Essentials of Microeconomics

Essentials of Microeconomics
‘Nguyen and Wait have written a clear and accessible introduction to modern
microeconomics. The exposition of marginal effects and of game theory should be
especially helpful to students concerned about their maths background but keen to master
economics.’
Richard Pomfret, Professor of Economics, University of Adelaide, Australia
‘Nguyen and Wait’s new textbook rejects anecdotes in favour of developing the key
practical areas of microeconomic theory with just enough technical detail to prepare
students to bring the theory to data in their careers. A textbook for a new generation of
business focused economics students planning on actually using economics.’
Kieron Meagher, Professor of Economics, Australian National University, Australia
‘Essentials of Microeconomics, by Bonnie Nguyen and Andrew Wait, prepares students

with introductory chapters, and not appendices, for quantitative analysis appropriate for a
rigorous yet accessible course in microeconomics principles. The students may better grasp
and retain core microeconomic concepts by using this textbook compared with other
textbooks that treat the tools as less integral to learning the microeconomic concepts.’
William Rieber, Professor of Economics, College of Business, Butler University, USA
Essentials of Microeconomics is an excellent introduction to microeconomics. It presents

the basic tools of microeconomics in a clear and concise manner. This book delivers a
parsimonious yet vigorous treatment of all relevant introductory microeconomic concepts.
In the text there is also an emphasis on modern economics – on game theory and imperfect
markets. The book is self-contained, with a chapter on the key mathematical skills required
included at the start of the text.
This book is ideal for an introductory microeconomics course at any good university. The
level of analysis of the book also makes it appropriate even for introductory level economics
taught at the postgraduate level; the emphasis on strategy makes this text well suited for an
introductory course in business economics.
Bonnie Nguyen and Andrew Wait

Bonnie Nguyen is an Honours graduate from the University of Sydney in both Law and
Economics. Bonnie’s research interests include an economic analysis of litigation and the
internal organization and ownership structure of firms. She has published research in the
Journal of Institutional and Theoretical Economics and the Australian Economic Review.
Bonnie has previously taught microeconomics at the University of Sydney.
Andrew Wait is an Associate Professor in the School of Economics at the University of Sydney.
He has a PhD from the Australian National University and a Bachelor of Economics (Honours)
from the University of Adelaide. Andrew’s research interests include industrial organization
and organizational economics. Andrew has published in the Rand Journal of Economics,
Essentials of
Microeconomics
the International Journal of Industrial Organization and the Journal of Law, Economics and
Organization. He is the co-convenor of the Annual Organizational Economics Workshop.
Economics Cover image: © Thinkstock
ISBN 978-1-138-89136-4
www.routledge.com/cw/wait
www.routledge.com Bonnie Nguyen and Andrew Wait
Routledge titles are available as eBook editions in a range of digital formats
9 781138 891364
6458 ESSENTIALS MICROECONOMICS-V_246x174mm 10/06/2015 15:46 Page i
Essentials of
Microeconomics
Essentials of Microeconomics is an excellent introduction to microeconomics. It

presents the basic tools of microeconomics in a clear and concise manner. This book
delivers a parsimonious yet vigorous treatment of all relevant introductory micro-
economic concepts. In the text there is also an emphasis on modern economics – on
game theory and imperfect markets. The book is self-contained, with a chapter on the
key mathematical skills required included at the start of the text.
This book is ideal for an introductory microeconomics course at any good university.
The level of analysis of the book also makes it appropriate even for introductory level
economics taught at the postgraduate level; the emphasis on strategy makes this text
well suited for an introductory course in business economics.
Bonnie Nguyen is an Honours graduate from the University of Sydney in both Law
and Economics. Bonnie’s research interests include an economic analysis of litigation
and the internal organization and ownership structure of firms. She has published
research in the Journal of Institutional and Theoretical Economics and the Australian
Economic Review. Bonnie has previously taught microeconomics at the University of
Sydney.
Andrew Wait is an Associate Professor in the School of Economics at the University

