions.

(2.6) Y = 3x + 2x2 – 0.1x3

Total physical product, abbreviated as (TPP), is the total output hereafter denoted as Y and the
variable input is denoted as x.

The average physical product, abbreviated as (APP), is the ratio of output to variable input,
obtained by dividing the total physical product by the variable input.

(2.7) APP = Y/x

This parameter measures the average rate at which inputs are transformed into output. The APP
therefore measures the efficiency of production or the efficiency of using variable inputs. The
shape of the APP depends on the production function. Geometrically the APP is defined by the
slope of a ray from the origin that intersects or is tangent to the TPP. From the production
function in equation (2.6);

(2.8) APP = 3 + 2x – 0.1x2

Note that for the production function given above, the APP is not defined when x = 0. Even
though the algebraic function (2.8) implies that APP is equal to 3 at this point, rather, 3
represents the limit of the APP as x approaches 0.

18
The marginal physical product, abbreviated as (MPP), is defined as the change in output
resulting form a unit incremental change in variable input. The shape of the MPP depends on the
shape of the production function. Geometrically, the MPP represents the slope of the production
function. There are two ways to compute MPP;

• Average MPP is computed using tabular data of a production function

(2.9) Average MPP = ∆Y/∆x

• Exact MPP is computed when a continuous differentiable production function is defined.

The exact MPP is the first derivative of the TPP obtained using differential calculus.

(2.10) Exact MPP = dY/dx

From the production function given in equation 2.6;

(2.11) Exact MPP = dY/dx = 3 + 4x – 0.3x2

Note that the average MPP between two points on a production function is an approximation of
the exact MPP of points that lie between the two points. Thus there will always be a difference
between the exact and average MPP computed for a given point using tabular data and calculus

For the classical production function under consideration,

• When the MPP is increasing , TPP is increasing at an increasing rate
• When MPP is decreasing but positive, TPP is increasing at a decreasing rate
• When MPP is zero, TPP attains a maximum
• When MPP is negative, TPP is decreasing

19
 Slope equal
Y
zero
or
TPP Slope equals MPP,
equals Max. APP
Maximum TPP

Maximum APP

Slopes
equal APP
Point of
inflection

APP Max. MPP

or Max. APP
MPP

APP

0 MPP

X
MPP

Figure 2.4 A Classical Production Function (Adapted from Debertin, 19…)

20
2.3 The Law of Diminishing Returns

The law of diminishing returns was developed by early economists to describe the relationship
between output and a variable input when other inputs are held constant. The law of diminishing
returns has widespread application. The law states that;

If increasing amounts of an input are added to a production process while all other inputs are
held constant, the amount of output added per additional unit of variable input will eventually
decrease.

The law of diminishing returns has invariably also been states as follows;

As units of a variable input are added to one or more fixed inputs, after some point, each
incremental unit of variable input produces less and less additional output.

or

As units of a variable input are added to units of fixed inputs the proportions change between
fixed and variable inputs

The LDR refers to the rate of change in the slope of the production function. The law of
diminishing returns is also referred to as the law of diminishing marginal returns,

For example, as nitrogen fertilizer is applied to a maize farm, after some point, each additional
unit of nitrogen would produce less and less additional maize

This law implies that there is a right amount of variable inputs to be used in combination with
fixed inputs. The farm manager should use neither too much or too little. The law of diminshing
returns says nothing about increasing returns even though production functions may have
increasing returns within their range.

21
The classical production function presented above initially exhibits increasing marginal followed
by diminishing marginal returns, but for other types of production functions marginal returns
could be decreasing right from the first unit of variable input applied to the production process.

Assumptions of the Law of Diminishing Returns

(1) The law assumes that the technology is constant. That is the method of production
remains the same as changes are made to the amount of variable input
(2) The law of diminishing returns applies to changing proportion between variable and one
set of fixed inputs. The law does not apply when all inputs are changed simultaneously.

The law of diminishing returns is sometimes referred to as the law of variable proportions or the
law of decreasing productivity.
Some people apply the LDR to APP and TPP, but this is not commonly done.

2.4 Three Stages of Production

The classical production function can be divided into three regions or stages based on the
efficiency of resource utilization

22
 Y

TPP

I II III

10 14 x
MPP
or
APP
MPP
APP

Figure 2.5 Three Stages of the Classical Production Function

As can be seen from figure 2.5:

• Stage I occurs when the MPP is greater than zero and greater than the APP and the APP is
increasing. This means the efficiency at which variable inputs are transformed into output, as
measured by the APP is increasing throughout stage I.
• Stage II occurs when the MPP greater than or equal to zero but less than the APP and the
APP is decreasing

The efficiency of using the variable input as measured by the APP. When the APP attains a
maximum at the beginning of stage II. Variable inputs are used most effeciently

23
On the other hand the efficiency of using fixed inputs is measured by dividing the output by a
one level of fixed inputs. Since the TPP attains a maximum at the beginning of stage III, at this
point the ratio of output to fixed inputs should be the highest, therefore representing the most
efficient use of fixed inputs.

• Stage III occurs when the MPP is negative and the TPP is decreasing. This occurs when
excessive variable inputs are combined with fixed inputs. That is why TPP begins to
decrease

Economic Interpretation

Although the price of inputs and outputs must be known in-order to determine the most
profitable level of production, some general recommendations can be made based on the
technical relationship between inputs and output as represented by the production function.
These recommendations include:

• If the product is of any value at all (and therefore it must be produced), then, varialble inputs
should be used at least up to the beginning of stage II. That is up to the point where the
efficiency of using variable inputs is highest. This is the point where the APP attains a
maximum.

• Even if inputs are free, no variable inputs should not be applied in stage III because
additional variable inputs would result in a decrease of the TPP

• The boundaries of stage II represent the limits of economic relevance. The optimal level of
variable inputs to be used (the level that yields the highest level of profit) occurs somewhere
in stage II. However, the exact amount of input can only be determined once input and
output prices are known.

Algebraic Interpretation

The boundaries of the three stages can also be determined from an algebraic representation of the
production function. Continuing with the function given in equation (2.6)

TPP Y = 3x + 2x2 – 0.1x3

24
APP APP = 3 + 2x – 0.1x2

At the boundary of stage I and II the APP attains a maximum, which means the slope of the APP
at this point is equal to zero. The value of x and this point can be determined by equating the
first derivative of the APP to zero and subsequently solving for x.

Slope of APP d APP/dx =2 – 0.2x =0

∴ X = 10

At this point, MPP = APP

The slope of the TPP is equal to zero when TPP is its maximum, which means the MPP is equal
to zero at this point. This marks the boundary between stage II and III. The value of x at this
point can be determined by equating the MPP to zero and solving for x.

MPP MPP = 3 + 4x – 0.3x2 = 0

X = [-4 ± √ (16 +3.6)]/0.6 = 14.04

The Elasticity of Production and the Point of Diminishing Returns

The elasticity of production is a measure of the responsiveness of output to changes in inputs. It

is defined as the percentage change in output given a unit percentage change input. The
elasticity of production like all other elasticities in unitless.

εp = Percentage change in output/ Percentage change in input

εp = ∆Y/ Y ÷ ∆X/ X
εp = (∆Y/ ∆x) (x/Y) = MPP/APP
One can compute:
• Average elasticity of production, if the production function is defined in tabular form or only
two coordinates of a production function are defined

25
• Exact elasticity of production, if a continuous differentiable production function is defined

Elasticity of Product Substitution and the Point of Diminishing Returns

At what point does the point of diminishing returns set in?
The MPP begins to diminish at the point of inflection
The APP begins to diminish at the point where MPP = APP
The TPP begins to diminish at the point where MPP = 0

The determination of the point of diminishing returns can be ambiguous, that is the point at
which returns begin to diminish. To avoid such ambiguity, some authors apply the law of
diminishing returns directly to the marginal physical product, thereby referring to the law of
diminishing marginal returns. In this case, the elasticity of product substitution provides an
unambiguous way to define the point of diminishing returns as follows;
% ∆ Output MPP
εp = --------------- = ------
% ∆ Input APP

In stage I εp > 1
In stage III εp ≤ 0
In Stage II 0 ≤ εp ≤ 1

The point of diminishing returns occurs at the point where MPP = APP and εp = 1. In the

classical production function, the point of diminishing returns occurs at the lower boundary of
stage II. This marks the point where variable inputs are used most efficiently, and this is where
the minimum amount of variable inputs should be used. At the upper limit boundary of stage II,
MPP = 0, and εp = 0. Consequently, the relevant interval of production for a variable input is
within the range where;
0 ≤ εp ≤ 1

26
2.5 Forms of Production Functions

Besides the classical production function, which is often used for illustrative purposed, there are
many other forms of production functions that are used in empirical work. These include:

1. Linear production functions e.g. Y = a + bx

2. Power production functions e.g. Y = bxn
3. Quadratic production functions e.g. Y = a + bx - cx2; Where a, b, c and n are constants

4. CES functions; these have a constant elasticity of substitution

5. Linear Programming
6. Input-Output relations

At this level, at most you will come across the first three types of functions and some examples
are given below:

(1) A linear function

Y = a + bx
MPP = b

For more than one variable input

Y = a + bx1 + cx2

MPPx1 = ∂ Y/∂ x1 = b ; MPPx2 = ∂ Y/∂ x2 = c

εp = b (x/Y)

27
(2) Power functions (e.g. the Cobb-Douglas function)

Y = axb

MPP = baxb-1
APP = axb-1

εp = (abxb-1)/ (axb-1) = b

• Power functions can be linearlized by logarithm transformation

Lny = a + bLnx

(3) Quadratic Production Function

Y = a + bx - cx2
MPP = b – 2cx

This kind of production function allows for interaction between inputs

e.g. Y = a + b1x1 + b2x2 – b3x12 – b4x22 + b5x1x2

In this case the MPP of x1 will be influenced by x2 and vise versa

MPPx1 = ∂Y/∂x1 = b1 –2b3x1 +b5x2

MPPx2 = ∂Y/∂x2 = b2 –2b3x1 +b5x1

Discussion Questions
1. Discuss the meaning of a production function
2. A farmer should handle all the land within the farm such that crop yields are maximized.
Comment
3. List four categories of inputs that are important in agricultural production

28
3. COST FUNCTIONS

Cost functions represent expenses incurred in organizing and carrying out production processes.
They include the cost of inputs and services used in production such as the cost of fertilizer,
labor and management. In the short run costs are viewed as fixed and variable costs. In the
long- run, all costs are considered to be variable.

