CHAPTER SIX

THE THEORY AND ESTIMATION OF PRODUCTION

— For the manager of a profit-maximizing firm, seeking the maximum total

revenue may not ensure attaining maximum profits. Profit is the difference
between total revenues and total costs.
— Total revenue depends on price and quantity.
— For a given price, under competition, a successful manager should aim at
two intermediate targets leading to the final target of profit maximization:
o First, in the short run, to choose the optimal quantity of each input
o Second, in the long run, to choose the optimal size of the firm or the
production capacity that minimizes the cost
— The manager will be able to tackle these targets only if he has enough and
accurate knowledge concerning the production function or the input output
relationships, and that is the focus of this chapter.
— The first part of the chapter introduces the production function in short run
and long run, while the second part of the chapter is concerned with the
estimation of the production function.

The Production Function:

— What level of output should the firm produce depends on the production
function it has, among other things.
— The production function is the technical relationship, which shows the
maximum attainable production levels using different combinations of input
at a given state of technology within a given period of time.
— Mathematically, the production function can be expressed as
Q = f (L, K, N, T, …)

Page 1 of 29
 Where Q is level of output produced; L is the number workers (including
entrepreneurship); K is the capital; N is land; T is the state of technology;
and … refers to other inputs used in the production process
— The relationship between inputs and output assumes:
a. Fixed state of technology
b. Efficient use of input combinations
c. Given time period
— For simplicity we will often consider a production function of two inputs:
labor and capital. Labor and capital are both composite inputs that include
all other factors of production, so, the production function is normally
written in the follows implicit form:
Q = f (L, K)
— The above equation tells us that the first partial derivatives of output, with
respect to each of the inputs, are positive, i.e., Q has a positive relationship
with L and K. The second partial derivatives of output with respect to each
of the inputs are negative, indicating that the production function has some
maximum point.
— Production depends on the time-frame in which the firm is operating (short
run and long run)
— The short-run production function shows the maximum quantity of good
or service that can be produced by a set of inputs when at least one of the
inputs used remains unchanged.
— The long-run production function shows the maximum quantity of good
or service that can be produced by a set of inputs, assuming the firm is free
to vary the amount of all the inputs being used.

Page 2 of 29
Short-Run Analysis of Total, Average, and Marginal Product
— We use alternative terms in reference to inputs such as factors, factors of
production, and resources
— Also alternative terms are usually used in reference to outputs such as
quantity (Q), total product (TP), and product
— The relationship between output and the quantity of labor employed,
assuming fixed capital, can be described using the following three concepts:
o Total product
o Marginal product
o Average product

Total Product:
— Total product is the total output produced in a given period.
— In the short run, the total production curve shows the maximum output
produced using a certain set of variable inputs (as labor and row
materials…) in addition to one or more of fixed inputs (as the size of the
plant, equipments, area of land).
— However, because of our assumption for simplicity, the production function
depends only on labor and capital. Hence, to increase output in the short
run, a firm must increase the amount of labor employed.
— The total product curve shows how total product changes with the quantity
of labor employed
— The area below TP curve is attainable at every level of labor, while the area
are above TP is unattainable.
— Example:
To understand the nature of the relation between the different measures
of production in the short run, let us take a simple example of a small
farm where capital (area of the farm, water well, and equipments) is
fixed, and the number of workers is the only variable input.

Page 3 of 29
— Under these set of assumptions, total production (TP), marginal Product
(MP), and average product (AP), may be represented by the following
hypothetical values:

Number of Total Marginal Average product of

Workers Product Product of labor labor
(L) (TP) (MPL) (APL)
0 0 --------- ---------
1 8 8 8
2 18 10 9
3 29 11 9.67
4 39 10 9.75
5 47 8 9.4
6 52 5 8.67
7 56 4 8
8 52 -4 6.5

— From the table above, you may notice that total product increases as more
workers are hired. First, total product increases at high speed making big
jumps, then it slows down but still in creasing as it approach its maximum,
then total product starts to decrease as more labor are employed.
— In scientific terms we may describe the behavior of the total product in the
short run by saying that, TP increases first at an increasing rate to reach an
infliction (turning) point, after that it keeps increasing but at a decreasing
rate until it reaches its maximum level, then it starts to fall.
— The following graph reflects these phases of production:

Page 4 of 29
 TP

Stage I Stage Stage III

II

TP, AP,
MP then, TP, AP,
MP > AP MP, TP
MP < AP TP, AP,
a MP > 0 MP,
MP < AP
MP < 0

AP
L
0 L1 L2
MPL

The Marginal Product:

— The marginal product of labor (MPL) is the change in output (or Total
Product) that results from a one-unit change in the variable input (quantity
of labor employed), or one additional hour of work, with all other inputs
remaining the same.
Change in total output ∆TP ∆Q
MPL = = =
Change in the quantity of labor ∆L ∆L
— Graphically, MPL measures the slope of the total product curve.
— As you see in the graph above, the slope of the TP curve starts very close
to zero at the origin, then increases (production speeds up) to reach its
maximum at the infliction (turning) point (point a) on the TP curve, where
MPL curve reaches its maximum. The slope of TP curve decreases
gradually (production slows down) to reach zero at L2 as MPL falls to

Page 5 of 29
 intersects the L axis. Total product starts to fall from there on as the slope
of the MPL curve become negative.
— From the above table notice how the Marginal product of labor increases to
reach its maximum of 11, then falls after that to take a negative value
where total product starts to fall.
— Almost every production process has two features: initially, increasing
marginal returns (IMR), and eventually diminishing marginal returns (DMR).
— Initially increasing marginal returns
o When the marginal product of a worker exceeds the marginal
product of the previous worker, the marginal product of labor
increases and the firm experiences increasing marginal returns.
o Increasing marginal returns arise from increased specialization
and division of labor.
— Then, diminishing marginal returns
o Diminishing marginal returns arises from the fact that employing
additional units of labor means each worker has less access to
capital and less space in which to work.
o Diminishing marginal returns will start to take effect, when MP
starts to decrease.
o Diminishing marginal returns are so pervasive that they are
elevated to the status of a “law.”
o The law of diminishing returns states that as a firm uses more
of a variable input with a given quantity of fixed inputs, the
marginal product of the variable input at some point will eventually
falls, where each additional unit contribute less production than
the preceded unit”.
o The law of diminishing returns is a short run concept that
describes the change in the marginal product of the variable input.

Page 6 of 29
— It is important to keep in mind that all units of the variable input are
assumed of equal productivity, and the only reason for variations in its
marginal product (productivity) can be attributed to its order in utilization in
the production process.
— Referring to our previous example, at some early stage you may notice
increasing returns, where the MP curve has a positive slope. The reason
here is clear because the larger the number of workers is the higher the
productivity of individual workers due to specialization and teamwork
privileges.
— As more workers are added to the same quantity of the fixed input, at some
point diminishing returns will be in effect due to crowdness in the work
place, and inadequacy of the fixed input, as the number of workers
increase each worker will have less fixed input.
— In general:
o When MP is, firm is experiencing increasing marginal returns.
o When MP is but positive, firm is experiencing diminishing
(decreasing) marginal returns.
o When MP is and negative, firm is experiencing negative
marginal returns.
— Example:
Given a production function:
Q = 15X – 5X2,
Find the level of input at which diminishing marginal returns (DMRs)
starts
MP = dQ/dX = 15 – 10X
Set MP = 0 and solve for X
15 – 10X = 0 ⇒ X = 15/10 = 1.5

Page 7 of 29
The Average Product:
— The average product of labor (APL) refers to the share of each worker in
the total production.
— It is equal to total product divided by the quantity of labor employed. It is
the total Product per unit of input used.
TPL Q
APL = =
L L
— Example:
Given a production function:
Q 15 X 5 X 2
Q = 15X – 5X , then AP = =
2
− = 15 − 5 X = 0
X X X
⇒ X = 15/5 = 3
(Notice that AP reach 0 at twice L as MP)

The relationship between TP and MP

— Because of IMR, TP increases initially at increasing rate. Then, because of
DMR, TP increases at decreasing rate.
— The point that TP change its pace from increasing at increasing rate to
increasing at decreasing rate (i.e., the point where DMR start its course) is
called the turning point. When TPL is at its turning point MPL is at its
maximum,
1. If MPL is increasing and positive, TPL is increasing at increasing rate
2. If MPL is decreasing and positive, TPL is increasing at decreasing rate
3. If MPL = 0, TPL is at its maximum
4. If MPL < 0, TPL is decreasing

