Chapter 4

Labour Demand

McGraw-Hill/Irwin
Labor Economics, 4th edition Copyright © 2008 The McGraw-Hill Companies, Inc. All rights reserved.
 4-2

Introduction

• Firms hire workers because consumers want to purchase a

variety of goods and services.
• Therefore, demand for workers is derived from the wants and
desires of consumers (it is ‘derived demand’).
• Central questions: How many workers are hired and what are
they paid?
 4-3

4.1 The Firm’s Production Function

• The production function describes the technology that the firm uses to
produce goods and services.
• Assume only two production factors: The firm’s output is produced by any
combination of capital (land, machines, etc.) and labour (employee hours
hired by the firm; if working hours constant, also simply the number of
workers hired). q=f(E,K)
• Note that labour is assumed to be homogenous (and so is capital).
• The marginal product of labour MPE is the change in output resulting
from hiring an additional worker, holding constant the quantities of other
inputs.
• The marginal product of capital MPK is the change in output resulting
from a one unit increase in capital, holding constant the quantities of other
inputs.
 4-4

More on the Production Function

• Marginal products of labour and capital are positive, so as more

units of each are hired, output increases.
• When firms hire more workers, total product rises.
• The slope of the total product curve is the marginal product of
labour.
• Law of Diminishing Returns: Eventually, the marginal
product of labour declines.
- Average product of labour APE: The amount of output
produced by the typical worker, i.e. q/E.
 4-5

Figure 4.1: The Total Product, the Marginal

Product, and the Average Product Curves
140 25

120 Average Product

20
100
Output

Output
80 15

60 Total Product
10
Curve
40
5 Marginal Product
20

0 0
0 2 4 6 8 10 12 0 2 4 6 8 10 12

Number of Workers Number of Workers

The total product curve gives the relationship between output q and
the number of workers hired by the firm E (holding capital fixed).
The marginal product curve gives the output produced by each
additional worker, and the average product curve gives the output per
worker.
 4-6

Marginal and Average Curves

• Note the following rule:

“The marginal curve lies above the average curve when

the average curve is rising, and the marginal curve lies
below he average curve when the average curve is
falling.”

This implies that “the marginal curve intersects the average

curve at the point where the average curve peaks”.
 4-7

Profit Maximisation

• We assume the objective of the firm is to maximise profits.

• The profit function is:
- Profits = pq – wE – rK
- Total Revenue = pq
- Total Costs = (wE + rk)

• In this chapter it is assumed that the firm is perfectly

competitive, i.e. it cannot influence prices of output or inputs
(they are assumed constant, irrespective of what the firm does)
• How much E and K should the firm hire?
 4-8

4.2 The Employment Decision

in the Short Run
• Definition of short-run: K fixed.
• Value of Marginal Product (VMP) of labour is the marginal
product of labour times the dollar value of the output:
VMPE=p∗MPE
• This indicates the $ increase in revenue generated by an
additional worker, holding capital constant.
• Value of Average Product (VAP) of labour is the dollar value of
output per worker: VAPE=p ∗APE
• A profit-maximizing firm hires workers up to the point where
VMPE=w (and VMPE is declining). This is the marginal
productivity condition. At that point the marginal gain due to
an additional worker is equal to the cost of the worker.
 4-9

Figure 4.2: The Firm's Hiring Decision

in the Short-Run

38
A profit-maximizing
firm hires workers up
VAPE to the point where the
wage rate equals the
22
value of marginal
VMPE product of labour. If the
wage is $22, the firm
hires eight workers.
8 Number of Workers
1 4
 4 - 10

The Short-Run Labour Demand Curve

• The short-run demand curve for labour indicates what happens

to the firm’s employment as the wage changes, holding capital
constant.
• The curve is downward sloping. It is the downward-sloping part
of the VMPE curve below its intersection point with the VAPE
curve.
 4 - 11

Figure 4.3: Short-Run Demand Curve for Labour

Because marginal product

eventually declines, the
short-run demand curve for
labour is downward
sloping. A drop in the wage
22
from $22 to $18 increases
18
the firm’s employment. An
increase in the price of the
VMPE
output (or an increase in
VMPE
worker productivity) shifts
the value of marginal
8 9 12 product curve upward (i.e.
Number of Workers
outward), and increases
employment.
 4 - 12