Fixed and variable costs

A resource or input is called a fixed resource if its quantity does not vary during the production
period, while the quantity of a variable resource used in a production process depends on the
quantity of output being produced (the level of production). Most inputs are not free as such
there are costs associated with them. In general costs of fixed inputs are called fixed costs, while
those of variable inputs are called variable costs. Fixed costs are incurred irrespective of
whether or not production is undertaken. Thus, they do not change in magnitude depending on
the amount of output produced. Examples of fixed costs include land taxes, principal and
interest payments on loans, insurance premiums, wages for permanent employees. These are
cash costs. Non-cash fixed costs include depreciation costs on buildings, machinery and other
equipment, interest charges on capital investments and charges on family labor and management.
Variable costs depend on the level of production. They include purchased inputs such as seed,
fertilizer, herbicides, insecticides and hired labor. In the case of livestock within a single
production season, feed and medicine are the major variable cost items.

Computation of costs

In the following section, we will discuss how to compute the various cost items.

Total Costs
1. Total fixed costs (TFC) represent the summation of all the fixed costs. This is constant
throughout the production process for a given period.

29
2. Total variable costs (TVC) are computed by multiplying the amount of variable input used
by the price of per unit input. TVC = Px.X. These costs vary as output changes.

• The shape of the TVC function depends on the shape of the production function.

3. Total costs (TC) are the sum of TVC and TFC, i.e.
TC = TVC + TFC
= Px.X + TFC

• The shape of the total cost function depends on the shape of the production function

From these basic cost functions, a number of other cost functions can be derived as presented
below:

Average Costs
1. Average Fixed Cost (AFC)

The average Total Cost is obtained by dividing the total cost by the amount of output.

AFC = TFC/Y Where Y is output

Cost

AFC

Output
Figure 3.1 Average Fixed Cost Curve

30
• AFC varies depending on the level of output.
• As Y increases, AFC declines.
• AFC always has the same shape regardless of the production function.
• Average fixed cost decreases with the level of output and eventually becomes asymptotic to
the horizontal axis. This effect is due to spreading fixed costs over more output.

2. Average Variable Costs (AVC)

The average variable cost (AVC) is obtained by dividing the total variable cost (TVC) by the
level of output (Y).

AVC = TVC/Y

The AVC varies depending on the level of production. The level of the AVC curve depends on
the unit cost of the variable input.
• The AVC is inversely related to the average physical product (APP).
When the APP is increasing, the AVC is decreasing. When the APP is at its maximum, the AVC
is at its minimum and when the APP is decreasing, the AVC is increasing.

AVC = TVC/Y
= Px.X/Y
= Px.(X/Y), But, (X/Y) = APP
∴ AVC = Px./ (APP)

For a production function the APP measures the efficiency of using variable inputs, likewise in
the case of cost functions, the AVC provides the same measure.

3. Average Total Cost (ATC)

The average total cost is the sum of the average fixed cost and the average variable cost.

ATC = AFC + AVC

• The shape of the ATC curve depends on the shape of the production function.

31
Traditionally, the ATC is always referred to as the unit cost of production. That is, the cost of
producing one unit of output. As the level of production increases from zero, the ATC initially
declines due to spreading fixed costs among an increasing number of units of output. Also due
to increasing efficiency in using the variable inputs ( as indicated by a declining average variable
cost curve). However, as output increases further, AVC attains a minimum and it begins to
increase. Thereafter, the AVC and AFC, which constitute the ATC, move in opposite directions
(one is decreasing while the other is rising). From this point onwards, initially, the ATC
continues to decline, implying that the rise in AVC is not strong enough to offset the declining
effect of the AFC. Once the declining effect of the AFC can no longer offset the rising effect of
the AVC, the ATC also begins to rise.
Cost

ATC

AVC

AFC

Output

Figure 3.2 Average Cost Curves

4. Marginal Cost

The marginal cost is defined as the change in total cost, per unit change in output. It is the cost
of producing one additional unit of output.

32
MC = ) TC/) Y

But, TC is the sum of TFC and TVC. Since TFC does not vary with the level of production, it
means that any variation in the TC must be due to variation in the TVC.

Thus ) TC = ) TVC

Therefore, MC = ) TVC/ ) Y

From a geometric point of view, the MC is the slope of the total cost curve.

• The shape of the MC curve is inversely related to that of the Marginal physical product
(MPP), as illustrated by the following mathematical derivation:

MC = ) TC/) Y
= ) TVC/) Y
= ) (Px.X)/) Y
= Px ()X/) Y) But, ) X/) Y = 1/MPP
= Px/MPP

MC
Cost

ATC
AVC

AFC

Output

Figure 3.3 Cost Curves

33
• As long as there is some fixed cost, the marginal cost curve crosses the ATC curve at its
minimum but this is at a greater level of output than the output at which AVC is at its
minimum

• Stage II of the classical production function begins at the point where MC = AVC

• At the boundary of stage II and III, we have already established that MPP = Zero. This
means:
MC = Px/MPP = Px/0 = Undefined

In other words at the boundary of stage II and III, the MC would be vertical and parallel to
the vertical axis. From an economic point of view, at this point the MC ceases to have any
meaning.

• It should be noted that the change in Total cost resulting from a unit change in variable input
is not the marginal cost. Rather, this is referred to as the marginal factor cost (MFC). In a
perfectly competitive market, the MFC = Price of the variable input (Px)

These various cost functions can also be represented in tabular form as shown in Table 3.1. As
part of the class exercise, fill in the remaining gaps in table 3.1 below.

Table 3.1 Various Cost Functions in Tabular Form

X Y TFC TVC TC AFC AVC ATC MC

Average Exact
0 0 1000 - 1000
1 4.9 1000 100 1100
2 13.2 1000 200 1200
3 24.3 1000 300 1300
4 37.6 1000 400 1400
5 52.5 1000 500 1500
6 68.4 1000 600 1600
7 87.4 1000 700 1700
8 100.8 1000 800 1800
9 116.1 1000 900 1900
10 130.0 1000 1000 2000
11 141.9 1000 1100 2100
12 151.2 1000 1200 2200
13 157.3 1000 1300 2300
14 159.6 1000 1400 2400
15 157.5 1000 1500 2500
16 150.4 1000 1600 2600

Source: Doll and Orazem (1978)

34
With regard to cost functions, it should be noted that:
• Managers/farmers are interested in the total cost of producing an output as well as the unit
cost of production at a given level of output. Thus, costs are usually expressed as a function
of output

• Marginal cost is a widely used concept in agricultural economics. It is the increase in total
cost resulting from one unit increase in output. NOTE that the marginal cost is not the
change in Total cost given a unit increase in inputs.

Empirical Estimation of Cost Functions

Cost functions can be estimated either directly or indirectly.

1. Direct derivation of cost functions

Direct estimation of cost functions involves making estimation of costs and output at various
levels of production for a given technology. This can be attained using cross sectional data from
a sample of firms (eg. Farms) that are involved in a given production process. Also data for
direct estimation of cost functions may be obtained from time series data for a particular firm
over time. In such a case, the firm must have changed the volume of production over time, using
different levels of inputs. Direct estimation of cost function therefore entails statistical analysis
of accounting data together with engineering or production data to derive appropriate cost
functions. Other cost functions may be derived as previously defined

TC = TVC + TFC
ATC = AVC + AFC
AFC = TFC/Y
AVC = TVC/Y
MC = dTC/dY = dTVC/dY

Example:

Senkondo (1988) made direct estimation on the Total cost function using production and
accounting data from Kilombero Sugar Company as follows;

35
TC = 29,890 + 6,941,620 Q – 294,150 Q2 + 3,990 Q3

Other cost functions were subsequently derived as follows:

ATC = 29,890/ Q + 6,941,620 – 294,150 Q + 3,990 Q2

= 6,941,620 + 29,890/ Q – 294,150 Q + 3990 Q2

AFC = 29,890/ Q

AVC = 6,941,620 – 294,150 Q + 3,990 Q2

MC = d TC/d Q
= d TC (Q)/d (Q)
= 6,941,620 – 588,300 Q + 11,970 Q2

For another example of a case where the cost function is known, see Doll and Orazem Chapter 2.
(pages 35 – 35)

2. Derivation of Cost Functions from Production functions

Cost curves express the cost of fixed and variable inputs, as functions of the amount of output.
Meanwhile, production functions express output as a function of inputs. But input costs are the
product of input quantity times the input price. It follows therefore that cost and production
functions must be inversely related. This means, knowledge of one implies knowledge of the
other, as long as input prices are known.

TVC = Px.X (Total variable cost function)

Y = f (X) (Production function)

Normally, cost functions are expressed as a function of output rather than input. In order to meet
this requirement, we express the inputs as a function of output. This is the inverse production
function represented as follows:

X = f-1 (Y)

NB. This inverse function does not imply cause and effect. In other words, the function above
does not imply that output (Y) causes input (X). Rather, the inverse equation (f-1) is a

36
relationship, which gives the minimum amount of input required to produce a given level of
output.

Example 1: Using a quadratic production function:

Y = 8X - ½ X2
APP = Y/X
= 8 - ½ X2
MPP = dY/dX = 8 – X

An inverse function of Y = f (X) can be obtained using the quadratic formula of the form:

0 = a X2 + bX – C as follows:
0 = ½ X2 - 8 X – Y Where a = ½; b = - 8; C = Y

Deriving the cost function:

X = - (-8) ± [82 – {4.(½).Y}]½ / 2 (½)

= 8 ± [64 – 2Y]½ / 1

= 8 ± [64 – 2Y]½
TVC = Px.X
= Px [8 – (64 – 2Y)½]
MC = dTVC/ dY (Using the product rule for differentiating a function)
= Px/ (64 – 2Y)½

Example 2: Using Power Function (Cobb-Douglas)

Y = 6 X½

APP = 6X½ / X
= 6 / X½

MPP = dY/dX
= 3X-½
= 3/ X½

Deriving cost function:

Y = 6X½
Y2 = 36X
X = Y2/36

TVC = Px.(Y2/36)

37
AVC = Px (Y/36) or (PxY)/ 36

MC = d TVC/ d Y
= 2Px (Y/36)
= (Px.Y)/18

Example 3: Using a linear function

Y = 2X

APP = MPP = 2

Deriving cost functions:

X =½Y

TVC = Px.X
= Px (½ Y)
= ½. Px.Y

AVC = MC = = ½. Px

Review Questions

See Doll and Orazem

38
4. PROFIT MAXIMIZATION DECISIONS

The main assumption for profit maximization is that farmers by their inputs in purely
competitive markets. It is likewise assumed that they sell their products in competitive markets.
This means as buyers, the volume of their purchase is too small to influence the market price,
also, the volume of their sale is too small to change the market price.