Page 8 of 29
The relationship between AP and MP
— When MP > AP (MP Curve above AP Curve), then AP is rising
o If your marginal grade in this class is higher than your grade point
average, then your GPA is rising
— When MP < AP (MP Curve below AP Curve), then AP is falling
o If your marginal grade in this class is lower than your grade point
average, then your GPA is falling
— When MP = AP, (MP Curve intersect AP Curve) then AP is at its maximum
o If your marginal grade in this class is similar to your grade point
average, then your GPA is unchanged

The Three Stages of Production in the Short Run:

— The relationship between TP, MP, and AP can be used to divide the SR
production function into three stages of production.
— Stage I:
o Starts from zero units of the variable input to where AP is maximized
(where APL = MPL).
o At this stage, AP is rising.
o In our example, stage I ends where four workers are being hired at
L1 on the figure above.
o For a profit-maximizing firm, it is irrational to limit production to any
level within the first stage. The rising AP throughout this stage
causes the average cost to fall and profits to rise as production
expands.
— Stage II:
o From the maximum AP to where MP=0 (or TP is maximized).
o At this stage, AP declining but MP is positive.
o In our example, Stage II is between L1 and L2 levels of employment.

Page 9 of 29
 o This is the rational stage of production, over which the firm manager
has to figure out the optimal number of workers that maximizes
profits.
— Stage III:
o From where MP=0 (or TP is maximum) onwards.
o At this stage, MP is negative or TP is declining.
o In our example, Stage III starts at L2 (MPL = 0) and above.
o Production in this stage is also irrational because the firm incurs
higher costs to hire more workers; while the total revenue is falling as
TP decreases.

At which stage should the firm operate?

— According to economic theory, in the short run, firms should operate in
Stage II. Specialization and team work continue to add more output when
additional variable input is used. Fixed input is being properly utilized ⇒
efficient use of resources
— Why not Stage I?
o Stage I indicates that the fixed input is underutilized since AP as
more and more of the variable input is used i.e., can increase output
per unit by increasing the amount of the variable input ⇒ inefficient
use of resources
— Why not Stage III?
o Firm uses more of its variable inputs to produce less output. Fixed
input is over utilized or overused ⇒ inefficient use of resources

Page 10 of 29
The Partial Elasticity of Production:

— In the short run, it is possible to measure the elasticity of production with

respect to each of the variable inputs.
— In a similar way to the concept of demand elasticity, production elasticity
measures the responsiveness of production to changes in one variable
input when other inputs are held constant.
— It measures the percentage change in the quantity produced due to one
percent change in labor input.
— It is known as the partial elasticity of production.
— The elasticity of production with respect to labor may be presented in the
following formula:
∆Q ∆Q
%∆Q ∆Q L ∆Q L
× = ∆L =
MPL
EL = = Q = × =
%∆L ∆L Q ∆L ∆L Q Q APL
L L
— By rearranging the basic elasticity formula, the elasticity of production
turned to be equal to the ratio of the marginal product to the average
product of the variable input.
— Look at the graphical illustration of the three stages of production; compare
the location of the marginal and the average product curves. You will easily
notice how the elasticity of production changes across the three stages of
production.
— In the first stage, when MPL > APL, then the labor elasticity, EL > 1.
o A one percent increase in labor will increase output by more than 1
percent.
— In the second stage, when MPL < APL, the labor elasticity 0< EL <1.
o A one percent increase in labor will increase output by less than 1
percent.
— In the third stage of production, when MPL is negative, EL< 0.

