UNIT THREE

THE THEORY OF PRODUCTION

Introduction
Production is the creation of any good or services that have economic value to either consumers
or producers. Production may be defined as the act of creating those goods/services which have
exchange value for sale (not for personal consumption).Raw materials yield less satisfaction to
the consumer by themselves. In order to get utility from raw materials, first they must be
transformed into output. However, transforming raw materials into final products require factor
inputs such as land, labor, capital and entrepreneurial ability. Thus, no production (transforming
raw material into output) can take place without the use of inputs.
3.1 Definition and Basic Concepts
Production Function
It is a purely technical relationship which connects factors (inputs) to products (outputs) at any
particular time period. It also shows the maximum quantity of output that can be produced from
the given amount of variable inputs for a given technology. It describes the technical relationship
between inputs and out put in physical unit.
Fixed Vs variable inputs
In economics, inputs can be classified as fixed and variable. Fixed inputs are those inputs whose
quantity cannot readily be changed when market conditions indicate that an immediate change
in output is required. In fact no input is ever absolutely fixed, but may be fixed during an
immediate requirement. For example, if the demand for Beer shoots up suddenly in a week, the
brewery factories cannot plant additional machinery over night to respond to the increased
demand. It takes long time to buy new machineries, to plant them and use for production
purposes. Thus, the quantity of machinery is fixed for some times such as a weak. Buildings,
machineries and managerial personnel are examples of fixed inputs because their quantity cannot
be manipulated easily in short time periods.
Variable inputs, on the other hand, are those inputs whose quantity can be changed almost
instantaneously in response to desired changes in output. That is, their quantity can easily be
diminished when the market demand for the product decreases and vise versa. The best example
of variable input is unskilled labor. In our previous example, if the brewery factory had idle
machinery before the market demand increased, the factory can easily and immediately respond
to the market condition by hiring laborers.
 Short run vs long run
In economics, short run refers to a period of time in which the quantity of at least one input is
fixed. For example, if it requires a firm one year to change the quantities of all the inputs, those
time periods below one year are considered as short run. Thus, short run is a time period which
is not sufficient to change the quantities of all inputs, so that at least one input remains fixed.
One thing to be noted here is that short run periods of different firms have different durations.
Some firms can change the quantity of all their inputs with in a month while it takes more than a
year for another type of firms. For example, the time required to change the quantities of inputs
in an automobile factory is not equal with that of flour factory. The later takes relatively shorter
time.On the other hand, long run is a time period (planning horizon) which is sufficient to change
the quantities of all inputs. Thus, there is no fixed input in the long run
3.2 Production in the short run: Production with one variable input
Short run production function is a production function which combines at least one fixed input
with variable input.
Assumption of short run production analysis
In order to simplify the analysis of short run production, the classical economists assumed the
following conditions:
1. Perfect divisibility of inputs and outputs
This assumption implies that factor inputs and outputs are so divisible that one can hire, for
example a fraction of labor, a fraction of manager can produce a fraction of output, such as a
fraction of automobile.
2. Limited substitution between inputs
Factor inputs can be substituted each other up to a certain point, beyond which they can not be
substituted each other. In other words, resources are not perfect substitutes for one another as
they are not identical.
3. Constant technology
It is assumed that level of technology of production is constant in the short run.Suppose a firm
that uses two inputs: Capital (which is a fixed input) and labor (which is variable input). Given
the assumptions of short run production, the firm can increase output only by increasing the
amount of labor it uses. Output varies with the variation of the variable input.Hence, its
production function is:
Q = f (L) K - being constant
 Where Q is the quantity of production (Output)
L is the quantity of labor used, which is variable, and
K is the quantity of capital (which is fixed)
The production function shows different levels of output that the firm can obtain by efficiently
utilizing different units of labor and the fixed capital. In the above short run production function,
the quantity of capital is fixed. Thus output can change only when the amount of labor used for
production changes. Hence, Q is a function of L only in the short run.
3.3 Total Product (TP), Marginal Product (MP) and Average Product (AP)
Total Product (TP)
Total product is the total amount of output that can be produced efficiently utilizing a specific
combination of labor and capital. The total product curve, thus, represents various levels of
output that can be obtained from efficient utilization of various combinations of the variable
input, and the fixed input. Dear learner, do you think that output can always be increased by
increasing the variable input while there is a fixed input?
Increasing the variable input (while some other inputs are fixed) can increase the total product
only up to a certain point. Initially, as we combine more and more units of the variable input with
the fixed input output continues to increase. But eventually, increasing the unit of the variable
input may not help output increase. Even as we employ more and more unit of the variable input
beyond the carrying capacity of a fixed input, output may tends to decline. Thus, increasing the
variable input can increase the level of output only up to a certain point, beyond which the total
product tends to fall as more and more of the variable input is utilized. This tells us what shape
a total product curve assumes. The shape of the total variable curve is nearly S-shape.
Marginal Product (MP) of a variable input is the addition to the total product attributable to the
addition of one unit of the variable input to the production process, other inputs being constant
(fixed). Before deciding whether to hire one more worker, a manager wants to determine how
much this extra worker (L =1) will increase output. The change in total output resulting from
using this additional worker (holding other inputs constant) is the marginal product of the worker.
If output changes by Q when the number of workers (variable input) changes by ∆L, the change
in output per worker or marginal product of the variable input, denoted as MPL is found as
Q dTP
MPL = or MPL 
L dL
Thus, MPL measures the slope of the total product curve at a given point. In the short run, the
MP of the variable input first increases, reaches its maximum and then tends to decrease to the
extent of being negative. That is, as we continue to combine more and more of the variable inputs
with the fixed input, the marginal product of the variable input increases initially and then
declines.
Average Product (AP) of an input is the ratio of total output to the number of variable inputs.
totalproduct TP
APlabor  
numberofL L
 The average product of labor first increases with the number of labor (i.e. TP increases faster
than the increase in labor), and eventually it declines. However, average product can never be
zero or negative as it is a ratio of two positive numbers (total output and employed labor).
Short Run Production Curves
The following figures shows how the TP, MP and AP of the variable (labor) input vary with the
number of the variable input.

