Managerial Economics, 8e

William F. Samuelson ● Stephen G. Marks

CHAPTER FIVE

Production
A GRAPHICAL APPROACH

Consider once again the production function of Example

3:
Q =40L - L2 + 54K - 1.5K2. We saw that the firm could
produce Q =636 units of output using L = 10 and K = 8
units of inputs. The same output, Q = 636, can be
produced using different combinations of labor and
capital: 6 units of labor and 12 units of capital, for
instance.

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
 An isoquant is a curve that shows all possible
combinations of inputs that can produce a given level
of output. The isoquant corresponding to Q = 636 is
drawn in Figure 5.2a.
 The amounts of the inputs are listed on the axes. Three
input combinations along the Q = 636 isoquant, (L =6,
K =12), (L = 10, K = 8), and (L =14.2, K = 6), are
indicated by points A, B, and C
 respectively. A separate isoquant has been drawn for
the output Q =800 units. This isoquant lies above and
to the right of the isoquant for Q = 636 because
producing a greater output requires larger amounts of
the inputs.

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
FIGURE 5.2
ISOQUANTS AND ISOCOST LINES

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
 The isoquant’s negative slope embodies the basic trade-off
between inputs. If a firm uses less of one input, it must use more
of the other to maintain a given level of output.
 For example, consider a movement from point B to point A in
Figure 5.2a—a shift in mix from (L =10, K =8) to (L = 6, K =
12). Here an additional 12 - 8 = 4 units of capital substitute for 10
- 6 =4 units of labor.
 But moving from point B to point C implies quite a different
trade-off between inputs. Here 4.2 units of labor are needed to
compensate for a reduction of only 2 units of capital. The
changing ratio of input requirements directly reflects diminishing
marginal productivity in each input. As the firm continually
decreases the use of one input, the resulting decline in output
becomes greater and greater. As a result, greater and greater
amounts of the other input are needed to maintain a constant level
of output. K=6 , L= 14.2
Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
 MARGINAL RATE OF TECHNICAL
SUBSTITUTION (MRTS)
 MRTS refers to the rate at which one input factor is substituted
with the other to attain a given level of output. The MRTS is the
slope of the Isoquant curve.
 It is measured as:
MRTS = changes in one input/changes in another input

Output Capital Labor

MRTS Combinations
(quantity) (Units) (Units)
---- 50 10 1 A
1:3 50 7 2 B
1:3 50 4 3 C
1:1 50 3 4 D
1:2 50 1 5 E
Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
 The general rule is that the slope of the isoquant at any
point is measured by the ratio of the inputs’ marginal
products: 
 ΔK/ΔL (for Q constant )= -MPL/MPK.
 Notice that the ratio is -MPL/MPK and not the other
way around. The greater is labor’s marginal product
(and the smaller capital’s), the greater the amount of
capital needed to substitute for a unit of labor, that
is, the greater the ratio ΔK/ΔL. This ratio is important
enough to warrant its own terminology. The marginal
rate of technical substitution (MRTS) denotes the
rate at which one input substitutes for the other and
is defined as: 
 MRTS = -ΔK/ΔL (for Q constant) =- MPL/MPK.
Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
 ISOCOSTS (EQUAL COSTS OF
PRODUCTION)
 Isocosts refers to that cost curve that represents
the combination of inputs that will cost the
producer the same amount of money.
 In other words, each Isocost denotes a particular
level of total cost for a given level of production.
If the given level of production changes, the total
cost changes and thus the Isocost curve moves
upwards. And vice versa.
 The slope of Isocost line = PL/Pk

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
 GRAPHING THE ISOCOST CURVE
 In the opposite figure, there are
three downwards sloping:
 straight line cost curves each
costing 1.0 $,1.5 $ and 2.0$ for
the output levels of 20,000,
30,000 and 40,000 units.
Isocosts farther from the origin,
for given input costs, are
associated with higher costs.
 Any change in input prices
changes the slope of isocost
lines

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
AT THE TANGENCY OF THE ISOCOST &
ISOQUANT, THE SLOPES OF THOSE
CURVES ARE EQUAL.
We found previously that:
the slope of the isoquant is MPL/MPK ,
& the slope of the isocost is PL/PK .
So at the tangency, MPL/MPK = PL/PK
This expression is equivalent to MPL/PL =
MPK/PK .

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
MAXIMIZING OUTPUT FOR A GIVEN
COST LEVEL
At points A & B, we’re
isoquants
spending the outlay
associated with this Isocost,
K
but we’re not producing as
much as we can. We’re only
making Q1 units of output.
A
We can’t produce Q3 or Q4
with this outlay. Those output
levels would cost more.
isocost E At point E, we’re producing
the
Q4 most for the money, where
Q3the Isocost is tangent to an
Q2
B isoquant.
Q1
L
Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
MINIMIZING COST FOR A GIVEN OUTPUT
LEVEL
At points A & B, we’re producing the
desired quantity, but we’re not using
the cheapest combination of inputs,
so we’re spending more than
K necessary.
We can’t produce the desired output
level at cost level C1. We need more
A money.
At point E, we’re producing the
desired output at the lowest cost,
E
where the isoquant is tangent to an
Isocost.