Page 11 of 29
— Using the above form of partial elasticity of production, the firm manager,
my easily find out the stage of his firm production. All he needs is
estimating the production functions and the quantity of the major variable
input used in production.
— The production elasticity of capital has the identical in form, except K
appears in place of L.
— Example:
Consider the production function, Q = 100 L – L2, find the stage of
production if the firm uses 20 workers.
Solution
MPL = 100 – 2L = 100 – 2 (20) = 60
APL = 100 – L = 100 – 20 = 80
EL = MPL / APL = 60 /80.
Since EL is positive and less than one, the firm is producing in stage II.
— Example:
If Q = 9L2 – L3
(Not that power 3 determines 3 stages)
a. Find the ranges of the 3 stages of production
b. Find at which stage the firm is operating if L= 5, and L = 3
c. Find L value at the starting of DMRs
Solution
a. Stage I: Between L= 0 and L where AP is maximized
Q 9L2 L3
AP = = − = 9L − L2
L L L
dAP/dL = 9 – 2L = 0 ⇒ L = 9/2 = 4.5
Stage I from 0 to 4.5
Stage II: From L= 4.5 to L where MP = 0
MP = dQ/dL = 18L – 3L2 = 0 divide by 3L
6–L=0⇒L=6

Page 12 of 29
 Stage II from 4.5 to 6
Stage III: 6 onward
b. Suppose L = 5, at which stage the firm is operating?
MP 18L − 3L2 18(5) − 3(5)2 15
= = = < 1 ⇒ at stage II
AP 9L − L2 9(5) − 5 2 20
Suppose L = 3, at which stage the firm is operating?
MP 18L − 3L2 18(3) − 3(3)2 27
= = = > 1 ⇒ at stage I
AP 9L − L2 9(3) − 3 2 18
c. Diminishing returns starts when MP reaches its maximum; i.e., when
dMP/dL = 0
MP = 18L – 3L2
dMP/dL=18 – 6L = 0 ⇒ L=18/6=3
— Example:
Q = 8L – 0.5L2
(Not that power 2 determines only 2 stages of production: II and III)
Law of diminishing returns starts immediately when production begins
Stage II: AP = 8 -0.5L MP, AP
dAP/dL = -0.5 ⇒ L= 0
Stage II starts where L = 0
Stage III: MP = dQ/dL = 8 – L = 0
⇒L=8
Stage III starts at 8
To find AP = 0
8 – 0.5L = 0 ⇒ L = 8/0.5 = 16
0 L
Notice that AP = 0 is at twice level II III
of L than when MP = 0 MP AP

— Exercise:
If Q = 50L + 6L2 - 0.5L3, find the 3 stages of production.

Page 13 of 29
Optimal level of the Variable input:
— Given that Stage II is the best for profit-maximizing firm, then what is the
optimal level of variable input should the firm use? In other words, at which
point on production function should the firm operate?
— The answer depends upon: how many units of output the firm can sell, the
price/ of the product, and the monetary costs of employing the variable
input.
— The Optimal quantity of the variable input is the quantity that allows the
firm to maximize its profits.
— The question facing the manager is: what is the optimal number of workers?
— To make things easy, let us assume that the firm buys its inputs and sell its
products in competitive markets, which means, it can hire any number of
workers at the market going wage (W) and sell any quantity of its product
at the market going price (P).
— Now we may present the firm profit function in the following form:
π = TR –TC = PXQ – (FC + WL)
Where: P is the Price of the good which is assumed constant, Q is the total
product, FC is the fixed input cost, W is the labor wage which is assumed
constant, and L is the number of workers.
— Now, the question is: how many workers the firm should hire in order to
maximize its profits? To find out the answer, we should take the first
derivative of the profit function with respect to L, set it equal to zero and
solve for the value of L as follows:
∂π ∂Q
=P* −W =0 or P * MPL = W
∂L ∂L
— Which says that, the optimal number of workers is that number at which the
value of the marginal product of labor is equal to the market wage rate; or
at which the marginal revenue product of labor (MRPL) equals the
marginal labor cost (MLC).