of Sydney. He has a PhD from the Australian National University and a Bachelor of
Economics (Honours) from the University of Adelaide. Andrew’s research interests
include industrial organization and organizational economics. Andrew has published
in the Rand Journal of Economics, the International Journal of Industrial Organization
and the Journal of Law, Economics and Organization. He is the co-convenor of the
Annual Organizational Economics Workshop.
6458 ESSENTIALS MICROECONOMICS-V_246x174mm 10/06/2015 15:46 Page ii
‘Nguyen and Wait have written a clear and accessible introduction to modern
microeconomics. The exposition of marginal effects and of game theory should be
especially helpful to students concerned about their math background but keen to master
economics. A welcome feature is the chapter on international trade, an application of
microeconomics too often omitted from inward-oriented texts but essential in our
globalized world.’
Richard Pomfret, Professor of Economics,
University of Adelaide, Australia
‘The past two decades have seen principles classes in economics shift to emphasizing
narrow anecdotes while in industry the combination of a core set of technical
microeconomics tools and big data have been revolutionizing areas like auction design,
strategy, bargaining and marketing. Nguyen and Wait’s new textbook rejects anecdotes
in favour of developing the key practical areas of microeconomic theory with just
enough technical detail to prepare students to bring the theory to data in their careers.
A textbook for a new generation of business focused economics students planning on
actually using economics.’
Kieron Meagher, Professor of Economics,
Australian National University, Australia
‘Essentials of Microeconomics, by Bonnie Nguyen and Andrew Wait, prepares students

with introductory chapters, and not appendices, for quantitative analysis appropriate
for a rigorous yet accessible course in microeconomics principles. After discussing
basic economic tools in Chapter 1, e.g. marginal analysis and ceteris paribus, the
authors explain mathematical tools in Chapter 2, e.g. simultaneous equations and
differentiation, and strategic tools in Chapter 3, i.e. game theory models. They present
the mathematical and strategic tools in a straightforward and engaging manner, which
should motivate students to review or learn the material. As a result, the students may
better grasp and retain core microeconomic concepts by using this textbook compared
with other textbooks that treat the tools as less integral to learning the microeconomic
concepts.’
William Rieber, Professor of Economics,
College of Business, Butler University, USA
‘Nguyen and Wait’s Essentials of Microeconomics is really covering the essentials of

the discipline. All concepts are explained clearly, in plain language and with no frills.
These features, however, do not compromise rigour of the content. This book can be
a great resource for the complete beginners and a useful refresher for the more
experienced students.’
Carlo Reggiani, Lecturer in Microeconomics,
University of Manchester, UK
‘The authors have done a great job explaining complex materials.’

Shahina Amin, Associate Professor Department of Economics,
University of Northern Iowa, USA
6458 ESSENTIALS MICROECONOMICS-V_246x174mm 10/06/2015 15:46 Page iii
Essentials of
Microeconomics
Bonnie Nguyen and

Andrew Wait
6458 ESSENTIALS MICROECONOMICS-V_246x174mm 10/06/2015 15:46 Page iv
First published 2016

by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
711 Third Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2016 Bonnie Nguyen and Andrew Wait
The right of Bonnie Nguyen and Andrew Wait to be identified as authors
of this work has been asserted by them in accordance with sections 77
and 78 of the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced
or utilized in any form or by any electronic, mechanical, or other means,
now known or hereafter invented, including photocopying and recording,
or in any information storage or retrieval system, without permission in
writing from the publishers.
Trademark notice: Product or corporate names may be trademarks or
registered trademarks, and are used only for identification and explanation
without intent to infringe.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
Nguyen, Bonnie.
Essentials of microeconomics/Bonnie Nguyen and Andrew Wait.
pages cm
1. Microeconomics. I. Wait, Andrew. II. Title.
HB172.N446 2015
338.5 – dc23
2015000456
ISBN: 978-1-138-89135-7 (hbk)