B Q1 isoquant

L
isocosts C1 C2 C3
Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
 Note that at the point of tangency, the slope of the isoquant
and the slope of the isocost line are the same. The isoquant’s
slope is -MPL/MPK. In turn, the isocost line’s slope is
-PL/PK. Thus, the least-cost combination of inputs is
characterized by the condition
MRTS = MPL/MPK =PL/PK.
 The ratio of marginal products exactly matches the ratio of
input prices. (If one input is twice as expensive as another,
optimal usage requires that it have twice the marginal
product.) This relationship can be rearranged to read
MPL/PL = MPK/PK.

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
OTHER PRODUCTION DECISIONS
Within the limits of its production technology, the firm’s
managers face a number of important decisions. We have
already discussed finding the optimal use of single input
in the short run and choosing the best mix of inputs in the
long run.
We now consider two other decisions:
(1) the allocation of a single input among multiple
production facilities and
(2) the use of an input across multiple products

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
 MULTIPLE PLANTS
 Consider an oil company that buys crude oil and
transforms it into gasoline at two of its refineries.
Currently it has 10 thousand barrels of oil under long-
term contract and must decide how to allocate it between
its two refineries. The company’s goal is to allocate
supplies to maximize total output from the refineries. Let
MA and MB represent the crude input at each refinery
and QA and QB the gasoline outputs. The firm’s problem
is:
 Maximize Q =QA + QB, subject to MA+MB = 10
thousand

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
 The key to maximizing total output is to compare
marginal products at the two refineries. Barrels of crude
first should be allocated to the refinery at which the
marginal product is greater. Let’s say this is refinery A.
As additional barrels are allocated to this refinery, its
marginal product diminishes, and it becomes worthwhile
to allocate oil to refinery B as well. In the final
allocation of all 10 thousand barrels, output is
maximized if and only if the marginal products of both
refineries are equal, that is, when
MPA = MPB.

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
 
 Why must this be the case? If marginal products differed
(say, MPA< MPB), barrels should be shifted from the
low-MP plant (refinery A) to the high-MP plant (refinery
B).
 Let’s apply this rule in a specific example. Based on
extensive studies, suppose that management has estimated
the following production functions for the refineries:
 Refinery A: QA = 24MA - 0.5MA 2
 Refinery B: QB = 20MB - MB2,

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
 where gasoline outputs are measured in
thousands of gallons and quantities of crude oil
are measured in thousands of barrels. Marginal
products are

 Refinery A: MPA = 24 – MA
 Refinery B: MPB =20 - 2MB.
the marginal product curve for each refinery and
two possible allocations

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
 One is a naive allocation calling for an equal
split between the two facilities: MA=MB=5
thousand barrels.
 Using the production functions, we find total
output to be 182.5 thousand gallons. However,
the figure immediately points out the
inefficiency of such a split.
 At this division, the marginal product of the
last barrel of crude at refinery A greatly
exceeds the marginal product of the last barrel
at refinery B (19 > 10). Barrels should be
reallocated toward refinery A.

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
 We can readily identify the optimal solution
from Figure 5.4:
 MA =8 thousand barrels and MB = 2 thousand
barrels. At these allocations, each refinery’s
marginal product is 16. (To check this, refer to
the marginal product expressions just given.)
The total output of gasoline from this
allocation is 196 thousand gallons—a
considerable improvement on the naive
allocation. Furthermore, no other allocation can
deliver a greater total output.

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
We can find these allocations directly using the
facts that MPA=MPB and MA+MB=10.
Equating marginal products implies
24 -MA = 20 - 2MB. Solving this equation and
the quantity constraint simultaneously gives the
solution, MA = 8 thousand barrels and MB =2
thousand barrels.