Students should review theories of monopoly, oligopoly, perfect competitive markets etc and
write brief notes on each type of market. Listed below are main characteristics that differentiate
markets.

Types of markets are classified according to:

• Size of the firms or farmers who sell the product
• Number of buyers who buy the products from the market and the volume of the product
purchased
• Homogeneity (similarity or lack of it) of the product sold or purchases
• Ease of entry or exit from the market (to produce the product)

The agricultural industry in Tanzania may be cited as one example whose characteristics make it
close to a competitive market. These characteristics include the following:
• There are many smallholder farmers who buy inputs from the same market and sell their
products in the same market.
• Agricultural products of a given type are similar in appearance and quality
• In most cases, smallholder farmers are price takers, unless they are selling a differentiated
product. Thus Prices are assumed to be constant at a given point in time.
• However, in local markets, farmers may influence the price of products in the market.

Before we proceed, we need to make additional assumptions about farmers and the market
conditions under which they operate.
• Farmers want to maximize profit, however, other objectives such as food self sufficiency also
apply.
• Prices and technical relationships between inputs and output are assumed to be known with
certainty
• In order to understand the process of profit maximization, we will initially assume that there
is only one variable input and one output

39
Definitions and concepts:
In a purely competitive market:
TPP = Y (4.1)
Py = constant
∴Py.TPP = Py.Y (4.2)
Total value of the product (TVP) = Total Revenue
For a farmer, TVP represents the monetary value received from the sale of a single commodity
such as maize, beans, milk beef or goats etc. In a purely competitive market individual sellers
are assumed to be price takers, which means the quantity of a product supplied by one farmer to
the market does not change the price a given point in time. Thus for the purpose of decision
making, the product price is assumed to be constant for the individual farmer, the TVP function
has the same shape as the TPP function. Only the units on the vertical axis change.

Total factor Cost or Total Resource Cost is the product of the quantity of the input used in a
production process times the price for that input. In the following discussion, in order to avoid
confusion between the notation for Total Factor Cost (TFC) and Total fixed Cost (TFC), we are
going to use Total Resource Cost (TRC). In a purely competitive market, individual buyers are
assumed price takers. This implies that the price of the input does not vary with the quantity an
individual farmer purchases from the market.

TRC = Px.X (4.3)

Maximizing Profit with One Variable Input

Profit is the difference between gross returns and total cost. Using he notation described above,
profit (Π) is the total value product (TVP) less the total cost (TFC). Profits are also referred to as
net returns. The profit function for a farmer can be written as:

Π = TVP – TC (4.4)
or
Π = Py.Y – Py.X

40
SHS

TVP

Inflection
Point

SHS

VMP
AMP
MFC MFC

AVP

O
X

VMP

Figure 4.1 The relationship Between TVP, MVP, AVP and MFC

41
Determining Optimal Amount of Input (Input Criterion)

The optimal level of inputs is that level, which produces the highest level of profit. This level of
input can be deduced using three criteria:

(1) Using the Net return criterion

Π = TVP - TFC
As pointed earlier, profit is equal to the net return obtained by a farmer, which is the difference
between the TVP and the TFC. Thus the profit can be computed directly and the optimal level of
input use is at the point where profit attains a maximum. Table 3 is given as an example of such
a computation.

Table 4.1 Determining Optimal level on Input

INPUT OUTPUT MPPP TOTAL COST TOTAL VALUE PROFIT (Π)

(X) (Y) (TC=TVC+TFC) PRODUCT (TVP)
0 0
20 16.496
40 35.248
60 55.152
80 75.104
100 94.000
120 110.736
140 124.208
160 133.312
180 136.944
200 134.000
220 123.376

NB.
• Total fixed cost = 0
• MRC = 0.15
• Py = 4

Once Table 4.1 has been completed, the point at which profit is highest can be determined, and
therefore the optimal level of input. This can also be illustrated graphically as in Figure 4.2.

42
 SHS

TVP
Π

TC

1,000

0 180 X
Figure 4.2 Optimum level of Inputs Using Net Return Criterion

(2) Using Profit as function of Input (Π = f(Xi)

This may be done graphically. Based on data from Table 2 for example, profit can be expressed
as a function of inputs. The optimum level of input occurs when profit is at maximum

SHS

1,000 Profit

0
A B

-1,000

X
0 180

Figure 4.3 Optimum Level of Input Using Profit as Function of Input

43
 (3) Using Marginal Criteria

The production function expressed in Table 4.1 may be expressed using a mathematical function
as follows:
Y = f(X)
Y = 0.75X + 0.0042X2 – 0.000023X3 (4.5)
Π = Py.Y – Px.X – TFC

If the profit function is differentiated with respect to the variable input then, the optimal level of
the input could be determined.

dΠ/dX = Py.dY/dX – Px = 0 (4.6)

= Py..MPP - Px = 0
or
Py.MPP = Px (4.7)
VMP = Px
VMP = Slope of the TVP curve
Px = Slope of TC curve

Equation (4.7) can also be written as:

MPP = Px/Py (4.8)

This is the input criterion for profit maximization. It expresses both TVP and TC in terms of X
Using this criterion, one can establish the optimal level of input and the corresponding level of
output. Profit is maximized when the two slopes are equal in the second stage of the classical
production function. Thus, the VMP curve after the point of maximum AVP is referred to as the
derived supply demand curve for the variable input (Why?)

Px
SHS Derived Supply
curve for
Variable Inputs

Optimum
VMP AVP
level of X
0
Px
X1 X4

Figure 4.4 Optimum level of Input using Value of the Marginal Product Criterion

44
Continuing with our example:

Y = 0.075X + 0.0042X2 – 0.000023X3 (4.9)

dY/dX = MPP = 0.75 + 0.0084X – 0.000069X2 (4.10)
∴VMP = Py (0.75 + 0.0084X – 0.000069X2 ) = Px (4.11)

Thus, if the price of X and Y is known, the optimal level of the input and output can be
computed. (Solve for X in equation 4.11).

Choice of Decision Criteria: Maximum Profit versus Maximum Yield

When we determine the optimum level of variable inputs, we assume that the farmer had made
the decision to pursue maximum profit as their goal. However, sometimes farmers may want to
pursue other goals such as maximum output for reasons that are best known to them. What is
important to note here is that, unless inputs are free, the point where profit attains a maximum
(X1) is not the same as the point at which output is maximized (X2). This is because the
efficiency of using variable inputs declines in stage II of the production function. Beyond the
point of maximum profit, the added inputs cost more than they earn, such that pursuing
maximum output may lead to lower net returns or profit. However, of inputs are free, both profit
and output are maximized at the same point, (X2).

TPP

MPP = Px/Py

Iso-profit line
Relative magnitude of
profit before fixed cost

X
X1 X2
Figure 4.5 Optimum level of input Using TPP Curve and Choice Indicator

45
 Determining the Optimum level of Output (Output Criterion) A

SHS TR

TC
Break even point

Fixed
Cost

Y1 Y
B
SHS
Break even point

-FC

0
Y1 Y
SHS ATC Break even
C
point
AFC ATC = Py Point of profit max
MC = Py
AVC
Py

MC

0 Y1 Y

Figure 4.6: Determining Optimum Amount of Output Using cost and Revenue Curves

46
The optimum level of output is that level of output, which corresponds to the maximum profit.
As is the case with the optimal level of input, optimal output can also be determined from a table
(e.g. Table 4.1), graphically or from a mathematical equation. Graphically, determination of
optimal output is Y1 as illustrated in figures 4.6 A, B and C.

Since farmers are assumed to sell in a purely competitive markets, then Py is constant (why?) and
therefore TR is a straight line with a slope of Py. Maximum profit occurs where the TR>TC and
the difference between them is greatest. In a purely competitive market, the product price (Py) is
also equal the Average Value Product (TVP/Y = Py.Y/Y = Py = AVP)

Profit = TR – TC
= Py.Y – Px.X – TFC
= Py.Y – Pxf-I.(Y) – TFC (4.12)

The inverse function is used to express input X as a function of Y (in stages I and II)

Differentiating equation 4.12 with respect to output and equation the derivative to zero (first
order condition for profit maximum)
dΠ/dY = Py – Px .dX/dY = 0
= Py – Px/MPP = 0
∴ Py = Px/MPP or
Py = MC

This is the output criterion for profit maximization. It may also be expressed as follows:

dΠ/dY = dTR/dy – dTC/dy = 0

∴ dTR/dy = dTC/dy
MR = MC

If MC < MR, the farmer could always increase profit by increasing the use of more variable
inputs until the point at which profits are at maximum is attained. This point occurs in stage II of
the classical production function. However, a farmer may not have enough funds to purchase all
the inputs required for profit maximization. In such a special case, the farmer could operate in I,
to the left optimal level of input, as long as net returns are positive.

47
Comparison of Input and Output Criteria for Profit Maximization
Input Criterion Output Criterion
MR = MC
VMP = MFC Py = Px/MPP
Py.MPP = Px Py = Px/dY/dX
Py.dY/dX = Px Py = Px.dX/dY
Py.dY = Py.dX Py.dY = Px.dX

Where VMP = Value of the marginal product = value of product from the last additional unit of
variable input used in the production process
MFC = Marginal factor cost = value of the last unit of variable input used
MR = Marginal revenue = Added return from the last additional unit of product produced
MC = Marginal cost = added cost from the last additional unit of product produced
MPP = Marginal physical product

Based on the graphic presentation and mathematical derivation above, the following points it is
important to note the following points:
• MC always intersect the AVC and ATC curves at their lowest points

• When Py is lower than the AVC, the producer is losing is losing all fixed costs and part of
variable costs

• The optimum output always occurs where Py intersect MC curve, a point above the AVC
curve. This is the portion of the MC in region II of the production function.