Page 14 of 29
— Total revenue product (TRP) refers to the market value of the firm’s
output, computed by multiplying the total product by the market price (Q * P)
— Marginal revenue product of labor (MRPL) is the change in the market
value of the firm’s output resulting from one unit change in the number of
workers used (UTRP/UL). It can be also computed by multiplying the
marginal product by the product price (MP *P)
— Total labor cost (TLC) refers to the total cost of using the variable input,
labor, computed by multiplying the wage rate (which assumed to be fixed)
by the number of variable input employed (W * L)
— Marginal labor cost (MLC) is the change in total labor cost resulting from
one unit change in the number of workers used
— Because the wage rate is assumed to be constant regardless of number of
inputs used, MLC is the same as the wage rate (w).
o If MRPL > MLC ⇒ π  as L 
o If MRPL < MLC ⇒ π  as L 
o If MRPL = MLC ⇒ π is maximum ⇒optimal input level
— In summary, A profit-maximizing firm operating in perfectly competitive
output and input markets will be using the optimal amount of an input at the
point in which the monetary value of the input’s marginal product is equal to
the additional cost of using that input (MRP = MLC) or (MRP = wage rate)
— As you may notice here again we are comparing marginal revenues and
marginal cost. Therefore, the decision rule is to hire more workers as long
as the value of the production contributed by the additional worker (P*MPL)
exceeds the cost of hiring an additional unit of labor which is equal to the
wage (W) paid to worker, under competition in the labor market.

Page 15 of 29
— Example:
Suppose the firm short run production function has the quadratic form: Q =
100L - L2,
The firm hires any number of workers at a wage of $80, and sells any
quantity of its output at a price of $2. Find how many workers should this
firm hire to maximize its profits.
Solution:
The optimal number of workers is reached when P*MPL = W
MPL = 100 - 2L
P*MPL = 2 * (100 – 2L) = 200 – 4L = 80
P*MPL = W ⇒ 200 – 4L =80 ⇒ 120 = 4L ⇒ L = 30 workers
— If the wage rate falls to $ 60, the optimal number of workers increases to 35
workers, which represents a movement along the MRPL curve.
— The MRPL curve thus shows the negative relationship between the wage
rate and the number of labor demanded.
— At a constant wage rate, the higher the price of the output is the greater the
number of workers demanded would be, for this reason, the demand for
labor (MRPL) is considered a derived demand from the demand for the
output.
— Exercise:
Q = 50L + 6L2 - 0.5L3
If the price is of the good is BD10 and the wage of each worker is BD400,
what is the level of production where profit will be at maximum level.

Page 16 of 29
The Optimal Mix of Variable Inputs:
— Suppose the firm production function has more than one variable input in
the short run. In order to maximize profits, the manager has to choose the
quantities of each input that will minimize cost.
— Let the production function be as follows:
Q = f (X, Y, K),
Where X and Y are the two variable inputs
— Let CX is the cost of X and CY be the cost of Y, while PQ is the price of the
firm output.
— For simplicity let us assume that the firm buys its inputs and sell its product
in competitive markets, where the firm can buy or sell any quantities at a
constant price.
— Now by taking the first derivative of the profit function with respect to each
of the variable inputs and equate it to zero we get:
π = TR - TC
= P * Q - (C X * X + C Y * Y )
∂π ∂Q
= PQ * − CX = 0 (1)
∂X ∂X
∂π ∂Q
= PQ * − CY = 0 (2)
∂y ∂Y
From (1) and (2)
PQ * MPX = C X (3)
PQ * MPY = C Y (4)
Divide (3) by (4)
MPX C X
= (5)
MPY C Y
This equation can re - written as
MPX MPY
= (7)
CX CY
— Thus, the optimal condition for a profit-maximizing firm employing two
variable inputs X and Y is attained where the ratio of the marginal products

Page 17 of 29
 of the variable inputs is equal the ratio of their costs. Or, the marginal
product of X to its cost must be equal to the marginal product of Y to its
cost.
— In other words, the marginal product per the amount of money spent on X
must be equal to the marginal product per the amount of money spent on Y.
— Political and economic risk factors may outweigh this relationship
— To get a better understanding of this condition, let us solve the following
simple example:
— Example:
o Suppose you are the production manager of a company that makes
computer parts in Malaysia and Algeria.
o At the current production levels and inputs utilization you found that:
ƒ Malaysian marginal product of labor (MPM ) =18 Units
ƒ Algerian marginal product of labor (MPA ) = 6 Units
ƒ Wage rate in Malaysia Wm = $ 6/hr
ƒ Wage rate in Algeria WA = $ 3/hr
o In which country should the firm hire more workers?
Solution
o Looking at the wage rates you might be tempted to hire more workers
and expand production in Algeria, where wages are relatively lower.
o However, production theory suggests that the firm should not only look
at input’s cost but also to the MP of each input relative to the cost.
o By examining the marginal product per dollar in each country, you will
find that:
MPM 18 MPA 6
= =3> = = 2;
WM $6 WA $3
Which means that: an additional dollar spent on labor in Malaysia would
yield 3 units, but would yield only 2 units if spent in Algeria.