Output
a

TP3

TP2 TP

TP1

L1 L2 L3 Units of labor (variable

APL, MPL input)

APL

L1 L2 L3 Units of labor (variable input)

MPL

Fig 3.1 Total product, average product and marginal product curves:
As the number of the labor hired increases (capital being fixed), the TP curve first rises, reaches
its maximum when L3 amount of labor is employed, beyond which it tends to decline. Assuming
that this short run production curve represents a certain car manufacturing industry, it implies
that L3 numbers of workers are required to efficiently run the machineries. If the numbers of
workers fall below L3, the machine is not fully operating, resulting in a fall in TP below TP3. On
the other hand, increasing the number of workers above L3 will affect the production process
negatively because only L3 number of workers can efficiently run the machine. Increasing the
number of workers above L3, rather results in lower total product because it results in
overcrowded and unfavorable working environment.
Marginal product curve increases until L1 number of labor are employed and reaches its
maximum at L1, and then it tends to fall. The MPL is zero at L3 (when the TP is maximal); beyond
which its value assumes negative value indicating that each additional worker above L3 tends to
create over crowded working condition and reduces the total product. Thus, in the short run
(where some inputs are fixed), the marginal product of successive units of labor hired increases
initially, but not continuously, resulting in the limit to the total production. Geometrically, the
MP curve measures the slope of the TP. The slope of the TP curve increases (MP increases) up
to L1, it decreases from L1 to L3 and it becomes negative beyond L3.
The average product curve increases up to L2, beyond which it continuously declines. The AP
curve can be measured by the slope of rays originating from the origin to a point on the TP curve.
For example, the APL at L2 is the ratio of TP2 to L2. This is identical to the slope of ray a.
The Relationship between AP and MP of Variable Input
The relationship between MPL and APL can be stated as follows:
 For all number of workers (Labor) below L2, MPL lies above APL .
 At L2, MPL and APL are equal.
 Beyond L2, MPL lies below the APL
 Total product reaches maximum when marginal product is zero.
Thus, the MPL curve passes through the maximum of the APL curve from above. This relationship
between APL and MPL can be shown algebraically as follows:
Suppose the production function is given as
TP = f (L), K -being constant
Given the total product function,
dTP df ( L) TP
MPL   and APL  = f ( L)
dL dL L L
To determine the relationship between APL and MPL, consider the slope of the APL function.
f ( L) df ( L) dL
d( ) .L  . f ( L)
dAPL L dL dL
Slope of APL = = = -------- (quotient rule of
dL dL L2
differentiation)
df ( L)
.L
f ( L) ab a b
Slope of APL = dL2 - 2
----------------------------- (note that   )
L L c c c
 df ( L) f ( L)
= dL - L
L L
MPL  APL df ( L ) f ( L)
Slope of APL = , because = MPL and = APL
L dL L
Now – when MPL > APL, Slope of APL is positive (APL rises)
 When MPL = APL, Slope of APL is zero (APL is at its maximum).
 When MPL < APL, Slope of APL is negative (APL falls)
1.4 The Law of Diminishing Marginal Returns (LDMRs): Short –Run
Law of Production
The LDMRs states that as the use of an input increases in equal increments (with other inputs
being fixed), a point will eventually be reached at which the resulting additions to output
decreases. When the labor input is small (and capital is fixed), extra labor adds considerably to
output, often because workers get the chance to specialize in one or few tasks. Eventually,
however, the LDMRs operates: when the number of workers increases further, some workers
will inevitably become ineffective and the MPL falls (this happens when the number of workers
exceeds L1 in fig 3.1)
The Law of diminishing marginal returns is based on two very important assumptions. These are
the existence of constant technology and homogeneous labor (the variable input). The LDMRs
operates (MP of successive units of labor decreases) not because highly qualified laborers are
hired first and the least qualified last. Diminishing marginal returns results from limitations on
the use of other fixed inputs (e.g. machinery), not from decline in worker quality. The LDMR
applies to a given production technology (when the level of technology is fixed). Over time,
however, technological improvements in the production process may allow the entire total
product curve to shift upward, so that more output can be produced with the same level of inputs.
1.5 Stages of Production in the short-run or Efficient Region of
Production in the short-run
Dear learner, we are now not in a position to determine the specific number of the variable input
(labor) that the firm should employ. Because, this depends on several other factors than the
productivity of labor such as the price of labor, the structure of input and output markets, the
demand for output, etc. However, it is possible to determine ranges over which the variable input
(labor) be employed.
To do best with this, let us refer back to fig 3.1 and divide it into three ranges called stages of
production. The production of a firm in the short run can be divided in to three stages of
production.
 Stage I – ranges from the origin to the point of equality of the APL and MPL.
 Stage II – starts from the point of equality of MP L and APL and ends at a point where
MP is equal to zero.
 Stage III – covers the range of labor over which the MPL is negative.
Graphically, the three stages of production are indicated bellow
APL
MPL

Stage I Stage II Stage III

APL

0 Units of labor (variable input)

L1 L2 L3
MPL
Fig 3.2The Three Stages of Production in the Short run
Dear learner, which stage of production do you think is efficient and preferable?
Dear learner! Since additional units of variable inputs are contributing negatively to the total
product a firm should not operate in stage III (MP of the variable input is negative).It is due to
overcrowded working environment i.e., the fixed input is over utilized.
Stage I is also not an efficient region of production though the MP of variable input is positive.
The reason is that the variable input (the number of workers) is too small to efficiently run the
fixed input; so that the fixed input is underutilized (not efficiently utilized)
Thus, the efficient region of production is stage II. At this stage additional inputs are contributing
positively to the total product and MP of successive units of variable input is declining (indicating
that the fixed input is being optimally used). Hence, the efficient region of production is over
that range of employment of variable input where the marginal product of the variable input is
declining but positive. This stage of production is called rational or economic stage of
production.
Example
Assume that a production function is given as
Q  f ( L)
Q  7 L  10L2  L3
Determine:
A) The amount of employment which maximizes average productivity
B) The amount of employment which maximizes total production or at which marginal
product is zero.
C) Define the three stages of production for the above production function.
Solution:
A) Average product reaches maximum when the slope of average product curve is zero. In
other words, average product reaches maximum when marginal product equals to average
product.