• In the short run, the producer will use input as long as the product price is equal to or greater
than the average variable cost. That is, as long as they are able to cover variable costs.

• As the product price increases or decreases, the optimum level of input increases or decreases
along the MC curve.

• The marginal cost curve above the average variable cost is considered as the supply curve for
the producer.

48
SHS
MC AVC Derived supply
curve for individual
producer

Py

0 Y
Y1 Y6

Figure 4.7 Derived Supply curve for Individual Producer

In the short-run a farmer will produce as long he/she covers his/her variable costs. The producer
adjusts the quantity of the product he/she produces depending on the prevailing or anticipated
price as indicated in Figure 4.7. For this reason, the marginal cost is equated to the product
price, in the region where the MC lies above the AVC (MC ≥ AVC). This portion of the MC
curve is referred to as the derived supply curve for the individual producer. It is important to
note that in a purely competitive market, the individual producer (farmer) is unable to influence
the market price for inputs or output.

Thus, if the firm’s cost structure is such that the minimum AVC is higher than the prevailing
market or equilibrium price , then, the firm will be driven out of business, because they will not
be able to meet their variable costs in the short run. Similarly, a firm that attempts to vary all its
inputs (variable and fixed), and operates where MC = MR is said to be operating in the long run.
. In the short run, a firm may continue to produce even if then do not cover all the fixed costs
(AVC < MC). However, this cannot go on indefinitely. In the long run, both variable and fixed
costs must be met (ATC ≥ MC). A firm that cannot meet its long run variable and fixed costs
will eventually be driven out of business (in the long run).

If a firm is operation at Py = MC = ATC, it implies that all costs are met and the managers’
return is exactly that required to keep him/her in business in the long run. When Py > ATC, the
profit is accruing to the firm. This excess profit or RENT accrues only in the short run. In the
long run, other producers will be attracted by the RENT to enter the industry. This will increase

49
 production (rightward shift of supply function), which will drive the price down up to the poing
where Py = MC = ATC, where rend is driven down to zero and no super profits exist.

Short Run Equilibrium

A firm or producer operating at the point where MC = MR and MC ≥ AVC is said to be in

equilibrium. If there are any fixed costs, which cannot be varied during the production period,
the firm is said to be operating in the short run. The equilibrium is likewise referred to as the
short run equilibrium. Suppose the market price for the product rises from OR to OS, as a result
of a right ward shift in the market demand from DD to D’D’. This shift may be induced by
several factors including changes in taste, income or the price of related goods
Individual producer supply curve Industry Supply Curve
Cost/ S
D D
Price
MC ’
ATC
S

R
AVC

D D’
S

O
L N Output (Y) R T Output (Y)

Figure 4.8 Producer and Industry Short Run Supply Curve

The profit maximizing firm will increase output from OL to ON where the marginal revenue
(MR) is equal to the marginal cost (MC). Based on similar arguments, changes in output by the
individual producer, in response to price changes, will trace a locus of points along the MC
curve, which effectively becomes the individual producer’s supply curve in the short-run, as
depicted in figure 4.7 and 4.8.

50
The supply curve for the industry is a horizontal summation of all the supply curves for the
individual producers or firms. Thus, the industry supply can be deduced from the individual
producers’ supply curves as illustrated in Figure 4.8.

Derived Demand for Inputs

Just as the marginal cost curve above the AVC is the short run supply curve for the individual
producer, likewise, the value of the marginal product curve (VMP) can be interpreted as the
derived demand curve for variable inputs for the individual producer. This is based on the input
criterion for profit maximization derived earlier, where at maximum profit;
VMP = PX
As the price of the variable input changes, the point at which PX = VMP would trace a locus of
points, which on the horizontal axis represent the amount of input which would be demanded by
the producer at that price. Thus as shown in Figure 4.9, the VMP may be viewed as a derived
demand curve for that variable input. This occurs in stage II or the classical production function,
from the point where the APP is maximum up the point where the MPP is equal to zero.

VMP

PX4
PX3

PX2 VMP

AVP
PX1

O
X
X1 X2 X3 X4

Opportunity Cost
The opportunity cost of a resource is defined as the return that a resource can earn when it is put
to its best alternative use. For example a farmer with 100 Kg of CAN fertilizer can apply it to
one ha. Of maize or one ha of Irish potatoes. Suppose from maize the farmer will get an
additional T.Shs 40,000of maize while from potatoes the farmer will get T.Shs 45,000 worth of

51
additional potatoes. If the farmer decides apply the fertilizer he/she foregoes 40,000 T.Shs worth
of additional maize. Thus the opportunity cost of the fertilizer is T.Shs 40,000. The opportunity
cost of a shilling spent on consumption for example is the interest it could earn if deposited in the
bank.

Concluding Comments
Profit maximizing conditions for the firm have been derived. Profits are maximum when the
level of output is chosen such that MC is equal to MR. The cost function is the inverse of the
production function that underlies it. The firm product supply curve can be derived from the
equilibrium MC = MR condition and is represented by the marginal cost curve above the average
variable cost curve. The firm’s demand curve can be derived from the equilibrium condition
VMP = Px

52
5 INPUT MIX DECSIONS (FACTOR – FACTOR DECISIONS)
Until now, we have presented and derived decision criteria for a profit maximizing firm manager
as if production involves only one variable input. Often however, the firm manager must
simultaneously decide on the amount to be used for more than one variable inputs. In this section
we will discuss criteria that are used for input mix decisions. We will maintain some of the
assumptions introduced earlier where we has only one variable input. The assumptions include
the following:
• Prices of inputs and output are determined by forces of supply and demand, which are
outside the farmer’s control. As such, input and product prices are taken as given.
• Farmers/managers buy inputs and sell their products in perfectly competitive markets
• In the production process, at least one of the inputs is fixed. This implies that the farmer is
operating in the short run where the law of diminishing returns applies. NOTE that if all
inputs are varied, the law of diminishing returns does not hold.

When only one variable input is used, for a given technology, a given level of output can be
produced only in one way (remember, there is a one to one relationship between inputs and
output in a production function). When there are two or more variable inputs, a given level of
output can be produced using different combinations of the variable inputs. This is because
substitution occurs between some factors of production. The main problem to the firm manager
is to find the right combination of inputs so as to optimize the manager’s goal or decision
criteria.

Production function with two variable inputs

Suppose production function involves two variable inputs as follows:

Y = f(X1, X2) (5.1)

or
Y = f(X1, X2x3, x4……….xn) (5.2)

53
Table5.1 Hypothetical Maize Response to Phosphate and Potash Fertilizer
Potash (Lb/acre)
Phosphate
(Lb/acre) 0 10 20 30 40 50 60 70 80
0 96 98 99 99 98 97 95 92 88
10 98 101 104 104 105 104 103 101 99
20 101 104 108 108 109 110 110 109 106
30 103 107 114 114 117 119 120 121 121
40 104 109 117 117 121 123 126 128 129
50 104 111 121 121 125 127 129 131 133
60 103 112 123 123 126 128 130 131 134
70 102 111 123 123 126 127 131 136 135
80 101 108 119 119 119 125 129 131 134

Based on Table 5.1 the following facts may be deduced:

• To some degree phosphate can substitute for potash maintain a given level of production
• Different combinations of potash and phosphate can produce the same level of maize
• Depending on the relative prices of the two inputs, a farm manager must choose the
combination of inputs, which maximizes profit.

An Isoquant
Consider a production function given by the mathematical function;
Y = 18X1 – X12 + 14X2 – X22
An isoquant is defined as the line that connects all the locus of points for different combinations
of the variable inputs (X1 and X2) which produce the same level of a given product (Y). The
word isoquant is derived from tow Greek words (iso = equal and quant = quantity).

Note that with the exception of the maximum and minimum level of production, all other output
levels can be produced using several combinations of inputs.

Conventionally, isoquants are represented graphically in two dimensions as in Figure 5.1.

However, if there are more than two imputs the isoquant would represent a surface.

54
 105
X1 115

Isoquant

Isoquant
Map

X2

Figure 5.1 An Isoquant and An Isoquant Map

Mathematically, an isoquant can be derived by solving for one of the variable inputs (X1) as a
function of the other variable input (X2) and Output (Y).

From the production function given earlier,

Y = 18X1 – X12 + 14X2 – X22 (5.3)
2 2
0 = 18X1 – X1 + 14X2 – X2 – Y
-18X1 + X12 = 14X2 – X22 – Y
81 – 18X1 +X12 = 81 + 14X2 – X22 – Y
(9 – X1)2 = 81 + 14X2 – X22 – Y
9 – X1 = [81 +14X2 – X22 – Y]½
-X1 = -9 + [81 +14X2 – X22 – Y]½
X1 = 9 - [81 +14X2 – X22 – Y]½ (5.4)

An isoquant can be defined for every output level or conversely, every output has an isoquant

55
Properties of Isoquants
Isoquants have two main properties, which are:
• Isoquants never cross each other. If they did they would violate a basic assumption of
production functions, that is; each combination of inputs can produce only one level of
output. However, many combinations of the variable inputs can result into a given level of
output. In fact, the entire loci of points along an isoquant represents different combinations
of the variable inputs for a given level of the product.
• Isoquants must be convex or quasi-convex to the origin. A convex isoquant implies that
production is within stage two of the (classical) production function, and the law of
diminishing returns applies. If the isoquants were concave to the origin, decision makers
would not be able to reach optimal decisions (maxim or minimum), which means, conditions
for economic efficiency would be violated.