ISBN: 978-1-138-89136-4 (pbk)
ISBN: 978-1-315-69033-9 (ebk)
Typeset in Times and Helvetica

by Florence Production Ltd, Stoodleigh, Devon, UK
6458 ESSENTIALS MICROECONOMICS-V_246x174mm 10/06/2015 15:46 Page v
Contents
List of illustrations vii
Part I Key concepts and tools 1
1 Key economic concepts 3
2 Key mathematical tools 7
3 Key strategic tools 13
Part II Gains from trade 25
4 Trade and the PPF 27
5 Bargaining 37
Part III Market fundamentals 45
6 Demand 47
7 Production and costs 53
8 Supply 65
9 Equilibrium and welfare 69
10 Elasticity 79
6458 ESSENTIALS MICROECONOMICS-PT2_246x174mm 18/06/2015 09:45 Page vi
vi Contents
Part IV Types of markets 89
11 Introduction to markets 91
12 Perfect competition 93
13 Monopoly 105
14 Monopolistic competition 123
15 Oligopoly 129
Part V Market failures 145
16 Price regulation, taxes and subsidies 147
17 Externalities 163
18 Public goods and common resources 179
19 The Theory of Second Best 185
Part VI International trade 187
20 International trade 189
Part VII Review 201
21 Questions and answers 203
Index 237
6458 ESSENTIALS MICROECONOMICS-V_246x174mm 10/06/2015 15:46 Page vii
Illustrations
Figures
3.1 The normal form of a game 15