The Short-Run Labour Demand Curve

for the Industry

• We do not get the industry demand curve for labour by

adding up individual firms’ labour demand curves
horizontally.
• Reason: For each firm the output price is given (i.e. the firm
cannot change it), but if all firms in an industy expand their
output, the output price will be reduced, and so will be the
VMPE (=p∗MPE) of each firm, shifting each firm’s labour
demand curve left.
 4 - 13

Figure 4.4: The Short-Run Labour Demand

Curve for the Industry

Wage Wage

D
20 20

10 10
D

15 28 30 Employment 30 56 60 Employment
 4 - 14

The short-run elasticity of labour

demand

• To measure the responsiveness of employment in an

industry to changes in the wage rate, we can calculate the
elasticity of labour demand:
- The percentage change in employment divided by the
percentage change in the wage, or

ΔE w
σ SR
=
Δw
SR
∗
E SR
 4 - 15

An Alternative Interpretation of the

Marginal Productivity Condition

• More familiar profit-maximizing rule for perfectly

competitive firms: Produce output up to the point where
Marginal Cost (MC) equals output price p (i.e. Marginal
Revenue MR).
• This is the same as the ‘marginal productivity condition’
derived earlier:
The cost of producing an extra unit of output equals:
MC = w ∗ 1/MPE

If we set this equal to P and re-arrange, we get the

marginal productivity condition:
w ∗ 1/MPE = p ⇒ w = p ∗ MPE
 4 - 16

Figure 4.5: The Firm's Output Decision

Dollars A profit-maximizing firm

produces up to the point
MC
where the output price
equals the marginal cost of
production. This profit-
maximizing condition is
p Output Price
the same as the one
requiring firms to hire
workers up to the point
where the wage equals the
value of marginal product
Output of labour.
q*
 4 - 17

Critique of Marginal Productivity

Theory

• A common criticism is that the theory bears little relation to

the way that employers make hiring decisions.
• Another criticism is that the assumptions of the theory are
not very realistic.
• However, employers act as if they know the implications of
marginal productivity theory (hence, they try to make
profits and remain in business).
 4 - 18

4.3 The Employment Decision in

the Long Run
• In the long run, the firm maximizes profits by choosing how
many workers to hire AND how much plant and equipment to
invest in.
• An isoquant describes the possible combinations of labour
and capital that produce the same level of output (the curve
“ISOlates the QUANTity of output). Isoquants…
- must be downward sloping
- do not intersect
- that are higher indicate more output
- are convex to the origin
- have a slope that is the negative of the ratio of the marginal products of
labour and capital. The absolute value of this is called the marginal
rate of technical substitution.
 4 - 19

Isoquant Curves
Capital

All capital-labour
combinations that lie along
a single isoquant produce
the same level of output.
X
The input combinations at
ΔK points X and Y produce q0
Y
q1
units of output. Input
q0
combinations that lie on
higher isoquants produce
more output.
ΔE Employment
 4 - 20

Isocost Lines

• The isocost line indicates the possible combinations of

labour and capital the firm can hire given a specified
budget.
• An isocost line indicates equally costly combinations of
inputs (i.e. they indicate a particular value of total cost).
• Higher isocost lines indicate higher costs.
• The firm’s costs are: C=wE + rK. Re-write so that this
function can be drawn in K, E space:
C w
K = − E
r r
 4 - 21

Isocost Lines

Capital

All capital-labour
C1/r combinations that lie along
a single isocost curve are
Isocost with Cost Outlay C1
C0/r equally costly. Capital-
labour combinations that lie
Isocost with Cost Outlay C0
on a higher isocost curve
are more costly. The slope
of an isocost line equals the
ratio of input prices (-w/r).
C0/w C1/w Employment
 4 - 22

Cost Minimization

• Profit maximization implies cost minimization.

• The firm chooses the least cost combination of capital and
labour to achieve its profit-maximizing output level
(determined by MC=p).
• This least cost choice is where the isocost line is tangent to the
isoquant.
• At that point, the marginal rate of (technical) substitution
equals the price ratio of capital to labour.
MP E
=
w
MP K
r

• This condition can also be derived from ‘value marginal

product of an input equals its price’, e.g. from w=p*MPE and
r=p*MPK.
 4 - 23

The Firm's Optimal Combination

of Inputs (for a given level of output)
Capital
A firm minimises the costs of
C1/r
producing q0 units of output
A by using the capital-labour
C0/r
combination at point P,
where the isoquant is tangent
to the isocost. All other
P
175 capital-labour combinations
(such as those given by
B points A and B) lie on a
q0
higher isocost curve.
100 Employment
 4 - 24