The Marginal Rate of (Input) Substitution (MRS)

The slope of an isoquant is referred to as the marginal rate of substitution (MRS). Some authors
call it the marginal rate of technical substitution. The MRS is a measurement of how one input
substitutes for another as one moves along an isoquant to maintain the same level of production.
In other words, the MRS is defined as the amount by which one input (X1) must be reduced to
maintain the same level of production when the other input (X2) is increased by one unit.
When X1 substitutes for X2, the slope of the isoquant is expressed as MRSX1,X2, to represent the
fact that X1 is increasing while X2 is decreasing. If X2 was increasing to replace X1 then the
inverse slope of the isoquant would be represented as MRSX2,X1. Isoquants are convex to the
origin, which means they are negatively sloped

The MRS may also be represented as:

MRS = ∆X1/∆X2 or ∆X2/∆X1, depending on which variable is increasing or decreasing
MRS = dX1/dX2 or dX2/dX1

Isoquants exhibit a diminishing marginal rate of substitution. This means, as X1 substitutes for
X2 for example, each incremental unit of X1 (∆X1) replace less and less units of X2 (∆X2) as
depicted in figure 5.2. The diminishing MRS for isoquants is directly related to the diminishing
marginal product for each of the variable inputs. Each production function has its own

56
 corresponding isoquant map. A convex isoquant implies production is taking place in stage II
and therefore, diminishing marginal productivity

X1

∆X1

∆X2

X2
Figure 5.2 Illustration of Diminishing MRSX2,X1

Average and Exact MRS

The average MRS is derived when only two coordinates of an isoquant are defined. The average
MRS is represented by the slope of the straight line connecting the arch between those two
coordinates. The exact MRS on the other hand is represented by the slope of the tangent to the
isoquant at the specific point of interest. The exact MRS can only be obtained if a continuously
differentiable isoquant is defined.
X1 (2,9) X1
B
(a) (b)

∆X1

A (3,6)
Y = 105 (3,6)
Y=96
∆X2 C

X2 X2
Exact MRSx2,x1 = dX1/dX2
Average MRSx2,x1= ∆X1/∆X2

Figure 5.3 Graphic Representation of Average an Exact MRS

57
Using the mathematical expression of an isoquant developed earlier (Equation 5.4), the MRS can
be derived as follows:

X1 = 9 - [81 +14X2 – X22 – Y]½

Average MRS:
Assume that the following coordinates of the isoquant are given;
(X2,X1) = (2,9) and (X2,X1) = (3,6)
The Average MRSX2,X1 = ∆X1 9–6 3
----- = ------ = --- = -3 (5.5)
∆X2 2-3 -1

The MRS is negative because isoquants are negatively sloped. For a convex isoquant, the
absolute value of the slope changes continuously through the entire range of the isoquant,
consistent with diminishing MRS as defined earlier.

Exact MRS:
This is defined by the line that is tangent to the isoquant. If the isoquant given in equation 5.4
represents the output level Y = 105 units, the isoquant can be re-written as;

X1 = 9 - [81 +14X2 – X22 – 105]½ (5.6)

= 9 – [14X2 – X22 –24]½
dX1/dX2 = ½[14X2 – X22 – 24]-½.(14 – 2X2) (Using the chain rule) (5.7)

(14 – 2X2)
MRSX2,X1 = -------------------------
½[14X2 – X22 – 24]½

= - (28 – 4X2)
-------------------------
[24 - 14X2 + X22]½ (5.8)

58
The MRS can also be computed using the marginal physical product of the respective variable
inputs.
Consider the production function:
Y = f(X1, X2)
∆Y/∆X1 = MPPX1 (5.9)
∆Y/∆X2 = MPPX2 (5.10)

Now, along an isoquant, the level of Y is constant, ∴ ∆Y = 0

This implies that:
[∆Y/∆X2]
----------- = [∆Y/∆X2][ [∆X1/∆Y] = ∆X1/∆X2 = MRSX2,X1 (5.11)
[∆Y/∆X1]

but,
∆Y/∆X2 = MPPX2 and ∆Y/∆X1 = MPPX1 (5.12)

Therefore, the MRS may also be expressed in terms of MPP as in equation (5.12).

MRSX2,X1 = ∆X1 = MPPX2

----- ---------- (5.13)
∆X2 = MPPX1

Apply this approach to our earlier production function given under Equation 5.3, and the
corresponding isoquant as given under equation 5.4, one may compute the exact MRSX2,X1, for
the coordinate (X1,X2) = (3,6) as follows;
Y = 18X1 – X12 + 14X2 – X22
MPPX1 = 18 – 2X1
MPPX2 = 14 – X2
∴MRSX2,X1 = 14 – 2X2 14 – 2(3) 8 4
---------- = ----------- = --- = --- (5.14)
18 – 2X1 18 – 2(6) 6 3

It was state earlier that a production function is a technical relationship between inputs and
output to guide production decisions. Isoquants are derived from production functions, and they
serve the same purpose. Movement along an isoquant is done by the manager during the
planning process to select the combination of inputs which will yield the highest level of profit.
That combination of inputs is referred to as the optimum combination of inputs.

59
 5.2 Relationship Between inputs
We said earlier that being convex or quasi-convex to the origin is a general characteristics of
isoquants. However, the exact shape of the isoquant map depends on the manner in which the
inputs substitute for each other in the production process. In this respect, inputs may substitute
for each other at a decreasing rate, at a constant rate, or in fixed proportions. The isoquant map
is therefore related to the nature of input substitution, which is in turn related to the underlying
production function.

Inputs as Technical Substitutes

Inputs are said to be technical substitutes when an increase in one input permits a decrease in the
other input while maintaining the same level of output. Such inputs are said to compete for each
other and their MRS is negative,

Technical substitutes with Decreasing MRS

If two inputs are technical substitutes, and the absolute value of the MRS decreases as one of the
variable inputs increases while the other input decreases, then the rate of substitution is said to be
decreasing (See illustration in Figure 5.2). Decreasing rates of substitution are caused by the law
of diminishing returns.
MRSX2,X1 = MPPX2/MPPX1 (5.15)
When diminishing returns are present,
• MPPX2 decreases as X2 increases
• MPPX1 increases as X1 decreases
This means the absolute value of the ratio in Equation 5.15). Examples of isoquants with
decreasing MRS are shown in Figure 5.4
X1
X1 X1
(a) (b) (c)

X2 X2 X2

Figure 5.4 Decreasing Rate of Substitution

60
 • In Figure (a), output can be produced using either X1 or X2 or a combination of the two
inputs.
• In figures (b) production of output requires both inputs. If too much of one input is used,
increasing amounts of the other input must be applied to maintain output at the same level.
The zone of economic relevance is the portion, which has a negative slope.
• In figure (c), the isoquants become asymptotic to the axes as the variable inputs increase
indefinitely. Beyond a certain unit, the absolute value of the MRS becomes very large since
large quantities of the second input are required to replace small quantities of the first input
in order to maintain the same level of output production.

Constant marginal rate of substitution

This occurs when the rate at which inputs substitute for each other remain constant throughout
the entire range of the isoquant. Constant rate of substitution requires that the slope of the given
isoquant remains constant (unchanged). This means the isoquant is a straight line with a constant
slope, but, the isoquants do not have to be equidistance nor do they have to be parallel. In
economic terms, inputs that substitute at a constant rate are referred to as “perfect substitutes”

X1 (a) (b) (c)

X1 X1

X2 X2 X2

Figure 5.5 Constant Rate of Substitution

61
 Complementary Inputs
Inputs that increase the level of output only when they are combined in certain fixed proportions
are called technical complements. Complementary inputs represent the opposite of substitutes.
Some examples of complementary inputs in agriculture include: (1) parts which make a machine
e.g. a tractor tires, (2) A tractor and a tractor operator, (3) fence and wire posts. (4) hydrogen
and oxygen to from water (H2O). In the case of complementary inputs, adding more of one of
the inputs, beyond the required proportion will not reduce output, but it will not increase it either.
The level of output will remain unchanged. Thus the input combination may be considered as
one input (eg tractor and tractor driver), which may substitute for other combinations

X1 X1

X2 X2

Figure 5.6 Complements (No Substitution)

When isoquants degenerate into a dot, it means only one combination of inputs may be used to
produce that given level of output.

Lumpy Inputs

Inputs that are not divisible are referred to as lumpy inputs (eg a tractor). Such inputs appear
only in discrete units. If both inputs are lumpy (tractor and driver), then the isoquants would not
be continuous. Points would be connected by dotted lines as in Figure 5.7.

62
 X1

X2

Figure 5.7 Lumpy Goods

Elasticity of Factor Substitution

The elasticity of factor substitution (εs) in factor-factor relationship is defined the ratio of the
percentage change in one input to the percentage change in the other input. [Compare this with
the elasticity of production (εp) defined earlier with respect to product-product relationship].
Thus;
∆X1 ∆X2 ∆X1 X2 X2
εp = ----- ÷ ----- = ------ × ---- = (MRSX2,X1) ----- (5.16)
X1 X2 ∆X2 X1 X1

= MPPX2 X2
---------- × ---- (5.17)
MPPX1 X1

Average and Exact Elasticity of Factor Substitution

As is the case with average and exact MPP or MC, the average elasticity of substitution is
derived when only two coordinates of an isoquant are defined. For instance, the average εs at the
arc between the coordinates ((X2,X1) = (4,5) and (X2,X1) = (3,6) would be:
5–6 3.5 Midpoint of 3 and 4
εs = ----- × ---- = - 0.64 (5.18)
4–3 5.5 Mid point of 5 and 6

63
 In principle, this is the elasticity of the line connecting the arc between the two coordinates.

The continuous elasticity of factor substitution is defined when a continuously differentiable

isoquant is defined, such that the MPP of both variable inputs can be computed and substituted
into the elasticity of factor substitution equation as in Equation 5.17.
Note the following important points;
• The elasticity of factor substitution is negative for inputs that are technical substitutes
• The elasticity of factor substitution is zero for inputs that are complements
• The elasticity of factor substitution is unitless
• Input substitution is meaningful in a production process when (1) the substitution increases
output or reduce cost and (2) the MPP of both inputs are positive (eg when a farmer
substitutes labour for land or machinery for labour or fertilizer for land)
• The fact that inputs are economic technical substitutes does not imply that they perform the
same technical or physiological function, only that input may be combined in different
proportions to produce different levels on a product (eg. Nitrogen fertilizer and plant
population in the production of maize or any other crop for that matter).