Page 18 of 29
 o Therefore, the firm’s profit will rise as it shifts reallocate its labor budget
away from Algeria toward Malaysia. Shifting one dollar from hiring labor
in Algeria would reduce production by 2 units, if this dollar is spent on
hiring more labor in Malaysia it would generate 3 units.
o By this move, the firm would make a net gain of one unit. The firm would
continue to cut down its number of workers in Algeria and expand its
employment in Malaysia until equality is restored.
o As the firm increases employment in Malaysia, MPM would start to fall
following the law of diminishing returns. At the same time as the firm
reduces employment in Algeria, the law of diminishing returns will start
to raise MPA until eventually equilibrium is reached, and profit is
maximized.
— Example:
o Suppose labor and capital are both variable inputs and some other input
such as land is fixed, and suppose that
MPL = 12 units, MPk =24, w =$6 and r = $8,
Solution
o MPL /w = 12/6 =2 ⇒ spending one additional dollar on labor gives two
units of output
o MPK /r = 24/8 = 3 ⇒ spending one additional dollar on capital gives
three units of output
o So use more capital and less labor since capital is cheaper per dollar
spent than labor (capital is more productive)
o Bur, as more capital is used its MP, and as less labor is used its MP.
o This will continue until the two ratios are equal.
o Suppose MPL  to 15 and MPK  to 20, then
MPL/w = 15/6 = 2.5 and MPK/r = 20/8 = 2.5

Page 19 of 29
— Example:
o A firm has the following production function:
Q = 20 E – E2 + 12 T – 0.5 T2;
o Where E is the number of engineers and T is the number of technicians.
o The average annual salary for engineers is B.D. 9600; and for
technicians is B.D. 4800.
o The firm budget for hiring engineers and technicians is B.D. 336000 per
year.
o Calculate the optimal number of engineers and technicians.
Solution
o The budget constraint: 336000 = 9600E + 4800T (1)
o The optimization condition:
MPE MPT 20 − 2E 12 − T
= ⇒ = (2)
CE CT 9600 4800
o So, 10 – E = 12 - T or E = (T - 2) (3)
o By substituting (3) in (1) then;
3360 = 96E + 48T
3360 = 96 (T-2) + 48T
3360 = 96T – 192 +48T
144T = 3552
T = 3552/144 = 24.7
E = 24.7 – 2 = 22.7
Budget Constraint = 9600 (22.7) + 4800 (24.7) =336000

Page 20 of 29
The Long-Run Production Function
— As you know by now, the long run is a period of time long enough to allow
the firm to change all its inputs. Effectively, all inputs are variable.
— As the firm increases all its inputs in the long run, it actually changes the
scale of its production activity.
— The total elasticity of production, the increase in production in response to
the firm proportional increase in the scale of the production process is
called returns to scale.
— If all inputs into the production process are doubled, three things can
happen:
1. Output can be more than double, increasing returns to scale.
A larger scale of production allows the firm to divide tasks into more
specialized activities, thereby increasing labor productivity. It also
enables the firm to justify the purchase of more sophisticated (hence,
more productive) machinery. These factors help in explaining why
the firm can experience increasing returns to scale.
2. Output can exactly double, constant returns to scale.
3. Output can be less than double, decreasing returns to scale.
Operating on a larger scale might create certain managerial
inefficiency (e.g., communication problems, bureaucratic red tape)
and hence cause decreasing returns to scale.
— Graphically, the returns to scale concept can be illustrated using the
following graphs.
IRTS CRTS DRTS
Q Q Q