4.4 Long Run Demand for Labour

• What happens to the firm’s long-run demand for labour when

the wage changes?
• If the wage rate drops, two effects take place that increase the
firm’s labour demand:
- Firm takes advantage of the lower price of labour by expanding
production (scale effect).
- Firm takes advantage of the wage change by rearranging its mix of
inputs (while holding output constant)(substitution effect).
• If labour is cheaper, the firm will increase its demand for labour.
What happens to the demand for capital depends on the relative
strength of scale (increased demand for capital) and substitution
effects (reduced demand for capital).
 4 - 25

Figure 4.9: The Impact of a Wage Reduction, Holding

Constant (Unrealistically!) Initial Cost Outlay at C0

Capital

A wage reduction flattens the

isocost curve. If the firm were to
C0/r
hold the initial cost outlay
constant at C0 dollars, the isocost
would rotate around C0 and the
R
75 P firm would move from point P to
point R. A profit-maximizing
′
q0 firm, however, will not generally
want to hold the cost outlay
q0 constant when the wage changes,
Wage is w1 i.e. Figure 4.9 is most likely
Wage is w0
wrong.
25 40
 4 - 26

Figure 4.10: The Impact of a Wage Reduction on Output and

Employment of a Profit-Maximising Firm

Dollars Capital

MC0 MC1

R
p P

150

100
100 150 Output 25 50 Employment

•A wage cut reduces the marginal cost of production and encourages the
firm to expand output (from producing 100 to 150 units). Scale effect.
•The firm moves from point P to point R, increasing the number of workers
hired from 25 to 50.
 4 - 27

Figure 4.11: The Long Run Demand

Curve for Labour
Dollars

The long-run demand curve for

labour gives the firm’s
w0
employment at a given wage and
is downward sloping.

w1

DLR

25 50 Employment
 4 - 28

Figure 4.12: Substitution and Scale Effects

Capital
A wage cut generates
D
substitution and scale effects.
C1/r
The scale effect (the move from
Q point P to point Q) encourages
C0/r the firm to expand output,
R
increasing the firm’s
P employment. The substitution
200
effect (from Q to R) encourages
D
the firm to use a more labour-
intensive method of production,
100
further increasing employment.
Wage is w1
Wage is w0

25 40 50 Employment
 4 - 29

Long-run elasticity of labour demand

• In the long run, the firm can take full advantage of the
economic opportunities introduced by a change in the wage.
As a result, the long-run labour demand curve is more
elastic (‘flatter) than the short-run labour demand curve
(‘steeper’). See Figure 4.13.

• Wide range of estimates. ‘Consensus’ estimates:

- Short-run elasticity between –0.4 and –0.5.
- Long-run elasticity around –1.0.
 4 - 30

4.5 The Elasticity of Substitution

• When two inputs can be substituted at a constant rate, the two

inputs are called perfect substitutes.
• When an isoquant is right-angled, the inputs are perfect
complements.
• The substitution effect is very large (infinite) when the two
inputs are perfect substitutes. Depending on the slope of the
budget line, only one of the inputs will be used in production!
• There is no substitution effect when the inputs are perfect
complements (since both inputs are required for production).
Input price changes do not alter the input mix.
• The curvature of the isoquant measures elasticity of substitution
 4 - 31

Figure 4.14: Isoquants when Inputs are either

Perfect Substitutes or Perfect Complements

Capital Capital

100

q 0 Isoquant q 0 Isoquant
5

200 Employment 20 Employment

Capital and labour are perfect substitutes if the isoquant is linear (in this
example, two workers can always be substituted for one machine). The two
inputs are perfect complements if the isoquant is right-angled. The firm then
gets the same output when it hires 5 machines and 20 workers as when it hires
5 machines and 25 workers.
 4 - 32

Elasticity Measurement

• The more curved the isoquant, the smaller the size of the substitution effect.
To measure the curvature of the isoquant we use the elasticity of
substitution.

• Intuitively, the elasticity of substitution is the percentage change in capital

to labour (a ratio) given a percentage change in the price ratio (wages to real
interest):

Percentage change in (K/E) divided by percentage change in (w/r)

• It is a positive number. (w↑ → capital intensity ↑) .