TP 7 L  10L2  L3
Slope of APL  dAPL  0 , and APL=   7  10L  L2
dL L L
d (7  10L  L2 )
Slope of APL  0
dL
 10  2L  0  L  5
L5
Thus, when employment of labor equals to five average product reaches its maximum.

B) MPL  d (TP )  0
dL

d (7 L  10L2  L3 )
MPL  0
dL
MPL  7  20L  3L2  0
 7  20L  3L2  0 Using Quadratic solution method, a= 3, b= -20 and c= -7

b b 2  4ac

2a

20 (20) 2  4(3)(7)

2(3)
 
20 484

6

20 22

6
20  22 20  22
 or  7 or  0.33
6 6
Thus, total product reaches maximum when employment of the variable input equals to seven.
A) The three stages of production is defined as follows:
Stage one from zero labor to 5 units of labor. Stage two from 5 labor to 7 units of labor
and Stage three above 7 units of employment
Graphically,

APL
MPL

Stage I Stage II Stage III

APL

0 L1 5 7
Units of labor (variable) input
MPL
Fig 3.3 Specific example for the three Stages of Production in the Short run
3.6 Long run Production: Production with two variable inputs
Dear learner, in the previous section we have examined the technological relationship between
inputs assuming that the firm uses one variable input (labor) and one fixed input (capital). Now,
let us turn to the long run analysis of production. Remember that long run is a period of time
(planning horizon) which is sufficient for the firm to change the quantity of all inputs. For the
sake of simplicity, assume that the firm uses two inputs (labor and capital) and both are variable.
The firm can now produce its output in a variety of ways by combining different amounts of
labor and capital. With both factors variable, a firm can usually produce a given level of output
by using a great deal of labor and very little capital or a great deal of capital and very little labor
or moderate amount of both. In this section, we will see how a firm can choose among
combinations of labor and capital that generate the same output. To do so, we make the use of
an isoquant. So it is necessary first to see what is meant by an isoquant and its properties.
Isoquant
Dear learner! An isoquant is a curve which shows alternative combinations of inputs leading to
the same level of output. It shows all possible efficient combinations of inputs that can yield
equal level of output. Every point on an isoquant represents the same level of output. If both
labor and capital are variable inputs, the production function will have the following form:
Q = f (L, K)
Thus, isoquant shows the flexibility that firms have when making production decision. They can
usually obtain a particular output (Q) by substituting one input for the other.
Isoquant Maps: Isoquant map is a collection of different isoquant curves each representing a
specific output level. When a number of isoquants are combined in a single graph, we call the
graph an isoquant map. An isoquant map is another way of describing a production function.
Each isoquant represents a different level of output and the level of output increases as we move
up and to the right. The following figure shows isoquants and isoquant map.

Capital

3
q3
2
q2
1 q1

1 3 6 Labor

Fig 3.4 Isoquant and isoquant map.

Isoquants show the fact that long run production process is flexible. A firm can produce q1 level
of output by using either 3 units of capital and 1 unit of labor or 2 units of capital and 3 units of
labor or 1 unit of capital and 6 units of labor or any other combination of labor and capital on the
curve. The set of isoquant curves q1 q2 and q3 is called isoquant map.
 Properties of isoquants
Dear learner! Most of the properties of an isoquant are the same as that of an indifference curve.
The major difference between them is that output is constant along an isoquant where as
indifference curves hold utility constant. Thus, isoquants are:
1. Isoquants slope down ward. Because isoquants denote efficient combination of inputs
that yield the same output, they always have negative slope and can never be horizontal,
vertical or upward sloping. For example, if isoquants are horizontal only one point on the
isoquant is efficient. See the following figures.

Capital
Capital Capital 100kg
100kg
wheat
teff

A 100kg
4 teff
B
A
A

5 5 Labor Labor
2 Labor
Panel a Panel b Panel c
Fig 3.5 Isoquants with zero or positive slope
In fig.3.5 panel (a), the firm can produce 100kg of Teff by using either 4 capitals and 2 labors,
or 4 capitals and 5 labors or any other combination of labor and capital along the curve.
Obviously, only the first alternative is efficient as it uses the least possible combination of inputs.
Thus, all points, except A, are inefficient and not part of isoquant. Thus, an isoquant can never
be horizontal.
In fig.3.5 panel (b), a firm can produce 100kg of wheat by using any combination of labor and
capital along the isoquant. But only point A is efficient. For example, point B shows the same
number of labor as point A, but higher capital. Thus point B is inefficient because it shows higher
combination of inputs. Thus, isoquants can never be vertical line
In fig.3.5 panel (c), all points above point A utilize higher combination of both inputs to produce
the same output (100 kg coffee). Point A shows the least combination of inputs that can yield
100 kg coffees. Thus, all other points are inefficient and not part of the isoquants.
Thus, efficiently requires that isoquants must be negatively sloped. As employment of one factor
increases, the employment of the other factor must decrease to produce the same quantity
efficiently.
2. The further away an isoquant lays from the origin, the greater the level of output it denotes.
Higher isoquants (isoquants further from the origin) denote higher combination of inputs. The
more inputs used, more outputs should be obtained if the firm is producing efficiently. Thus,
efficiency requires that higher isoquants must denote higher level of output.
3. Isoquants do not cross each other. This is because such intersections are inconsistent with the
definition of isoquants.
Consider the following figure.

Capital

q=50
Q=20
Fig 3.6 Crossing Isoquants

Efficiency requires that isoquants do not

cross each other, because the point of
K* their intersection implies that there is
inefficiency at this point.

L*
This figure shows that the firm can produce at either output level (20 or 50) with the same
combination of labor and capital (L* and K*). The firm must be producing inefficiently if it
produces q = 20, because it could produce q = 50 by the same combination of labor and capital
(L* and K*). Thus, efficiency requires that isoquants do not cross each other.
4) Standard isoquants are convex to the origin
The slope of an isoquant is called marginal rate of technical substitution. Marginal rate of
technical substitution declines as we substitute on input for the other. This makes isoquants
convex to the origin. We will investigate this concept later.