Isocost Lines
Each combination of inputs has a cost associated with it. The costs are variable because variable
inputs are involved. Suppose the per unit cost (or price) of X1 = PX1 and that of X2 = PX2, then
the Total cost (TC) is given as;
TC = TFC + TVC
TVC = PX1.X1 + PX2.X2 (Assume total fixed cost is zero; TFC = 0)

X1
Isocost line for TVC = 100

Isocost line for TVC = 80

X2

Figure 5.8 Isocost Line

64
An isocost line traces all the locus of points on the cost surface having the same cost. In other
words, isocost lines determine all combinations of inputs that cost the same amount. An isocoost
line may be thought of as a budget constraint. That is the maximum amount of money available
for the producer to purchase inputs. Note that isocost lines correspond to total cost (TC)
functions, just as isoquants relate to production functions.

If the prices of variable inputs are known, the TVC can be computed for all input combinations.
Based on this, one can derive the isocost line as follows;

TVC = PX1X1 + PX2X2

PX1X1 = TVC - PX2X2
TVC PX2X2
X1 = ------ - --------
PX1 PX1
or

TVC PX2
X1 = ------ - ----- X2 (5.19)
PX1 PX1

Note that for any given combination of inputs, TVC, PX1 and PX2 are constant
This implies that in equation 5.19, the terms, TVC/PX1 and PX2/PX1 are constants. The first term
is the vertical intercept while the second is the slope of the isoquant (Figure 5.9)

X1 TVC
------- , 0
PX1

TVC = PX1X1 + PX2X2

TVC
0, -------
PX2

O X2
Figure 5.9 Determining Coordinates and Slope of Isoquant

65
The slope of the isocost line may be determined from Figure 5. As follows;

TVC TVC PX2

0 - ------ ÷ ------ - 0 = - ------ (5.20)
PX1 PX2 PX1

The two important aspects of an isocost line are it distance from the origin and its slope.
Changes in the TVC when input prices are constant shift the isocost line up or down accordingly.
Changes in the input prices change the slope of the isocost line as more of the cheaper input will
be substituted for the more expensive input.

THE LEAST COST CRITERION

We said earlier that producers strive to maximize profit as their primary objective. However,
sometimes cannot access all the variable inputs in quantities that are adequate to maximize
profit, often because they face budgetary constraints. Now if a producer has a given outlay of
funds to purchase inputs, they would like to produce the output using those inputs at the least
cost possible. That combination of inputs, which minimizes the cost of producing a given level
of input is referred to as the least cost combination (LCC) of inputs.

Graphically, for isoquants that are convex to the origin, the least cost combination occurs at the
point where the isoquant is tangent to the isocost line.

X1
Isocost line

Least Cost Combination

(X2,X1)

Isoquant

X2
Figure 5.10 Least Combination of Variable Inputs
At the point of tangency, slope of isoquant = slope of isocost line = - PX2/PX1

66
Algebraically;

We know that the slope of the isoquant = MRSX2,X1 = -  MPPX2/MPPX1 

We also know that the slope of the isoquant = - PX2/PX1

Therefore at the point of tangency where the slope of the two curves are equal;

∆X1 MPPX2 PX2

MRSX2,X1 = ----- = --------- = - ------
∆X2 MPPX1 PX1
∴

MPPX2 MPPPX1
At the LCC, --------- = ---------- (5.21)
PX2 PX1

Equation (5.20) is the criterion for LCC when there are two variable inputs. Equation (5.21)
implies that in the presence of a budgetary constraint (TVC = PX1X1 + PX2X2), the product Y will
be produced at the least cost if the last shilling spent of X1 and X2 returns the same number of
physical units of X1 and X2 In the case of n variable inputs, equation (5.21) can be generalized
further as follows;
MPPPX1 MPPPX2 MPPPX3 MPPPXn
At the LCC, ---------- = -------- = -------- =………….= -------- (5.22)
PX1 PX2 PX3 PXN

Another way of looking at this is as follows, as X1 and X2 substitute for each other along an
isoquant one of them will be negative (decrease) as the other is positive (incerease)
At the LCC, ∆X1/∆X2 = - PX2/PX1

∴ -PX1.∆X1 = PX2.∆X2

Now if the level of production along a given isoquant is not at the LCC, it will mean that either,

-PX1.∆X1 > PX2.∆X2 or -PX1.∆X1 < PX2.∆X2

The LCC will be obtained by either reducing X1 while increasing X2 or vice versa

67
SPECIAL CASES OF LEAST COST COMBINATIONS

Although we have said at the LCC the isocost line is tangent to the isoquant, this is only true for
isoquants that are convex to the origin. In this case the following conditions are met;
• Inputs are infinitely divisible (isoquants are smooth i.e continuous and differentiable). They
are not represented by dots
• The LCC requires that both inputs should be used
• The MRS is decreasing rather than being constant
However, there are other special cases where tangency does not exist but the LCC may
nevertheless be established. These include;

(1) LCC for Lumpy goods

By definition, the isoquant for lumpy goods is a set of points for certain input combinations, and
these points are connected by dotted lines because the isoquant is not defined everywhere. This
implies that the MRS is not defined, as such a point of tangency where MRS = slope of isocost
line does not exist. Nevertheless, the LCC may be geometrically determined as illustrated in
figure 5.11.

X1

Isoquant

E = LCC
Isocost line

X2

Figure 5.11 LCC for Lumpy Inputs

The LCC solution (E) for a lumpy good is often stable because it remains range for some range
of the input prices ratio. It requires a large shift in input prices for the LCC of a lumpy good to
change from E to another point.

68
 (2) Corner Solution
(a) Convex isoquants
Generally, convex isoquants are asymptotic to the axes and they require that both inputs should
be used in order to produce a given level of output. However, sometimes an isoquant that is
convex to the origin intersects the axes, implying that production of the output can take place by
using only one of the variable inputs (in the absence of the other). In this case, depending on the
relative prices (price ratio)s, the point at which the isoquant touches the isocost line may be at
one of the axes. While this is not a point of tangency, it nonetheless defines a point of least cost
for that price ratio. This LCC is referred to as a corner solution since it occurs at a corner of the
isoquant as illustrated in Figure 5.12. If the isoquants intersect the axes, the LCC may occur at
one of the access

X1 Corner solutions

Isocost lines for

different price ratios

X2

Figure 5.12 Corner Solution for Convex Isoquant

(b) Linear Isoquants

Linear isoquants also intersect both axes, therefore, the same argument advanced for a corner
solution in (a) applies for linear isoquants, which either have a corner solution or an infinite
number of LCC as illustrated in Figure 5.13.

I II
X1
X1 Corner solution for Isoquant coincides with
isocost lines of different isocost line ∴infinite LCC
slope

69
 X2
X2
Figure 5.13 Corner Solution for Linear Isoquant Infinite Least cost Combinations
Mathematical Example
Y = 18X1 – X12 + 14X2 – X22
MPPX1 = 18 – 2X1
MPPX2 = 14 – 2X2
If PX1 = 2 and PX2 = 3
MPPX2 PX2
Then, MRS = ---------- = -----
MPPX1 PX1

14 – 2X2 3
= ----------- = - -----
18 – 2X1 2

28 – 4X2 = 54 – 6X1
14 – 2X2 = 27 – 3X1
3X1 – 13
∴ X2 = -----------
2
If the isoquant represented is for Y = 105,
Then, the LCC can be solved for by substitution
Y = 18X1 – X12 + 14X2 – X22
105 = 18X1 – X12 + 14(3X1 – 13)/2 – (3X1 –13)2/4
= 18X1 – X12 + 7(3X1 – 13) – (9X12 – 78X1 + 169)/4
Using the quadratic equation, the LCC can be solved for and the value of X1 = 6.2 while that of
X2 is 2.8

Isoclines, Expansion Paths and Profit Maximization

There are other definitions that are related to curves, which are important for decision makers in
the course of maximizing profit. These are illustrated in Figure 5.14

70
Isocline
The isocline is a curve that passes through points of equal marginal rate of substitution. Thus an
isocline will pass through all isoquants.

Ridge lines
Isocost line
X1
Point of global
output
Isoquants maximum
Ridge lines

Point of global
profit maximum

Least Cost
Combinations

Isoclines

X2

Figure 5.14 Isoclines and Expansion Paths

Expansion Path
An expansion path is a special kind of an isocline, for a given input price ratio (which implies a
specific slope of the isocost line), the expansion path is the isocost line that connects all points
with the least combination of inputs. This means, along the expansion path the slope of the
isocost line is equal to that of the isoquant. MRS X2,X1 = - (PX2/PX1. This is a necessary
condition for economic efficiency. Just as there is an isoquant for each output level, there is an
expansion path for each price ratio.
The following points may be noted with respect to the expansion path;
• If inputs are technical substitutes and the MRS is decreasing, any change in input prices such
that the price ratio changes will shift the expansion path.

71
• If the expansion path is a straight line from the origin, then the inputs will be used in the
same proportion along the entire expansion path, that is, at all levels of output
• If the expansion curve is curved, then the proportion of inputs will vary along the expansion
path, that is, it will be different at different levels of output.
The mathematical equation for the expansion path my be derived as shown below;
PX2
MRS = ------
PX1

MPPX2 PX2
-------- = -----
MPPX1 PX1

If the equation for the production function is known, then the MPP for each input can be derived,
and then the variable inputs can be expressed one in terms of the other.
Example;
Y = 18X1 – X12 + 14X2 – X22
MPPX1 = 18 – 2X1
MPPX2 = 14 – 2X2
X1
If PX1 = 2 and PX2 = 3
long the expansion path, the following relationship holds;
MPPX2 7 – X2 3
--------- = -------- = ---
MPPX1 9 – X1 2

3(9 – X1) = 2(7 – X2)

13 2
∴ X1 = ---- + --- X2 Expansion path, slope = 2/3
3 3
X2
Note the following points;
• On an isoquant map, there is an expansion path for every price ratio
• There are many (infinite) points of LCC along any expansion path
• However, at only one of the LCC points is profit maximized
• Thus LCC is a necessary condition for profit maximization, but it is not sufficient. (What is
the sufficient condition?)

72
Ridge Line
The ridge line is a curve that connects all the locus of points where the MPP is equal to zero.
Thus the ridge line represent the limits of economic relevance because they represent the
maximum output that can be produced from each variable input given a fixed amount of all the
other input. The ridgeline is therefore the boundary beyond which the isoclines and the isoquant
map cease to have economic meaning. Outside this boundary, the necessary conditions for
economic efficiency are not met.