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
EXAMPLE
Enpar manufactures engine parts for an
automotive manufacturer. It operates two plants,
A and B, which have the following production
functions:
QA = 30SA – 0.25SA2 and QB = 40SB – 0.5SB2,
where QA and QB denote the outputs of engine
parts from each plant and SA and SB denote the
amounts of steel used in each plant. Suppose that
total steel availability is 40 units. What is the
optimal allocation of steel between the two
plants?
Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
ANSWER: Enpar maximizes profit when the
marginal product for steel is equal at the two
plants. The marginal product of steel at plant A is
given by: MPA = 30 – 0.5SA. The marginal
product of steel at plant B is: MPB = 40 – SB.
Equating MPA and MPB gives:
30 – 0.5SA = 40 – SB. In addition, the availability
constraint is: SA + SB = 40,
so that SB = 40 – SA. Substituting this last
equation into MPA = MPB yields the single
equation: 30 - 0.5SA = 40 – (40 – SA), so 30 =
1.5SA. Thus, SA = 20 and SB = 20
Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
MULTIPLE PRODUCTS
 Firms often face the problem of allocating an input in
fixed supply among different products. The input may
be a raw material—for instance, DRAM computer
chips allocated to the various models of personal
computers manufactured by the firm—or it may be
capital. Frequently, the input in shortest supply is
managerial labor itself. Which products of the firm are
in greatest need of managerial attention? Which top-
level managers are best suited to improve performance
in a given product line?

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
Consider a variation on the oil company’s
decision. Suppose two of the company’s product
managers are engaged in a heated debate. The
first manager oversees the company’s production
and sale of gasoline; the second is responsible for
production of synthetic fiber. Both products use
crude oil as an essential input. The problem is
that the current demands of the managers for this
input exceed the firm’s available crude oil
supply, 20 thousand barrels. Each manager is
arguing for a greater share of the input.

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
How can economic analysis be used to resolve this
dispute? Given a limited resource and two products,
gasoline and fiber, management’s ultimate goal must
be to allocate the crude oil to each product (in
quantities MG and MF) to maximize total profit
subject to the constraint of 20 thousand total barrels.
The form of this decision is very similar to that of the
multiplant decision. Here total profit is maximized if
and only if the input is allocated such that the
products generate identical marginal profits per unit
of input.
 MπG = MπF.
Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
If fibers had a higher marginal profit per unit input than
gasoline, gallons of crude should be switched from gasoline
production to fiber production. Here is a concrete example.
Suppose the production functions are
Gasoline: G = 72MG - 1.5MG 2
Fiber: F = 80MF - 2MF 2
Here gasoline output is measured in thousands of gallons, fiber output in
thousands of square feet, and crude oil in thousands of barrels. The
products’ profits per unit output are $0.50 per gallon for gasoline and
$0.75 per square foot for fiber. Then the respective marginal profits are
MπG = ($0.50)MPG =($0.50)(72 - 3MG) =36 - 1.5MG
MπF = ($0.75)MPF = ($0.75)(80 - 4MF) = 60 -3MF.
Setting these equal to each other and rearranging gives
MF =0.5MG + 8.

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
Solving this equation and the constraint MG +
MF = 20 implies MG = 8 thousand barrels and
MF = 12 thousand barrels. This allocation
generates 480 thousand gallons of gasoline and
672 thousand square feet of fiber. The firm’s total
profit is $744 thousand (less the cost of the
crude).

Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
 FINAL REMARKS With respect to both the plant and
product decisions, two comments are in order.
First, the appropriate marginal conditions are extended easily to
the case of multiple (more than two) plants and decisions. (For
instance, if there are three plants, the marginal product condition
becomes MPA = MPB = MPC.)
Second, the decision framework changes significantly if
management is able to vary the amount of the input. If
management has access to as much crude oil as it wants (at a
price), the problem can be dealt with plant by plant or product
by product. Indeed, we have already considered the solution to
this decision earlier: For each plant or each product, use of the
input should be expanded to the point where its marginal
revenue product equals its marginal cost per unit input (i.e., the
input’s price).
Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
 EXAMPLE
Sleak Teak builds yard furniture, using domestic hardwoods
and (in a smaller shop) handcrafted knick-knacks from the
same sort of wood. Although hardwoods were readily available
in the past, recently they have been much more difficult to
obtain. Consultation with the plant managers of the two lines
has resulted in the following production functions for hardwood
usage in the two products. For yard furniture:
Y = 2TY – .001TY2, while for Knick-knack: K = 20TK – .01TK2.
Here Y and K denote units of the two types of products and TY
and TK denote teak used in yard furniture and knick-knacks
respectively. Yard furniture can be sold at a profit of $100 per
unit, and knick-knacks can be sold at a profit of $25 each. Sleak
Teak has 1,300 units of teak available. How should it be
allocated across the two products? Explain.
Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.
ANSWER: The marginal products of teak in the
two goods are: MPY = 2 – .002TY and MPK = 20
– .02TK. The marginal profits of the two
activities are: MY = (100)MPY = 200 – .2TY, and
MK = (25)MPK = 500 – .5TK. Equating the two
marginal profits: 200 – .2TY = 500 – .5TK. In
addition, TY + TK = 1,300.
Therefore, 200 – .2TY = 500 – .5(1,300 – TY), or
350 = .7TY. Thus, TY = 500 units. In turn, TK =
800 units. From the production functions: Y =
750 units and K = 9,600 units.
Copyright © 2015 John Wiley & Sons, Inc. All rights reserved.