Capital

Fig.3.7: Convex Isoquants Labor

Special Shapes of isoquants
Isoquants can have different shapes (curvature) depending on the degree to which factor inputs
can substitute each other.
1-Linear isoquants
Isoquants would be linear when labor and capital are perfect substitutes for each other. In this
case, the slope of an isoquant is constant. As a result, the same output can be produced with only
capital or only labor or an infinite combination of both. Graphically,

10

q=100
8

12 15
L
Fig.3.8: Linear isoquant. Capital and labor can perfectly substitute each other so that the same
output (q=100) can be produced by using either 10k or 8K and12L or 15L or an infinite
combinations of both inputs.
2. Fixed Factor Proportion or L-shaped Isoquants
When a production function assumes fixed proportion between capital and labor, the isoquant
takes “L”-shape. It is also called Leontief isoquant. This assumes strict complementarities or zero
substitutability of factors of production. In this case, it is impossible to make any substitution
among inputs. Each level of output requires a specific combination of labor and capital:
Additional output cannot be obtained unless more capital and labor are added in specific
proportions. As a result, the isoquants are L-shaped. See the following figure. For example to
derive a car we always need one driver for one car which a one to one ratio in this case

q3

q2
K2

q1
 K1

L
L1 L2
Fig. 3.9 L-shaped isoquant when isoquants are L-shaped, there is only one efficient way of producing
a given level of output: Only one combination of labor and capital can be used to produce a given level
of output. To produce q1 level of output there is only one efficient combination of labor and capital
(L1 and K1). Output cannot be increased by keeping one factor (say labor) constant and increasing
the other (capital). To increase output (say from q1 to q2) both factor inputs should be increased by
equal proportion.
3. Kinked isoquants
This assumes limited substitutability between inputs. Inputs can substitute each other only at some
points. Substitution of inputs is possible only at the kink points. Thus, the isoquant is kinked and there
are only a few alternative combinations of inputs to produce a given level of output. These isoquants
are also called linear programming isoquants or activity analysis isoquants. See the figure below.

12 A

7 B

5 C

D
3
Q=100

L
1 3 5 9

Fig. 3.10: kinked isoquant In this case labor and capital can substitute each other only at some
point at the kink (A, B, C, and D). Thus, there are only four alternative processes of producing
q=100 out put.
4. Smooth, convex isoquants
This shape of isoquant assumes continuous substitution of capital and labor over a certain range,
beyond which factors cannot substitute each other. Basically, kinked isoquants are more realistic:
There is often limited (not infinite) method of producing a given level of output. However, traditional
economic theory mostly adopted the continuous isoquants because they are mathematically simple to
handle by the simple rule of calculus, and they are approximation of the more realistic isoquants. From
now on we use the smooth and convex isoquants to analyze the long run production.

∆K=4

∆L=1

∆K=2 ∆K=1/2

∆L=1 Q
∆L=1

L
Fig: 3.11: Smooth and convex isoquant.
This type of isoquant is the limiting case of the kinked isoquant when the number of kink is infinite.
The slope of the isoquant decrease as we move from the top (left) to the right (bottom) along the
isoquant. This indicates that the amount by which the quantity of one input (capital) can be reduced
when one extra unit of another inputs (labor) is used (so that out put remains constant) decreases as
more of the latter input (labor) is used.
3.7 Slope of an Isoquant: Marginal Rate of Technical
Substitution (MRTS)
The slope of an isoquant (-K/L) indicates how the quantity of one input can be traded off against
the quantity of the other, while out put is held constant. The absolute value of the slope of an isoquant
is called marginal rate of technical substitution (MRTS). The MRTS shows the amount by which the
quantity of one input can be reduced when one extra unit of another input is used, so that output remains
constant. MRTS of labor for capital, denoted as MRTS L, K (Marginal rate of technical substitution
between labor and capital) shows the amount by which the input of capital can be reduced when one
extra unit of labor is used, so that output remains constant.
MRTS L, K decreases as the firm continues to substitute labor for capital (or as more of labor is used).
In fig.3.11 to increase the amount of labor from 1 to 2, the firm reduces 4 units of capital (K=4), to
increase labor from 2 to 3, the firm reduce 2 units of capital (K=2), and so on. Hence, the firm reduces
lower and lower number of capital for the successive one unit of labor. Dear learner, why does this
happen?
The reason is that when the number of capital is large and that of labor is low, the productivity of
capital is relatively lower and that of labor is higher (due to the low of diminishing marginal returns).
Thus, at this point relatively large amount of capital is required to replace one unit of labor (or one unit
of labor can replace relatively large amount of capital). As the employment of labor increases and that
of capital decreases (as we move down ward along the isoquant), quite the reverse will happen. That
is, productivity of capital increases and that of labor decreases. Hence, the amount of capital that needs
to be reduced increase when one extra labor is used decreases. The fact that the slope of an isoquant is
decreasing makes an isoquant convex to the origin.
MRTS L, K (the slope of isoquant) can also be given by the ratio of marginal products of factors.
K MPL
That is, MRTS L , K   
L MPK
This can be shown algebraically as follows:

Let the production function is given as:

Where q- is output
q= f (L, K)
L- is unit of labor employed

K-is the amount of capital employed.

Given this production function, the equation of a specific isoquant can be obtained by equating the

production function with a given level of output, say q .