Profit Maximization and the Optimum Combination of Inputs

So far we have used graphics to determine the least cost combination of inputs, which include
the point at which profit is maximized. However, our primary goal is to derive decision criteria
for a profit maximizing firm manager. Let us therefore develop more general criteria for profit
maximization. Several methods may be used;

(1) Direct Methods

(a) Equi-marginal value product criterion

(2) This method is done by maximizing profit directly, a profit function (π) is defined, which is

then maximized with respect to the variable inputs

π = PY Y - PX1X1 - PX2X2 - TFC (Assume TFC = 0)

Y = f(X1, X2)
δπ δY
---- = PY ----- - PX1 = 0
δX1 δX1

Thus, PY.MPPX1 = PX1

VMPX1 = PX1 (5.24)

Likewise,
δπ δY
---- = PY ----- - PX2 = 0
δX2 δX2

PY.MPPX2 = PX2
VMPX1 = PX2 (5.25)

73
Based on equations 5.23 and 5.24, we may say that the profit-maximizing criterion requires that
the earnings from each input must be equal to its cost, and this must be true simultaneously for
all inputs. This occurs when MPP < APP and both are decreasing, i.e. In stage II of classical
production function.

This may be generalized further as follows;

From equations 5.23 and 5.34;

VMPX1 VMPX2
--------- = 1 and ----------- = 1 (5.25)
PX1 PX2
In more general terms, this may be expressed in general terms for more than two inputs as;

VMPX1 VMPX2 VMPX3 VMPXN

---------- = ---------- = --------- =… = ---------- ≥ 1 (5.26)
PX1 PX2 PX3 PXN

Equation 5.26 implies that all variable inputs must be earning as much as they cost on the
margin. Equation 5.25 also implies that;
MPPX1 = PX1/PY
MPPX2 = PX2/PY

(b) The optimum level of output may be established by estimating the average variable cost and
the marginal cost curves using empirical data. Then, equate the MC = PY to determine the
profit maximizing level of output (optimum Y) as illustrated in Figure 5.16. We previously
determined that using the output criterion, profit is maximized where MC = MR = PY
∴ Optimum output level = YO

Shs MC
AVC

PO

74
 Y
YO

Figure 5.16 Empirical Determination of Optimum Output

The MC may also be derived from the production function since the TVC is inversely related to
the production function (see Doll and Orazem, Chapter 4, for details). Then MC is equated to
PY to find optimum output.

(3) Indirect Method

(a) Least Cost Criterion (Cost Minimization)
We have already established that the least cost criterion for profit maximization is that the MRS
should be equal to the slope of the isocost line.

MPPX2 PX2 MPPX2 PX2

---------- = - ---- = --------- = - -----
MPPX1 PX1 MPPX1 PX1

This may be written in general terms for more than two inputs as follows;

MPPX1 MPPX2 MPPX3 MPPXN

---------- = ---------- = ---------- = ………= ---------- (5.27)
PX1 PX2 PX3 PXN

Thus by the cost minimization criterion, profit is maximized when the ratio of the MPP for each
variable input to its corresponding input price is equal across all inputs. The cost minimizing
combination of inputs is always the LCC for profit maximization, but the converse is not
necessarily true. For instance, in some cases, an isoquant may be defined at (X2,X1) = (0,0),
While this would qualify as a cost minimizing combination of inputs, it would not result in any
profit for the producer since no production would take place.

75
Note that if the numerator of each term in equation 5.27 is multiplied by PY, then equation 5.27
and 2.26 become identical.

5. OUTPUT MIX DECISIONS

So far we have been deriving various decision criteria for a profit maximizing firm manager as if
they are involved in producing only one output. However, in practice, particularly for farmers,
they must also choose what to produce from a range of alternative agricultural products. Such
decisions are guided by the relationship that exists between products as defined below.

Assume that a farmer owns one acre of land and has 100 Kg of Nitrogenous fertilizer. The crops
that can grow in the area include maize, bean, sunflower, groundnuts, sorghum, millet, cassava,
Cow-peas and Pigeon peas. Suppose also that the farmer can only plant two crops at a time. This
farmer would be faced with choosing which two crops to produce.

(a) The Production Possibility Curve.

The production possibility curve (PPC) traces all combinations of products that can be produced
with a given set of inputs. It is also referred to as an iso-resource curve. It may be viewed as the
boundary or frontier delineating the combination of products that can be produced using
available inputs for a particular technology.

Y2

Y1*

Y2*

Y1
Figure 5.1 Production Possibility Curve
The following may be said about PPC
• The slope of the PPC is referred to as the Marginal rate of product substitution (MRPS)
• PPC that are based on the same bundle of inputs and technology do not cross each other
• PPC are either concave to the origin or linear depending on the underlying production
function.

76
• The entire area bounded by the PPC including the axes is called the feasible set or the
attainable set of output.
Product – Product relationship
The relationship between products or enterprises may take different forms. Products may be
competitive, complementary or supplementary to each other.
Competitive products
Products are said to be competitive if when the production of one increases, the production of the
other must decrease.

Y2 Y2

O
O
Y1 Y1
Figure 5.2 Competitive products
• The MRPS for competitive products is negative. (i.e. the PPC has a negative slope)
• A PPC that is concave to the origin represent a production function with decreasing marginal
returns. As increasing amounts of the variable input are transformed into one product, the
marginal product of the input in that use becomes increasingly smaller, while the marginal
product of the input for the other product becomes larger. Thus, each additional unit of the
product that is increasing requires successively larger sacrifice of the competing product.
• Linear PPC represent a constant rate of substitution between the two products (eg. maize
substituting for Irish potatoes on a 10 ha homogenous piece of land where ha is the variable
input).

Complementary Products
Two products are complementary if an increase in one product causes the second product to also
increase, when the total inputs are held constant. A common example of complementary
products is maize and beans. However, beyond a certain limit the products become competitive.

77
 Complementary products produce by-products that are in fact inputs in the production process,
thereby changing the bundle of inputs. For instance, legumes fix nitrogen form the air, adding to
the stock of nitrogen available for plants.
Y2
A Y2 C

B Y1 O
O Y1
Figure 5.4 Complementary Products
Within the range OB, OC and OD products Y1 and Y2 are complementary.
• The MRPS within the complementary range is positive
• Within the complementary range, one can only reduce the level of one of the products by
reducing the level of inputs used.
• Complementary product eventually become competitive (eg maize and beans plant
population)
• Changes in technology may cause the relationship of complementary products to change
from complementary to competitive. (eg. the introduction of nitrogenous inorganic fertilizer
has changed the relationship between legumes and other cereals to a competitive one with
respect to land, labour, capital etc.)
• Products may be competitive within one period of time but complementary over a longer
duration of time (eg. crops in a rotation cycle).

Supplementary Products
Two products are said to be supplementary if one can increase the production of one without
increasing or reducing the production of the other.

Y2 E Y2 F

78
 Y1 Y1
O O
H
Figure 5.5 Supplementary Products

• The MRPS for supplementary products is zero or undefined

• Supplementary products occur through time when or surplus resources are available at a
given point in time (eg. a tractor purchased for cultivation during the growing season may be
used for transportation during the dry season. A harvester may be used to harvest two crops
that mature at different times of the year, eg. sorghum and millet.
• Supplementary relationships between products depend on amounts of unused resources.

Joint Products
These are products, which result from the same production process. In extreme cases, the two
are combined in fixed proportions such that one cannot be produced without the other. In such a
case, joint products may be considered as one product. Other joint products may have a narrow
range of substitution. The proportions of the joint products may be changed through
technological developments, such as new breeds, new processing methods etc. Examples in
agriculture include skimmed milk and butter, mutton and wool.

Y2
Y2

O Y1 Y1

Figure 5.6 Production Possibility Curve for Joint Products

REVENUE MAXIMIZING COMBINATION OF PRODUCTS

79
The marginal rate of product substitution (MRPS) refers to the amount by which one product or
output must change in quantity when the other product increases by one unit along a production
possibility curve. The MRPS is the slope of the production possibility curve.
∆Y2
MRPSY1,Y2 = ----- = Slope of PPC (5.1)
∆Y1

• The average MRPS may be computed for the midpoint of a line joining an arch when only
two coordinates of a production possibility curve are defined. It is an approximation of the
MRPS of points on the arch
• The exact MRPS may be computed when a continuously differentiable mathematical
function of the PPC is given

As stated previously;
MRPS for competitive products is negative
MRPS for complementary products is positive
MRPS for supplementary products is zero or undefined.

Iso-revenue line
For a profit maximizing producer, maximizing revenue that is obtained from the combination of
products or enterprises is of paramount importance. Total revenue is the sum of all revenue from
all the products or enterprises.
TR = PY1.Y1 + PY2.Y2
Thus on a PPC, each combination of products is associated with a particular total revenue.
An iso-revenue line connects all combinations of products (Y1 and Y2) that earn the same level
of revenue. In Figure 5.21 an iso-revenue line is given for Y1 and Y2 given the respective
product prices of PY1 = 2 and PY2 = 1

Y2
(0,80)
A = 80/PY2
(10,60)
80
 (20,40)

(30,20)
(80,0)
O
Y1
B= 80/PY1
Figure 5.7 Iso-revenue line
The slope of the iso-revenue line may be determined from figure 5.21. If the product prices are
given, then the total revenue can be computed and the vertical and horizontal intercepts of the
iso-revenue line can be computed, based on which the slope may be determined.
0 - TR/PY2
Slope of iso-revenue line = -----------------------------
TR/PY1 - 0

∆Y2 TR PY1
----- = - ---- ÷ -----
∆Y1 PY2 TR

PY1
= - ----- (5.2)
PY2

Revenue Maximizing Combination of Products

PPC represent all possible combination of two products using a given amount of variable inputs
The challenge to the manager is how to determine the revenue maximizing combination. Within
a PPC the amount of inputs being used is constant, which means, costs are also constant. Thus
he combination, which will maximize revenue, will also maximize profit, for a given price ratio
between products. Using the analogy between isoquants and the iso-cost line, in the case of the
PPC, revenue would be maximized at the point where the slope of the PPC equals that of the iso-
revenue line. If the PPC is given in tabular form then the revenue from each combination of
products would be computed to determine the one that maximized revenue.