3.2 A game with a dominant strategy equilibrium 15
3.3 A game with two Nash equilibria 16
3.4 The prisoner’s dilemma 19
3.5 The coordination game 19
3.6 Battle of the sexes 20
3.7 Stag hunt 20
3.8 Extensive form of a game 21
3.9 The normal form of a game 22
3.10 Extensive form of a game 23
4.1 The production possibility frontier (PPF) traces out combinations
of the quantity of two goods (X and Y) that can be produced if all
resources are used 29
4.2 A production possibility frontier for goods X and Y 30
4.5 Michelle and Rodney’s PPFs 34
5.1 Negotiation in which A makes a single offer of p 38
5.2 Negotiation in which B makes a single offer of p 39
5.3 Negotiation in which A makes an offer of p1 and B makes a
counter-offer of p2 40
6.1 A typical marginal benefit curve 48
6.2 An individual consumer’s demand curve is given by his or her
marginal benefit curve 49
6.3 A movement of the demand curve itself is called a
‘change in demand’ 50
6.4 The market demand curve (DM) can be derived by summing
horizontally the individual demand curves (D1 and D2) 51
7.1 A typical short-run production function, where L represents
the amount of labour employed and q(L) represents the level
of output 54
7.2 A typical total cost function, where q represents the level of
output and TC represents total cost 57
6458 ESSENTIALS MICROECONOMICS-V_246x174mm 10/06/2015 15:46 Page viii
viii Illustrations
7.3 The typical shape of the average total cost curve, the average
fixed cost curve, the average variable cost curve and the
marginal cost curve 60
7.4 The long-run average cost curve can be obtained by taking the
lower envelope of all short-run average cost curves 61
7.5 When the LRAC curve is downward-sloping, the firm is
experiencing economies of scale; when the LRAC curve is
upward-sloping, there are diseconomies of scale. When the
LRAC curve is flat, there are constant average costs 61
8.1 An individual firm’s supply curve is given by its marginal
cost curve 66
8.2 A movement of a firm’s supply curve itself is called a
‘change in supply’ 67
8.3 The market supply curve (SM) can be derived by summing
horizontally the individual supply curves (S1 and S2) 68
9.1 A market in equilibrium 70
9.2 When market price is above the equilibrium price, there is an
excess of supply in the market 70
9.3 When market price is below the equilibrium price, there is an
excess of demand in the market 71
9.4 The market for corn chips when there is an increase in the
price of potato chips 72
9.5 The market for cars when there is an increase in the price
of steel 72
9.6 The area of consumer surplus in this market is denoted by the
shaded area 74
9.7 When the market price falls from p1 to p2, the area of consumer
surplus increases from A to A + B + C 74
9.8 The area of producer surplus in this market is denoted by the
shaded area 75
9.9 When the market price increases from P1 to P2, the area of
producer surplus increases from A to A + B + C 75
9.10 The area of total surplus in this market is denoted by the
shaded area 76
10.1 When demand is perfectly inelastic, the demand curve is vertical
(as seen the left panel). When demand is perfectly elastic, the
demand curve is horizontal (as seen the right panel) 82
10.2 When the demand curve is linear, elasticity changes as we move
along the curve 83
10.3 Total revenue can be calculated by multiplying price and
quantity 84
10.4 When supply is perfectly inelastic, the supply curve is vertical
(as seen the left panel). When supply is perfectly elastic, the
supply curve is horizontal (as seen the right panel) 86
6458 ESSENTIALS MICROECONOMICS-V_246x174mm 10/06/2015 15:46 Page ix
Illustrations ix
12.1 The short-run supply curve of a firm is traced out by the part of
the MC curve that lies above AVC 95
12.2 A firm in a perfectly competitive market, making a profit 96
12.3 A firm in a perfectly competitive market, making a loss 97
12.4 The long-run supply curve of a firm is traced out by the part of
the MC curve that lies above ATC 98
12.5 When the market price is above average total cost, firms in the
market are making profits 99
12.6 When the market price is below average total cost, firms in the
market are making losses 99
12.7 In the long run, there are zero profits in a perfectly competitive
market 100
12.8 In the long run, free entry and exit means that the price in a
constant-cost industry will always be driven back to ATCmin 101
12.9 Following an unanticipated increase in demand, in the short run
price rises and firms in the industry make positive economic
profits. However, in the long run, entry forces prices back down
to the p* = ATCmin . Each firm sells q* units and economic profits
are zero 102
12.10 In an increasing-cost industry, the long-run industry supply curve
is upwards sloping 103
12.11 In a decreasing-cost industry, the long-run industry supply curve
is downward sloping 103
13.1 When the demand curve is linear, the marginal revenue curve
has the same vertical intercept and twice the slope of the
demand curve 107
13.2 A profit maximizing monopolist sets MR = MC, and thus produces
quantity Qm . At this quantity, the market price will be Pm 109
13.3 The monopolist’s profit is given by ␲ = Q(P – ATC) 109
13.4 The quantity traded in the market is lower under monopolist
than in a competitive market 110
13.5 Consumer surplus and producer surplus under a monopoly 111
13.6 Total surplus and dead weight loss under a monopoly 112
13.7 When the monopolist engages in first-degree price discrimination,
the efficient quantity is traded in the market 114
13.8 When the monopolist uses a two-part tariff, it charges a fixed fee
(F) equal to the size of the lighter shaded triangle 114
13.9 When a monopolist engages in third-degree price discrimination
(with constant MC s, it maximizes profit where MRA = MRB = MC 115
13.10 When a firm has a fixed cost and a constant marginal cost, the
average total cost curve will be downward sloping for all values
of Q; this industry will be a natural monopoly 119
13.11 Under marginal-cost price regulation, the government sets the
monopoly price at P = MC (assuming constant MC for simplicity) 120
6458 ESSENTIALS MICROECONOMICS-V_246x174mm 10/06/2015 15:46 Page x
x Illustrations
13.12 Under average-cost price regulation, the government sets the