q = f (L, K) = q

Total differential of q measures the total change in q that happens as a result of a simultaneous
change in L and K. i.e.,
q f
dq  .dL  .dk  d q
L k
But since q is constant, d q is zero (d q =0)
q q
So, .dL  dk  0
L k
q q
(But,  MPl and  MPk )
L k
Thus, the above equation can be written as:
MPL. dL + MPK.dk = 0
 MPL  dK
  = MRTSL,K
MPk dL
Therefore, the slope of an isoquant can be given as the ratio of marginal products of inputs.
Elasticity of Substitution
MRTS as a measure of the degree of substitutability of factors has a serious defect. It depends on the
units of measurement of factors. A better measure of the ease of factor substitution is provided by the
elasticity of substitution, δ. Elasticity of substitution is percentage change in capital labor ratio divided
by percentage change in marginal rate of technical substitution. The elasticity of substitution is defined
as
K K
% %
 L = L
%MRTS MPL
%
MPK
K
( )
d L
K
= L
 MPL  MPL
d  ( )
 MPK  MPK
The elasticity of substitution is a pure number independent of the units of measurement of K and L,
since the numerator and the denominator are measured in the same units and be cancelled.
Factor Intensity
A process of production can be labor intensive or capital intensive or neutral process. A process of
production is called labor intensive if it uses relatively large amount of labor than capital. If it uses
many capitals and relatively few labor it is called capital intensive technology. On the other hand, if
the process uses equal proportion of both it is called neutral technology. The factor intensity of any
process is measured by the slope of the line through the origin representing the particular process.
Thus, the factor intensity is the capital-labor ratio. The higher the capital-labor ratio is the higher the
capital intensity but the lower the capital-labor ratio is the higher labor intensity of the process.
 Capital

A
K1

B
K2 X

O
L2
L1 Labor
Fig 3.12 Factor Intensity
Process A uses k1 and L1 units of labor and capital to produce X amount of output. The factor intensity
of this process can be measured by the slope of OA, which equals
OK 1 K 1
AL1/OL1 = 
OL1 L1

K2
Similarly, factor intensity of process B is given by
L2

K1 K 2
Since  , process A is more capital intensive than process B or B is more labor intensive than
L1 L2
A. The upper part of the isoquant includes more capital intensive processes and the lower part, labor
intensive techniques.
Now let us illustrate the above concepts with the most popular and applicable form of production
function, Cobb-Douglas production function
The Cobb- Douglas production function is of the form
X  b0 Lb1 K b 2
From this production function
X
1. MPL =  b1b0 Lb11 K b2
L
Lb1 b 2
= b0 b1 K since X  b0 Lb1 K b 2
L
X
 b1
L
= b1APL
 K
MPK =  b2 b0 Lb1 K b 21
L
X
= b2 since X  b0 Lb1 K b 2
K
= b2 APK
2. Marginal rate of technical substitution
X
b1
MPL L  b1 . K
(MRT SLK) = 
MPK X b2 L
b2
K
3. The elasticity of substitution
K
 
d L 
K
 L
MPL
d( )
MPK
MPL
MPK

K K
   
d L  d L 
K K
 L  L
d (b1 K L) (b1 ) d ( K L)
b2 b2
b1 K b1 K
b2 L b2 L
 1
4. Factor intensity is measured by the ratio b1/b2. The higher the ratio, the more labor intensive the
b1
technique. Similarly, the lower the ratio of , the more capital intensive the technique.
b2
5. The efficiency of production. This is measured by the coefficient b0. It is clear that if two firms have
the same K, L, b1 and b2 and still produce different quantities of output, the difference could be due to
the superior organization and entrepreneurship of one of the firms, which results in different
efficiencies. The more efficient firm will have a larger b0 than the less efficient one.
3.8 The Efficient Region of Production: Long run
In principle, the marginal product of a factor may assume any value, positive, zero or negative.
However, the basic production theory concentrates only on the efficient part of the production function,
i.e. over the range of output over which the marginal product of factors are positive and declining. In
the short run production function, efficient region of production prevails in stage two (stage II), where
MPL
MPL >O, but < 0.
L
Similarly, efficient region of production in the long run prevails when the marginal product of all
variable inputs is positive but decreasing. In a given isoquant all points are not economically efficient.
Graphically this can be represented by the negatively slopped part of an isoquant. The locus of points
of isoquants where the marginal products of factors are zero form the ridge lines. The upper ridge line
implies that the MP of capital is zero. MPK is negative for all points above the upper ridge line and
positive for points below the ridge line. The lower ridge line implies that the MPL is zero. For all points,
below the lower ridge line the MPL is negative and positive for points above the line. Production
techniques are technically efficient inside the ridge lines. Symbolically in the long run efficient
production region can be illustrated as:
MPL
MPL >0, but <0
L
MPk
MPk >0, but <0
K
Graphically, efficient region of production is shown as follows:

Capital
Upper ridge line

Lower ridge line

q3

q2

q1

Labor

Fig 3.13: Economic Region of an Isoquant

Thus efficient region of production is defined by the range of isoquants over which they are convex to
the origin. Or the area between the two ridge lines is called economic region or technically efficient
region of production
MPL
From our previous section we know that MRTS L, K  and along the lower ridge line, since MPL
MPK
0 MP
is zero, MRTS L, K   0 .Along the upper ridge line MRTS L, K  L .since MPK is zero,
MPK MPK
MPL
MRTS L, K    .Note that MRTSK,L equals to zero along the upper ridge line
0
3.9. The Long Run Law of Production: The Law of Returns to Scale
The laws of production describe the technically possible ways of increasing the level of production.
Output may increase in various ways. In the long run, output can be increased by changing all factors
of production. This long run analysis of production is called Law of returns to scale.
In the short run, output may be increased by using more of the variable factor, while capital (and
possibly other factors as well) are kept constant. The expansion of output with one factor (at least)
constant is described by the law of variable proportion or the law of returns to the variable factor.
In the long run, all inputs are variable. Expansion of output may be achieved by varying all factors of
production by the same proportion or by different proportions. The law of returns to scale measures
the change in output when all inputs are altered by same or different proportion.
The traditional theory of production concentrates on the first case, i.e. the study of output as all inputs
change by the same proportion. The term returns to scale refers to the change in output as all factors
change by the same proportion.
Due to the proportionate change in inputs, output may change in any of the following three ways.
A) Output changes more proportionately than the change in inputs. This is the case of
Increasing returns to scale (IRS)
B) Output changes less proportionately than the change in inputs. When the percentage change in
output is less than that of the inputs we have what is called decreasing returns to scale.
C) When output changes by the same percentage that inputs have been changed, the production
function is said to exhibit constant returns to scale (CRS)
Suppose initially the production function is
X0 = f (L, K)
If we increase all factors by the same proportion c, we clearly obtain a new level of output X* where,
X* = f (cL, cK)
 If X* increases by the same proportion c or if X* = cX0, we say that there is constant returns to
scale.
 If X* increases less than proportionally with the increase in the factors (or if X* increases by a
proportion less than c), we have decreasing returns to scale.
 If X* increases more than proportionally with the increase in the factors (by a more than c
proportion), we have increasing returns to scale.
Returns to scale and homogeneity of production function
Suppose we increase both factors of the function X0=f (L, K) by the same proportion ‘c’, and we get
the new level of output X = f (cL, cK).
If c can be factored out (that is, may be taken out of the brackets as a common factor), then the new
level of output X* can be expressed as a function of c (to any power v) and the initial level of output.
Such production function is called homogeneous production function. X* = c v f (LK) or X* =c V
X0
If c cannot be factored out, the production function is non-homogeneous. Thus, a homogeneous
function is a function such that if each of the input is multiplied by c, then c can be completely factored
out of the function. The power v of c is called degree of homogeneity of the function and is measure
of returns to scale.
If v =1, we have constant returns to scale. This production function is sometimes called linear
homogeneous.
If v <1, decreasing return to scale prevails
If v >1, increasing return to scale prevails
For a Cobb-Douglas production function the sum of the power of L and K measures the nature of
returns to scale. Assume a Cobb-Douglas production function given as;
X = b0Lb1 Kb2, v = b1 +b2 and it is a measure of returns to scale.
Proof: Let L and K increase by c. The new level of output is
X* = b0 (cL) b1 (ck) b2
X* = b0 cb1 lb1 cb2kb2
X* = b0Lb1Kb2 cb1+b2
X* = X (c b1+b2)
Thus v = b1+b2
Example: 1. If a production function is given by Q0=100 L0.3K 0.4
check whether the production
function exhibits CRS, DRS or IRS?
Solution: Change all inputs by a certain percentage c
Q1 =100 (cL) 0.3 (cK) 0.4
=100(c0.3) L 0.3 (c0.4) K0.4 by distributing the power
=100 c0.3+0.4 (L 0.3 K0.4) by adding the exponent when the base is the same
=c 0.7 (100 L 0.3 K0.4) factoring out the common term
Q1 = c 0.7 (Q0) v=0.7 < 1.
Therefore, the production function exhibits decreasing returns to scale or since the above production
function is an example of Cobb-Douglas production function, v= b1 + b2.
v=0.3 +0.4 =0.7(DRS)
2. Assume a production function given as Q0 = 10L+5K. Does the production function have CRS, DRS
or IRS?
Solution: Similar to the previous example, change all inputs by a certain proportion (c)
Q1 =10(cL) + 5 (cK)
= c (10L) +c (5K)
= c (10L+5K), by factoring out c
Q1 = c Q0, since v=1, the production function exhibits constant returns to scale
For a homogeneous production function the returns to scale may be represented graphically in an easy
way. Before explaining the graphical representation of the returns to scale it is useful to introduce the
concept of product line and isoclines.
3.10 Product Lines
Dear learner! A product line shows the (physical) movement from one isoquant to another as we change
the employment of either factors or a single factor, and it describes the technically possible alternative
paths of expanding output. The path actually chosen by the firm will depend on the prices of factors of
production.
The product line (curve) passes through the origin if all factors are variable. If only one factor is
variable (the other being constant) the product line is a straight line parallel to the axis of the variable
factor. For such product lines the K/L ratio diminishes along the product line
 Product
line

Product
line
Product
line Product
line

Product
line
O
O O
Product line for homogeneous Product line for non -
Product line for one
production function homogeneous production
factor (K) constant
function
Fig 3.14 Different forms of a Product line
A product line along which the MRTS of factors is constant is called an isocline. So an isocline is the
locus of points of different isoquants at which MRTS of factors is constant. If the production function
is homogeneous the isoclines are straight lines through the origin. Along any one isocline the K/L ratio
is also constant (as is the MRTS of the factors). But K/L ratio and the MRTS are different for different
isoclines. If the production is non homogeneous the isocline will not be straight line, but their shape
will be twiddle. The K/L ratio changes along each isocline and on different isoclines.
3.11 Graphical Presentation of Returns to Scale for Homogeneous
Production Function
The returns to scale may be shown graphically by the distance (on an isocline) between successive
“multiple level-of-output” isoquants, i.e. isoquants that show levels of output which are multiple of
some base level such as X, 2X, 3X etc.
Constant returns to scale
Along any isocline the distance between successive multiple- isoquant is constant. Doubling the factor
inputs doubles the level of initial output; trebling inputs trebles output, and so on.

c
3k

b
2k
3X

a 2X
k
X
O
1L 2L 3L L

Fig.3.15 Constant returns to scale: oa= ab = bc

 Decreasing returns to scale
Here, the distance between consecutive multiple- isoquants decreases. By doubling inputs, output
increases by less than twice of its original level.

c
3k
b 3X
2k 2.5X
a 2X
k
X 1.7x
O
L 2L 3L
Fig 3.16 Constant returns to scale: oa< ab < bc

Increasing returns to scale

The distance between consecutive multiple isoquants increases, by doubling the inputs, output is more
than doubled.
K
3.75X

3K1

c
B’ c 4.5 X
2K1
b
3X
a
K1 2.5X
2X
X
L
O
L1 2L1 3L1

Fig 3.17 Increasing Returns to Scale

ReturnsK to
Doubling andscale aretousually
L leads B’ whichassumed to be
lies above an the same denoting
isoquant every where on the production
2X(i.e.,2.5X),and triplingsurface
K and Li.e., the
results
sameinalong
an isoquant
all the which lies above
expansion 3X(i.e.,4.5X)
product lines. Alland so on. are assumed to show the same returns to
processes
scale over all ranges of output. Either constant returns to scale every where, or decreasing returns every
 where, or increasing returns everywhere. However, the technological conditions of production may be
such that returns to scale may vary over different ranges of output. Over some range, we may have
constant returns to scale, while over another range we may have increasing or decreasing returns to
scale.

K
3.75
XX

3K1 b
7X
c
2K1 c 6X
5X
a 4X
K1 3X
2X
X
L
O
L1 2L1 3L1

Fig3.18 Up to point C, increasing returns to scale prevails in the firm, from C to B constant
returns to scale prevails, and beyond B decreasing returns to scale prevails.