 4 - 33

4.6 Application: Affirmative Action and Production

Costs - The Importance of Empirical Evidence

• Whether affirmative action (the mandated hiring of certain

types of workers, e.g. women, ethnic minorities) improves the
profitability of a firm depends on whether or not the firm
discriminated before the policy was introduced.
- A) If the firm was discriminating, affirmative action can improve
its profitability because discrimination will have shifted the
hiring decision away from the cost minimization tangency point
on the isoquant. “Discrimination is not profitable”.
- B) If the firm was not discriminating but is forced to adopt
affirmative action measures, its costs will increase and its profit
will decline. In the extreme case, the firm might go bankrupt.
 4 - 34

Figure 4.15a: Affirmative Action and the

Costs of Production

Black Labour
The discriminatory firm chooses
the input mix at point P, ignoring
the cost-minimizing rule that the
isoquant be tangent to the
isocost. An affirmative action
program can force the firm to
Q move to point Q, resulting in
more efficient production and
lower costs (despite black labour
P

q*
being less productive, i.e. having
White Labour
lower wage, than white workers).
 4 - 35

Figure 4.15b: Affirmative Action and the

Costs of Production
Black Labour

A non-discriminating firm is at
point P, hiring relatively more
whites because of the shape of
the isoquants. An affirmative
action program increases this
Q firm’s costs.
Note that the way the isocost
lines are drawn, black labour is
P
more productive (has higher
q
*

wage) than white labour.

White Labour
 4 - 36

4.7 Marshall’s Rules of Derived Demand

• Marshall’s famous four rules about factors that

generate elastic industry labour demand curves.

• Labour demand in a particular industry is more elastic

if:
- Elasticity of substitution is greater.
- Elasticity of demand for the firm’s output is greater.
- The greater labour’s share in total costs of production.
- But note the qualification mentioned in Note 12, p. 131.
- The greater the supply elasticity of other factors of
production (such as capital).
 4 - 37

4.8 Factor Demand with Many Inputs

• There are many different inputs (skilled and unskilled labour; old and new
machines, natural resources etc.). They can all be incorporated into the
production function.

• Cross-elasticity of factor demand:

Percentage change in xi / percentage change in wj
- i.e. responsiveness of demand for input i with respect to price of input j.
- If the cross-elasticity is positive, the two inputs are said to be substitutes
in production (if it is negative, they are complements).
- Unskilled labour and capital usually found to be substitutes, skilled
labour and capital found to be complements. The capital-skill
complementarity hypothesis. Policy implications?
 4 - 38

Figure 4.16: The Demand Curve for a Factor of Production

is Affected by the Prices of Other Inputs

Price of Price of
input i input i

(a)
(b)

D1 D0
D0 D1

Employment of Employment of
input i input i

The demand curve for input i shifts when the price of another input changes.
(a) If the price of a substitutable input rises, the demand curve for input i shifts up.
(b) If the price of a complement rises, the demand curve for input i shifts down.
 4 - 39

4.9 Labour Market Equilibrium

(more on this topic in chapter5!)

Dollars

Supply Figure 4.17: In a

whigh competitive labour
market, equilibrium is
w* attained at the point
where supply equals
wlow demand. The “going
Demand wage” is w* and E*
workers are employed.
ED E* ES Employment Everybody who wants
to work for the wage
w* can find work.
 4 - 40

4.10 Application: The Employment Effects of

Minimum Wages (the simple case)

• The ‘standard model’ of the effects of a minimum

wage:
- It creates or increases unemployment of the low skilled (some
formerly employed people will lose their jobs, other people will
be attracted into the market, but labour demand declines).
- It benefits some employed low skilled workers but disadvantages
others.
- The unemployment rate is higher the higher the minimum wage
and the more elastic the labour supply and demand curves.
 4 - 41

Figure 4.19: The Impact of the Minimum Wage on

Employment (the simple case or ‘standard model’)

Dollars

A minimum wage set

S
above the equilibrium
w−
wage forces employers to
cut employment (from E*
w* to E bar). The higher
wage also encourages (ES
- E*) additional workers
to enter the market. The
D
minimum wage creates
E− E* ES Employment unemployment (E bar –
ES).
 4 - 42

The Impact of Minimum Wages on the Covered and

Uncovered Sectors

Dollars Dollars
(If workers migrate to
SU covered sector)
SC
SU
w−

SU
(If workers migrate to
w* w* uncovered sector)

DC DU

E− EC Employment EU EU EU Employment

(a) Covered Sector (b) Uncovered Sector

If the minimum wage applies only to jobs in the covered sector, the displaced workers might move to
the uncovered sector, shifting the supply curve to the right and reducing the uncovered sector’s wage. If
it is easy to get a minimum wage job, workers in the uncovered sector might quit their jobs and wait in
the covered sector until a job opens up, shifting the supply curve in the uncovered sector to the left and
raising the uncovered sector’s wage. Labour movements between the sectors will stop when the
expected wage in the min. wage sector equals the sure wage in the uncovered sector.
 4 - 43