monopoly price at P = ATC 121
14.1 A firm in a monopolistically competitive market in the short run 125
14.2 A firm in a monopolistically competitive market in the long run 127
15.1 Price-setting game in an oligopoly 131
15.2 Advertising game in an oligopoly 131
15.3 Location-choice game in an oligopoly 134
15.4 Platform choice for game developers 135
15.5 Entry game in an oligopoly 137
15.6 Viking invasion game 138
15.7 Simultaneous technological choice 140
15.8 Technological choice game with a first-mover advantage 140
15.9 Natural monopoly entry game 141
15.10 Free-riding game 142
15.11 A ‘matching pennies’ game 143
–
16.1 A price floor set at P 148
–
16.2 The welfare effects of a price floor set at P 149
–
16.3 A price ceiling set at P 150
–
16.4 The welfare effects of a price floor set at P 150
16.5 A tax on consumers causes the demand curve to shift downwards
by the size of the tax 152
16.6 A tax on consumers creates a new market price and quantity at
(Qt , Pt ) 153
16.7 A tax on producers causes the supply curve to shift upwards by
the size of the tax 153
16.8 A tax on producers creates a new market price and quantity at
(Qt , Pt ) 154
16.9 The welfare effects of a tax 154
16.10 A tax on consumers in the market for tomatoes 156
16.11 If demand is perfectly inelastic, when a tax of t per unit is
instituted, consumers pay for all of the tax; following the
introduction of the tax consumers pay P* + t whereas suppliers
continue to receive P* 157
16.12 When a tax of t per unit is implemented and supply is perfectly
elastic, consumers pay for all of the tax; following the introduction
of the tax, consumers pay P* + t whereas suppliers continue to
receive P* 158
16.13 The deadweight loss generated by a tax depends on the
responsiveness of supply and demand 158
16.14 A subsidy for consumption creates a new market price and
quantity at (Qs , Ps) 160
16.15 A subsidy for production creates a new market price and
quantity at (Qs , Ps ) 161
16.16 Consumer surplus and producer surplus as a result of a subsidy 161
6458 ESSENTIALS MICROECONOMICS-V_246x174mm 10/06/2015 15:46 Page xi
Illustrations xi
16.17 Government revenue and deadweight loss as a result of a subsidy 162

17.1 The relationship between MPB and MSB in the presence of a
positive externality 165
17.2 The relationship between MPC and MSC in the presence of a
negative externality 166
17.3 The market equilibrium and the socially optimal outcome in the
presence of a positive externality 167
presence of a negative externality 167
presence of a negative consumption externality 168
presence of a positive production externality 169
17.7 A subsidy granted to the consumer and to the producer, used to
address a positive externality 173
17.8 A tax imposed on the consumer and on the producer, used to
address a negative externality 173
18.1 The market demand curve (MBT ) for a public good is obtained
by the vertical summation of each individual’s marginal
benefit curve 181
20.1 The market equilibrium for a country in autarky is given by
the intersection of the domestic demand curve and the domestic
supply curve 190
20.2 A country that is open to international trade is an exporter if the
domestic equilibrium price (Pd) is lower than the world price
(Pw) 191
20.3 Consumer surplus and producer surplus for an exporting country 192
20.4 A country that is open to international trade is an importer if
the domestic equilibrium price (Pd ) is higher than the world
price (Pw) 192
20.5 Consumer surplus and producer surplus for an importing country 193
20.6 The effect of a tariff on domestic market outcomes for an importing
country 194
20.7 The welfare effects of a tariff 195
20.8 The effect of a quota on domestic market outcomes for an
importing country 196
20.9 The left panel depicts the welfare effects of a quota 197
Table
11.1 Types of market structure 92

6458 ESSENTIALS MICROECONOMICS-V_246x174mm 10/06/2015 15:46 Page xii
Page Intentionally Left Blank

6458 ESSENTIALS MICROECONOMICS-V_246x174mm 10/06/2015 15:46 Page 1
PA R T I
Key concepts
and tools

6458 ESSENTIALS MICROECONOMICS-PT2_246x174mm 18/06/2015 09:45 Page 3
C H A P T E R
1
Key economic
concepts
1.1 Introduction
Economics is the study of choice under scarcity. Typically consumers want

more goods and services than they can afford to buy. Similarly, businesses
face constraints in relation to the funds and resources they have access to.
Governments and countries also face the same type of problem: a government
might want to tackle a large number of social problems but only have limited
resources to draw on. Economics is about understanding how a party (that is,
a consumer, a business, a country and so on) deals with the fact that when
they use their resources to pursue one option, they cannot use those resources
to do something else. And so, a consumer may have to choose between a new
pair of shoes or a textbook, a firm may have to choose between developing a
new product or launching a marketing campaign, and a government may have
to choose between improving education or targeting crime.
To help understand these issues, economics has developed a set of analytical
tools. This book provides an introduction to these tools. They can be used to
help understand economic problems wherever they arise, be it businesses
understanding the markets they compete in, or governments trying to develop
social policy, or families trying to manage their households. These tools
are not meant to capture everything that is occurring in any given situation.
Rather, they are designed to simplify (or to model) complicated and poten-
tially messy real-world issues into a tractable form that can provide valuable
insights.
Given that resources are limited, the key questions that an economy needs
to ‘decide’ are: (a) what to produce; (b) how to produce it; and (c) who should
get what is made. In modern economies, the answers to these questions are
largely determined by the market – that is, through the interaction of sellers
and buyers.1 Sometimes, however, the government also helps determine the
4 1 Key concepts and tools
answer to these questions by regulating or intervening in the market. Consequently,