Causes of Increasing and Decreasing Returns to Scale

Technical and /or managerial indivisibility is one of the factor that leads to increasing returns to scale.
Mostly, processes of production can be doubled but it may not be possible to half them. When the
production system expands, workers will specialize in one extreme and their productivity increases.
On the other hand, the most common causes of decreasing returns to scale is ‘diminishing returns to
management’. If we expand the output beyond optimum, the top management personnel will be
overburdened and the productivity of additional unit of the variable inputs decline eventually. E.g.,
doubling fishing fleet may not double fish catch.
3.12 Technological process and production function
Technological improvement (progress) makes factors of production more productive or it makes
production system more efficient; so that the firm will get higher output from the same combinations
of labor and capital than before.
Graphically, this can be shown by upward movement of the total product curve (indicating higher
output level can be achieved from the same input) and down ward movement of isoquant denoting
lower combinations of factors of production can produce equal level of output. See the figures
TP2 TP after
technological
advancement
Isoquant before
TP before technological
technological advancement
K1
advancement
TP1
K2 Isoquant after
technological
advancement

L1 L2 L1
Fig 3.19 Technological progress shifts the TP curve up ward and the isoquant down ward
3.13 Equilibrium of the firm: Choice of optimal combination of factors of production

Dear learner, in our previous discussion we have said that an isoquant denotes efficient combination
of labor and capital required to produce a given level of out put. But, this does not mean that the
monetary cost of producing a given level of out put is constant along an isoquant. That is, though
different combinations of labor and capital on a given isoquant yield the same level of out put, the cost
of these different combinations of labor and capital could differ because the prices of the inputs can
differ. Thus, isoquant shows only technically efficient combinations of inputs, not economically
efficient combinations.
Technical efficiency takes in to account the physical quantity of inputs where as economic efficiency
goes beyond technical efficiency and seeks to find the least cost (in monetary terms) combination of
inputs among the various technically efficient combinations. Hence, technical efficiency is a necessary
condition, but not a sufficient condition for economic efficiency. To determine the economically
efficient input combinations we need to have the prices of inputs.
To determine the economically efficient input combination, the following simplifying assumptions
hold true:
Assumptions

1. The goal of the firm is maximization of profit (  ) where   TR  TC

Where  -Profit, TR- Total Revenue and TC-is total cost outlay.
 2. The price of the product is given and it is equal to PX .

3. The prices of inputs are given (constant).Price of a unit of labor is w and that of capital is r .
Now before we go to the discussion of optimal input combination (or economically efficient
combination), we need to know the isocost line, because optimal input is defined by the tangency of
the isoquant and isocost line.
Isocost line
Dear learner, do you remember what the budget line denotes?
Isocost lines have the same properties as that of budget lines. An isocost line is the locus points
denoting all combination of factors that a firm can purchase with a given monetary outlay, given prices
of factors.
Suppose the firm has C amount of cost out lay (budget) and prices of labor and capital are w and r
respectively. The equation of the firm’s isocost line is given as:
C  rK  wL , where K and L are quantities of capital and labor
respectively.
Given the cost outlay C , the maximum amounts of capital and labor that the firm can purchase are
C C
equal to and respectively. The straight line that connects these points is the isocost line. Note
r w
that just like a budget line; the slope of the isocost line is given by the factor price ratios. That is
w
Slope of isocost equals to  .See the following figure:
r
Capital

C/r

Isocost
line

Labor
C/w
Fig: 3.20: Isocost line: shows different combinations of labor and capital that the firm can buy
given the cost out lay and prices of the inputs.
Now we are in a position to determine the firm’s optimal in put combination. However, the problem
of determining optimal input combination (economic efficiency) takes two forms. Sometimes,
situations may happen when a firm has a constant cost outlay and seek to maximize its output, given
this constant cost outlay and prices of inputs. Still, there are also situations when the goal of the firm
is to produce a predetermined (given) level of output with the least possible cost. Below we will discuss
Isoquant: x=100kg
the two situations separately.
Case1: Maximization of Output Subject to Cost Constraint
Suppose a firm having a fixed cost out lay (money budget) which is shown by its isocost line. Here,
the firm is in equilibrium when it produces the maximum possible out put, given the cost outlay and
prices of inputs. The equilibrium point (economically efficient combination) is graphically defined by
A
the tangency of the firm’s isocost line (showing the budget constraint) with the highest possible
w MPL
isoquant. At this point, the slope of the isocost line ( ) is equal to the slope of the isoquant ( ).
r MPK
The condition of equilibrium under this case is, thus:

w MPL MPL MPK

 or 
r MPK w r
This is the first order (necessary) condition. The second order (sufficient) condition is that isoquant
must be convex to the origin (tangent to the isocost). See the following figure:

Capital

Q3
K1 E

Q2
Q1

L1 B Labor
Fig: 3.21: Optimal combination of inputs ( L1 and K1 )
Optimal combination of inputs is defined by the tangency of the isocost line (AB) and the highest
w
possible isoquant ( X 2 ), at point E. At this point the slope of isocost line ( ) is equal to the slope of
r
MPL
isoquant ( ).The second order condition is also satisfied by the convexity of the isoquant.
MPK
Mathematical derivation of the equilibrium condition
The problem can be stated as:
Maximize X  f ( L, K )......... .......... .......Objective function
Subject to C  wL  rK.......... .......... ......Constraint function
or C  wL  rK  C  0
We use the lagrangian method to solve the problem.
The lagrangian equation is written as:
  X   (C )
  
Then we find , , and and set all of them equal to zero to solve for L and K.
L K 
That is,
  X   (wL  rK  C)
 X MPL
And,   wL  0  MPL  w   
L L w
 X MPK
  r  0  MPK  r   
K K r

  wL  rK  w  0  wL  rK  C

Solving these equations simultaneously, we obtain the equilibrium condition
MPL MPK w MPL
 or 
w r r MPK
The second order condition (the convexity of isoquant) would be insured when:
2
2 X 2 X  2 X   2 X   2 X 
 0 ,  0 and  2   2 
   