The Employment Effects of Minimum

Wages (the ‘non-standard model’)
• Lots of empirical studies on the impact of minimum wages on
employment (often using data on teenagers).
• US findings: minimum wage up, teenage employment down (elasticity
between –0.1 and –0.3).
• Some recent studies seem to contradict this ‘consensus’: Imposing a
minimum wage might actually increase employment! (5 reasons
that might explain this results, pp. 144/5)
- NOTE: In this chapter only the empirical evidence is introduced. A
theoretical model that might be able to explain these findings is
contained in chapter 5, section 5.8, of the Borjas textbook.
• Some facts from two NZ studies were mentioned in class (Hyslop and
Stillman, 2007; Pacheco, 2007). Full references are given in your
Course Reading List. A two page handout on the NZ minimum wage
was also provided.
 4 - 44

4.11 Adjustment Costs and Labour Demand

• So far our labour demand model assumed that a firm can

instantly change the level of employment. This is not the case
in reality.
• The expenditures that firms incur as they adjust the size of their
workforce are called adjustment costs.
- Variable adjustment costs are associated with the number
of workers the firm will hire and fire.
- Fixed adjustment costs do not depend on how many
workers the firm is going to hire and fire.
• The two types of adjustment costs will have different
implications for firm behaviour.
 4 - 45

Figure 4.21: Asymmetric Variable Adjustment Costs

Variable
Adjustment Costs

C0

Change in
Employment
-25 0 +50

If there are variable adjustment costs, changing employment quickly is

costly, and these costs increase at an increasing rate. If government policies
prevent firms from firing workers, the costs of trimming the workforce will
rise even faster than the costs of expanding the firm.
 4 - 46

Slow Transition to a New Labour

Equilibrium
Employment Variable adjustment
costs encourage the firm
to adjust the employment
150
B level slowly. The
expansion from 100 to
100 A 150 workers might occur
more rapidly than the
50
C contraction from 100 to
50 workers if
government policies
Time “tax” firms that cut
employment.
 4 - 47

Adjustment Costs and Labour Demand

• If there are fixed adjustment costs, they have to be paid whether the firm
hires or fires one or many workers.
• The firm faces a cost-benefit situation: Is it profitable to increase the number
of workers now (extra revenue) and pay the fixed adjustement costs (extra
cost)? Sometimes it will be, sometimes it will not. It depends on whether
variable or fixed adjustment costs dominate:
- If fixed adjustment costs are large, employment changes in a firm will
either be sudden and large, or they will not occur at all.
- If variable adjustment costs are large (compared to fixed adjustment
costs) employment changes are likely to occur slowly.
• Both types of adjustment costs are important in reality. We can say that
firms usually take a while to adjust to a new equilibrium when economic
circumstances change (6 months to adjust half-way?).
 4 - 48

Adjustment Costs and Labour Demand

• There are some other interesting topics in section 4.11 that students should
read and try to understand but that are not covered in class:
- The likely impact of employment protecton legislation.
- The distinction between workers and hours (including the Theory at
Work piece on ‘work-sharing in Germany’). Increasing fixed costs of
employment (p. 152)
- Job creation and job destruction: Lots of it is going on at the same time!
This is also the case in NZ:
• In the year to December 2005, there was an average quarterly
worker turnover rate of 17.6 percent. By firm size, it was highest in
firms with 1-5 employees (19.2%), while firms with 100+
employees had the lowest (15.8%). See Statistics NZ “Linked
Employer-Employee Data – December 2005 quarter”
(published February 2007, availble on-line).
 4 - 49

4.12 Shifts in Labour Supply and Labour Demand Curves

Generate the Observed Data on Wages and Employment

S0

Dollars

S1
Z
P
w0
R
w2
Z
Q
w1
D1

D0

E0 E1 E2 Employment
 4 - 50

How to estimate labour demand and labour

supply elasticities?

• This is very tricky. The textbook gives an example, which

should be of interest to those students that also do, or intend to
do, a course in econometrics.
• Essentially, because observed wage and employment data are
almost always due to both shifts in the demand and supply
curve, empirical economist somehow have to hold constant one
curve to estimate the other. This is done in econometrics using
the method of instrumental variables. Good instrumental
variables are often difficult to find.
 4 - 51

End of Chapter 4