Isocost line with

Y2
slope = -(PY1/PY2)

81
 Revenue maximizing
combination of
Y1 and Y2
A

O
Y1

Figure 5.8 Revenue Maximizing Combination of Products

In Figure 5.8. at point (A) the PPC and iso-revenue line are tangent to each other

∆Y2 PY1
∴ ------ = - -----
∆Y1 PY2

∆Y2PY2 = - (∆Y1PY1)  (5.3)

Opportunity cost and the Marginal Criteria for Resource Allocation

Along the PPC the quantity of products that are produced is limited by the quantity of resources
available to the producer. Thus for any given level of inputs, the amount of one product that is
produced increase only if the other product is reduced as resources are shifted from the
production of Y2 to Y1. Denote the quantity of variable inputs shifted as ∆X. Then we may
rewrite Equation 5.3 as;
∆Y2PY2 = -∆Y1PY1
∆Y1 ∆Y2
PY1 -------- = PY2 -----
∆X ∆X

∴ PY1.MPPY1 = PY2.MPPY2
or VMPX,Y1 = VMPX,Y2
Generalizing to N products;
VMPX,Y1 = VMPX,Y2 = VMPX,Y3 = …… = VMPX,YN (5.4)

82
Based on Equations 5.3 and 5.4 we may say that when resources are limited, revenue will be
maximized when he value of the marginal products or the variable input will be the same for all
products or enterprises.

Most resources can be used in the production of more than one product, i.e. they have an
alternative use. As indicated in the PPC, the variable input X can be used to produce Y1 or Y2
The opportunity cost of a resource is defined as the value of that resource when it is put to its
best alternative use. Thus, when one unit of X is shifted from the production of Y2 to Y1, the
opportunity cost of that unit of X is the value of Y2 that is foregone in order to increase the
amount of Y1. According to equation 5.3, when profit is maximized the VMPX,Y2 = VMPX,Y1,
which means, the opportunity cost of the resource should be equal across all products.

6 CONSTRAINED OPTIMIZATION

83
Thus far, in deriving the profit maximizing decision criteria we have assumed that resources are
not limited. However, in practice there are many constraints facing any producer, the money to
buy inputs may not be enough, inputs may not be obtained in desired quantities. Therefore, often
producer have to settle for less than maximum profit. They strive to get as high a profit as
possible (optimize profit) subject to prevailing constraints. Similarly a producer who is trying to
minimize cost may be facing certain constraints, which must be met. The producer will strive to
minimize cost subject to the constraint(s). So the producer’s problem becomes one of
constrained maximization or minimization.

A maximization example
Suppose the farmer is producing two products, milk and butter. Let the price of milk be T.Shs
600 per unit while that of butter is T.Shs 800 per unit. The farmer’s goal is two maximize
revenue, which is defined by;
R = PY1Y1 + PY2Y2
Where R = Total revenue
Y1 = Number of tins of milk
Y2 = Number of tins of butter
But, milk and butter are joint products, they are related through a PCC, given by;
18Y1 = 84 – 3Y22
Thus in the effort to maximize revenue, the farmer is limited by this constraint.
The farmer’s problem is therefore formulated as follows;
Maximize R = PY1Y1 + PY2Y2 (6.1)
Subject to 18Y1 = 84 – 3Y22
This problem can be solved using several approaches (eg by substitution, followed by
maximization). In the following discussion, we will illustrate how to use the Lagrangean
multiplier in solving constrained optimization problems. This approach is preferred in
economics because in addition to the optimum solution, it provided additional information
(shadow prices in the case of inputs), which is useful for planning purposes. The farmers
problem in Equation 6.1 may be solved using the Lagrangean method by following these steps;

(1) Define the objective function to be maximized (or minimized)

84
 Objective function; R = PY1Y1 + PY2Y2 (6.2)

(2) Rewrite the constraint

18Y1 - 84 – 3Y22 = 0 (6.3)

(3) Multiply the constraint by a factor λ

λ (18Y1 - 84 + 3Y22) = 0 (6.4)

(4) Define a Lagrangean function, constituted by the objective function (Equation 6.2) and the
constraint as expressed in equation 6.4

L = PY1Y1 + PY2Y2 + λ (18Y1 - 84 + 3Y22) (6.5)

Note that adding Equation 6.4 to the objective function does not really change the objective
function, Thus, for all practical purposes, Equation 6.5 is the same as equation 6.2. Note also
that the Lagrangean function contains three unknown variables on the left had side (Y1, Y2 and
λ). We can solve for these unknown variables by obtaining the derivatives of the Lagrangean
function (Equation 6.4) with respect to each of the unknowns. When L is at its maximum, each
of the first order derivatives will be equal to zero.

(5) Derive the first order derivatives of the lagrangean function (first order conditions) and set
the derivative equal to zero
δL
----- = 600 + 18λ = 0 (6.6)
δ Y1

δL
----- = 800 + 6λY2 = 0 (6.7)
δ Y2

δL
----- = 18Y1 - 84 + 3Y22 = 0 (6.8)
δλ

This gives 3 equations with three unknown. We can therefore solve for Y1, Y2 and λ to
determining the constrained optimizing combination of products (Y1 & Y2)

- 600
From Equation 6.6; λ = ----- (6.9)
18

85
 - 800
From Equation 6.7; λ = ---- (6.10)
6Y2

Equating 6.9 and 6.10;

600 800
----- = -----
18 6Y2

∴ 3600Y2 = 14400
Y2 = 4

By substitution;
18Y1 – 84 + 48 = 0
18Y1 = 36
Y1 = 2

∴ Y1 = 2; Y2 = 4 and λ = - 33.33

A minimization example
Suppose that a production of an output (Y) involves two inputs, Labour (W) and Capital (K).
The isoquant is given by Equation 6.11.
100
K = ----- + 30 (6.11)
2
W
The price of labour and capital are T.Shs 6 and 3 per unit respectively. The farmer would like to
minimize the cost of producing Y, subject to the technical constraint as defined by the isoquant.
So the farmer’s problem may be stated as;
Minimize: TC = 6W + 3K (6.12)

100
Subject to: K = ----- + 30
W2

Following steps of the Lagrangean method;

100
L = 6W + 3K + λ ----- + 30 - K (6.13)
W2

δL 200 λ

86
----- = 6 - ------- = 0 (6.14)
δW W3

δL
----- = 3 - λ = 0 (6.15)
δK

δL 100
----- = ----- + 30 - K = 0 (6.16)
δλ W2

Solving for the three unknown variables we get;

λ = 3; W = 4.624; K = 34.643

The lagrangean multiplier (λ) has a usefull interpretation in economics. It represents the value
by which the objective function would change if the constraint is relaxed by one unit.

87
7 ECONOMIES OF SIZE AND ECONOMIES OF SCALE
In chapter 2 we introduced the concepts of short run and long run as they relate to fixed and
variable inputs. In Chapter three we discussed about average variable costs and average total
costs that a producer faces in the short run. Thus in the short run a manager faces a given plant
size because some inputs are fixed. In the long run however, the manager can change the plant
size, often, with the objective of improving efficiency. We said earlier that improvement in
economic efficiency is associated with lower average variable cost. Thus as the firm manager
improves the efficiency of his/her plant, successive plant sizes will be associated with lower
AVC. If one traces the locus of these AVC points, a long run average cost curve (LRAC) is
designed. When the firm is small expansion of output usually increases efficiency and average
cost will fall due to;
• Increasing possibilities of division and specialization of labour and capital
• Increasing possibility of using advanced technologies , which lower the average cost of
producing output
• The firm is able to spread fixed cost over a lager amount of output
• The same machinery is able to do more field work
• Buildings may be housing the maximum number of livestock
• Some of the variable cost may be reduces ( e.g. one may be able to get discounts when
buying inputs in large quantities)

However, as the plant size increases, beyond a certain point, LRAC begins to rise due to;
• Managerial limitations
• Bureaucratic red tape (communication, coordination)
• Frequency of mistakes go up

Within the range where LRAC is decreasing due to increasing plant size, the firm is said to have
economies of size. In the region where LRAC is rising with increasing plant size, the firm is said
to have dis-economies of size.

The LRAC is a planning curve. Once a given plant size has been built/established, the plant is
operating in the long run.

88
 Shs
SRAC LRAC
(Envelop curve
SRMC

Most efficient
level of output
for this firm size
in the long run

Y
Economies of size Dis-economies of size

Figure 7.1 Long Run Average Cost Curve and Economies of Size

ECONOMIES/DISECONOMIES OF SCALE
From figure 7.1we can say that;
• For each plant size, there is a corresponding level of variable input (in the short run), that
minimizes cost
• This occurs at the point where
MPPX1 MPPX2 MPPX3 MPPXN
-------- = ------- = -------- = ……… = ----------
PX1 PX2 PXN PXN

This means, if we regard all the inputs as the set of variable inputs while plant size is treated as
the other input, In the long run one can trace the long run expansion path.

89
X1
All inputs

Long run
expansion path

X2
Plant size

Figure 7.2 Long run Expansion Path

Along the expansion path;

LRTC + PX1X1 + PX2X2

LRAC = PX1X1 + PX2X2

--------- --------
Y Y
Returns to scale
Returns to scale are related to economies and dis-economies of scale, not only in the short run,
but also in the long run. They refer to the effect of an increase in output upon average cost when
all inputs are increased by the same proportion.
Exmple, for a long run production function given as;
Y = f(X1, X2, X3,…………XN) (7.1)
Where Y = output
Xi = inputs
If each of the variable inputs is changed (increase or decrease) by the same factor (K)
Y may change by a factor = K
Y may change by a factor < K
Y may change by a factor > K

∴ YKN = f(KX1 + KX2 +………+ KXM) (7.2)

If N = 1 The function is said to have constant return to scale

90
 N > 1 The function is said to have increasing returns to scale
N < 1 The function is said to have decreasing return to scale

The isoquant map for the three scenarios are represented in figure 7.3.
Increasing
Constant decreasing
X1

Figure 7.3 Returns to Scale

91