X,Y X,Y X,Y

Page 21 of 29
— One way to measure returns to scale is to use a coefficient of output
elasticity:
%∆ Q
EQ =
%∆ in all inputs
o If EQ > 1, production function shows increasing returns to scale (IRTS)
o If EQ = 1, production function shows constant returns to scale (CRTS)
o If EQ < 1, production function shows decreasing returns to scale (DRTS)
— Example:
If Q = 5L + 7K; and L = 10 & K = 10
Q1 = 5(10) + 7 (10) = 120 units
Now if each input increases by 25%, then L = 12.5 & K = 12.5
Q2 = 5(12.5) + 7 (12.5) = 150 units
%UQ = (150 - 120)/120= 25%
A 25% increase in L & K led to a 25% increase in Q ⇒ CRTS
— Example:
Q = 50X + 50Y +100
X = 1, Y = 1 ⇒ Q 50(1) + 50 (1) +100 = 200
If X =2, Y = 2 ⇒ Q 50(2) + 50 (2) +100 = 300
%UQ = (300 - 200)/200 = 50%
%U in all inputs = 100%
⇒ EQ = 50%/100% = 0.5 < 1 ⇒ DRTS
— Example:
Q = 50X2 + 50Y2
X=1, Y=1 ⇒Q = 50+50 =100
X=2, Y=2 ⇒Q = 200+200 =400
%UQ = (400 - 100)/100 = 300%
⇒ EQ = 300%/100% = 3 > 1 ⇒ IRTS

Page 22 of 29
Forms of Production Functions
1. Short run: existence of a fixed factor to which is added a variable factor
o One variable, one fixed factor: Q = f(L)K

2. Increasing marginal returns followed by decreasing marginal returns

o Cubic function: Q = a + bL + cL2 – dL3

3. Diminishing marginal returns, but no Stage I

o Quadratic function: Q = a + bL - cL2

4. Power function: Q = aLb

ƒ If b > 1, MP increasing
ƒ If b = 1, MP constant
ƒ If b < 1, MP decreasing
o Can be transformed into a linear equation when expressed in
logarithmic terms
logQ = log a + bLog L
5. Cobb-Douglas Production Function:
o Cobb and Douglas have introduced their production function in
1928.
o The Cobb-Douglas production function is a non-linear power or
exponential function in the following form:
Q = a Lb K1-b
o Both capital and labor inputs must exist for Q to be a positive
number
o b and 1- b are the elasticities of production with respect o labor and
capital.
o b and 1- b are constants. b + b -1 = 1 ⇒ CRTS

Page 23 of 29
o In later version, Cobb-Douglas relaxed this requirement and
rewrote the equation as follows
Q = aLbKc
o Can be increasing, decreasing, or constant returns to scale
ƒ b + c > 1, IRTS
ƒ b + c = 1, CRTS
ƒ b + c < 1, DRTS
o Permits us to investigate MP for any factor while holding all others
constant
o Each of the coefficients is usually less than one showing that the
production takes place in stage two.
o Each exhibits diminishing marginal returns
o b and c represents the partial elasticity of production:
b = MPL/APL, c = MPK/APK
So b can be found using MPL and APL. Same thing for C.
o Can be estimated by linear regression analysis
log Q = log a +b log L + c log K

o Can accommodate any number of independent variables

Q = a X1b X2c X3d …. Xkn
o Does not require that technology be held constant
o Shortcomings:
ƒ Cannot show MP going through all three stages of production.
Cubic function is necessary
ƒ Cannot show a firm or industry passing through increasing,
constant, and decreasing returns to scale
ƒ Specification of data to be used in empirical estimates`

Page 24 of 29
— Example:
Q = 75 X0.25 Y0.75
X=1, Y=1 ⇒Q = 75*1*1 =75
X=2, Y=2 ⇒Q = 75*1.19*1.68 = 149.94 = 150
%UQ = (150 - 75)/75 = 100%
⇒ EQ = 100%/100% = 1 ⇒ CRTS
— For Cobb-Douglas production function
o Sum of Exponents = 1 ⇒ CRTS
o Sum of Exponents > 1 ⇒ IRTS
o Sum of Exponents < 1 ⇒ DRTS

Page 25 of 29
Choosing the Optimal Capacity: (Reading)