L2 K 2  L   K   L K 
Example: Suppose the production function of a firm is given as X  0.5 L1 / 2 K 1 / 2 prices of labor and
capital are given as $ 5 and $ 10 respectively, and the firm has a constant cost out lay of $ 600.Find
the combination of labor and capital that maximizes the firm’s output and the maximum output.
MPL MPK MPL w
Solution: the condition of equilibrium is  or 
w r MPK r
X
MPL   0.25L1 / 2 K 1 / 2
L
X
MPK   0.25L1 / 2 K 1 / 2
K
Thus, the equilibrium exists when,
0.25L1 / 2 K 1 / 2 $5
1 / 2 1 / 2

0.25L K $10
K 1
  L  2 K .......... .......... .......... .....(1)
L 2
The constraint equation is:
wL  rK  C
5L  10K  600.......... .......... .......... ........( 2)
Solving equation (1) and (2) would give us the optimal combination of L and K.
L  2K
5L  10K  600
 L=60 units and K=30 units.
Thus, the firm should use 60 units of labor and 30 units of capital to maximize its production (out put).
(Check the second order condition).
The maximum output can be found by substituting 60 and 30 for L and K in the production process.
Case -2: Minimization of cost for a given level of output
In this case, consider an entrepreneur (a firm) who wants to produce a given output (for example a
bridge or a building or x tones of a commodity) with minimum cost outlay. That is, we have a single
isoquant which denotes the desired level of output, but there are a set of isocost lines which denotes
the different cost outlays. Higher isocost lines denote higher production costs. The production costs of
a desired level of output will therefore be minimized when the isoquant line is tangent to the lowest
possible isocost line (see fig) At the point of tangency, the slope of the isoquant and isocost lines are
w MPL
identical. That is 
r MPK

Capital
e

c
K1
a
E

L1 b d f
Labor

Fig: 3.22 Equilibrium Combination of factors

The equilibrium combination of factors is K1 and L1 amounts of capital and labor respectively. Lower
isocost lines such as ‘ab’ are economically desirable but unattainable given the desired level of output.
So point E shows the least cost combination of labor and capital to produce X amount of output.
Now let us see the mathematical derivation of the equilibrium condition. As mentioned earlier, we
minimize the cost of producing a given level of output.
Thus, the problem can be stated as:
 Minimize C = f (q) = WL + rK -------------------------------------- (Objective function)
Subject to q = f (L, K) ------------------------------------------------ (Constraint function)
Or f (L, K) – q = 0
We use the Lagrange an method to obtain the equilibrium condition. Accordingly, the Lagrange an
function will be:
  C   ( f ( L, K )  q )
  WL  rK   ( f ( L, K )  q )
  
The condition of equilibrium will be obtained by finding , and and then solving them
L K 
simultaneous after equating each to zero.
 f ( L, K )
That is,  w 0
L L
w
w  MPL  0   
MPL
 f ( L, K )
 r  0
K K
r
r  MPK  0   
MPK

 f ( L, K )  q  0

w r
Thus, the equilibrium condition is 
MPL MPK
MPL w
Rearranging the above condition, we obtain 
MPK r

 MPL W 
This condition    is only a necessary condition .
 MPK r 
The sufficient condition is that the isoquant must be convex to the origin. That is
2
 2q  2q   2 q   2 q    2 X 
 0,  0 and  2  2    
L2 k 2  L  k   LK 
Example
Suppose a certain contractor wants to maximize Profit from building a bridge. The contractor uses both
labor and capital, and efficient combinations of Labor and capital that are sufficient to make a bridge
 1 1
2 2
is given by the function 0.25 L K . If the prices of labor (w) and capital (r) are $ 5 and $ 10
respectively, find the least cost combination of L and K, and the minimum cost.
Solution: The contractor wants to build one bridge. Thus, the constraint equation can be written as
1 1
2 2
0.25 L k =1
1 1
MPL = 0.125 L 2 K 2

1 1
MPK = 0.125 L 2 K 2

MPL W
The equilibrium condition is 
MPK r
1 1
0.125L 2 K 2 $5
1 1

2 2
$10
0.125L K
K 1
  L  2K
L 2
Substituting L = 2K in the constraint equation we obtain
1 1
0.125 (2K) 2 K =1 2

0.125 2 . K=1
1 8
K= K
0.125 2 2
16
L = 2K 
2
16 8
Therefore, efficient combination (least cost combination) of L and K are and respectively.
2 2
 16   8  160
The least cost is C = 5   + 10  =$
 2  2 2
3.14 Cost minimization with varying out put levels and the derivation of long run total cost
curve
Dear learner, in the previous chapter we saw how a cost minimizing firm selects a combination of
inputs to produce a given level of output. Now we extend this analysis to see how the firm’s costs
depend on its output level. To do best with, refer to the following figure.
 Expansion
K path

40 80

70
B
3.5 60
A
3 50
2 30
20
10
O
3 5 6 L
Fig 3.23 Expansion Path
In the above figure, the firm is assumed to have increasing returns to scale up to point A (the distance
between successive multiple levels of output decreases), constant returns to scale between points A
and B (the distance between successive multiple levels of out puts is constant) and decreasing returns
to scale beyond point B (the distance between successive multiple levels of output increases).
The curve passing through the points of tangency between the firm’s isocost lines and its isoquants is
called expansion path. The expansion path denotes least cost combination of labor and capital required
to produce different levels of output.
To produce 10 units of output, the firm uses 2K and 3L, to produce 20 units, the firm uses 3K and 5L,
to produce 30 units it uses 3.5K and 6L, and so on.
Hence, as output expands at a constant amount, the units of labor and capital increases at a decreasing
rate up to point A. That is, total cost of production increases at a decreasing rate up to point A. From
point A to B, combination of labor and capital increase at a constant rate as output increases at a
constant rate. Hence, the long run total cost increases at a constant rate up to point B.
Beyond point B, to expand output at a constant rate, combination of labor and capital should be
increased at an increasing rate. Assuming that the prices of inputs are constant, the long run total cost
of production increases rapidly (at increasing rate) beyond point B.
From the above discussion, we infer that the long run total cost curve assumes an inverse S-shape.