— Production levels do not depend on how much a company wants to

produce, but on how much its customers want to buy.
— Although the manager may be well informed about the three stages of
production for his firm, in the short rum, he might realize that the market
demand is consistently lower than expected.
— Capacity Planning: refers to the planning the amount of fixed inputs that
will be used along with the variable inputs. Good capacity planning
requires:
o Accurate forecasts of demand
o Effective communication between the production and marketing
functions
— Suppose the firm is forced to produce in the first stage as a result of the
inadequate demand in the short run. In this case the firm is underutilizing
its capacity or its fixed inputs, in other words, the firm would have excess or
idle capacity. Fixed costs are the cost of fixed inputs, that part of the firm
total cost, which is independent of the level of production. A firm producing
in stage one with some idle capacity, would have high average cost (per
unit cost), and therefore, will not maximize profits.
— In the long run, such firm should consider reducing its production capacity
to the optimal size that enables it to maximize profits
— By the same token, a firm facing a growing demand beyond its production
capacity should consider expanding its capacity in the long rune to meet
the market demand while producing in the rational range that would allow
maximum profits.

Page 26 of 29
Statistical Estimation of Production Functions (Reading)
Model specification;
— Specifying the model means to choose the exact explicit equation that
would best illustrate the production function of the firm. To make the
appropriate choice, the researcher may plot output against each of the
independent variables to find out in which stage the firm is producing. It is
possible to estimate the production function in the short run when at least
one input is constant, if it takes a form that shows all the three stages of
production, as in the case of
Q = 200 + 0.2L + 0.25 L2 – 0.1L3,
or if it shows only two stages as in the case of
Q = 200 +0.2L - 0.25 L2
which shows only the second and the third stages of the production.
— The most popular form of the production function is the power function. In
its implicit form, the power function with n variables is:
Q = (AX1bX2c……..Xnm),
which has the following appealing properties:
1. It can be transformed into a log-linear form that suits the use of LS
method for linear regression analysis.
2. In the short run, it allows the estimation of the marginal product of one
variable holding the others constant.
3. In the long run, it can be used to estimate returns to scale, as all inputs
are variable.
4. The powers measure the partial elasticity of production with respect to
each variable input. If b>1, then marginal product exceeds the average
product and the later is rising in the first stage. If 0<b<1, the marginal
product is less than the average product, and the later is falling in the
second stage. If b<0, the marginal product is negative and the
production takes place in the third stage.

Page 27 of 29
Data measurement problems;
1. In studies at the micro levels, i. e., at the firm or a group of firms’ level,
researchers depend on firm’s records as the main source of data.
However, in studies at the macro level, sectoral or aggregate economy
level, data will be obtained from published national accounts statistics
(GDP, GDP by sector, Employment, Investment, Inflation rate, the stock
of real capital….).
Gathering data for aggregate functions can be difficult.
2. For a firm producing one output, the output Q is normally measured in
physical units (tons, barrels, ..). But for a firm producing more than one
product with different measurement units, or in the case of macro
studies that use aggregate data of output, the output in physical units
should be converted into values using per unit costs or market retail or
whole sale products prices. However, changes in these values should
reflect the changes in the quantities demanded of the products.
Therefore, such values must be deflated annually to eliminate any
inflation distortion from the output data.
3. Inputs should be measured as “flow” rather than “stock” variables, which
is not always possible.
o Usually, the most important input is labor.
o Most difficult input variable is capital.
— Inputs should be measured as “flow variables” (for instance, Gallons
per month) to mach the way in which output is measured. However,
sometimes input data may not be available as flows, particularly in
the case of capital, where data on the flow of capital used in the
production process might not be readily available. Instead, annual
depreciation of capital may be used. But again depreciation taken
from the firm books is not an accurate measure of the capital input
used in production; it is always based on some accounting

Page 28 of 29
 techniques, which are mostly affected by the tax law. Furthermore,
some fixed assets are not depreciated at all like land. As a way out
of these problems, researchers use gross capital stock (the historical
value of plant and equipments) or net capital stock (gross capital
minus accumulated depreciation).

Time Series versus Cross Section Data:

— If the researcher uses time series data, it is recommended to include a time
variable or a dummy variable to capture the effect of technological
improvements on output. It is also important to adjust data measured in
monetary units to the rate of inflation.
— In cross section studies, researchers should take into consideration the
variations between firms using levels of technology, and the should adjust
variables measured in monetary terms to price and wage differentials
between various geographical regions.

Page 29 of 29