our focus in this microeconomics text will be on the behaviour of individuals
(consumers, firms, and governments) and their interaction in markets.
This chapter provides a few key concepts that underpin the analysis in the rest of
the book, as well as economic analysis in general.
1.2 Scarcity and opportunity cost
As noted above, it is usually the case that resources are limited, so that not all wants
can be met. We call this situation scarcity.
Scarcity also means that individuals, businesses and societies face trade-offs; by
choosing one thing, a person must give up or miss out on something else. For example,
if a consumer uses their money to buy product X, they cannot then use that same money
to buy another product.2 We use the concept of opportunity cost to capture this trade-
off; the opportunity cost of any choice is the value of the next best forgone alternative.
In the example above, if the consumer buys product X, and the next best thing they
could have done is buy product Y, the opportunity cost of buying X is forgoing Y.
Individuals also face opportunity costs in terms of their time – that is, if a person
spends his time doing one thing, he cannot also spend that time doing something else.
Example. Suppose Elizabeth prefers to spend her Saturday afternoon walking. The
next best thing that she could have done is to sleep, and her third best choice is
to go swimming. Therefore, if Elizabeth goes for a walk, the opportunity cost of
going for a walk is not sleeping, as this is her best foregone opportunity. The option
of swimming is not relevant here because it is not the next best opportunity.
Opportunity costs include both explicit costs and implicit costs. Explicit costs are
costs that involve direct payment (or, in other words, would be considered as costs by
an accountant). Implicit costs are opportunities that are forgone that do not involve an
explicit cost.3
Example. Suppose Stephen decides to go to university, and his next best option
is to work at a construction site and earn $80k over the year. The explicit costs
are those that Stephen must directly pay to go to university, such as student fees,
the cost of textbooks, and so on. The implicit costs are the opportunities that
Stephen must forgo – in this case that is working at the construction site and earning
$80K.
It is important to note that opportunity cost only includes costs that could change
if a different decision were made. Opportunity cost does not include sunk (or unrecover-
able) costs. Sunk costs are costs that have been incurred and cannot be recovered no
matter what. For example, if Katrien spends the weekend reading an accounting
textbook, no matter what she does (such as whether or not she decides to continue
Key economic concepts 5
studying accounting), she cannot get that time back. Similarly, if a business spent
$100K on an advertising campaign last year, regardless of what they decide to do this
year, that money (and effort) cannot be recovered.
1.3 Marginal analysis
Typically, we assume that economic agents are rational and act to maximize the benefit
they receive from any economic transactions.4 For example, consumers seek to
maximize their benefits from consumption and firms seek to maximize their profits
from production. One way that economic agents can solve this maximization problem
is by considering the additional benefit or additional cost of any action. This sort of
analysis is referred to as marginal analysis and it is a recurring theme both in this
book and economics generally.
For instance, consider a consumer faced with the decision of whether to buy one more
unit of a particular good. That consumer might consider the extra benefit he derives
from buying that extra unit; this is referred to as the marginal benefit of that extra
unit of the good. The consumer will also consider the additional cost of buying one
more unit; this is referred to as the marginal cost of purchasing another unit, which
is typically the price of the good. In making their final decision, the consumer will weigh
the marginal benefit against the marginal cost of buying that extra unit. For example,
if a consumer is considering buying another cup of coffee, and the marginal benefit is
$5 and the marginal cost is $3, the consumer will be better off buying the extra coffee.
Each of the marginal terms noted above, and many others, will be discussed at length
throughout the book. What is crucial to note is that the term ‘marginal’ simply means
additional or extra. That is, we are interested in what happens if we increase something
(such as the number of coffees bought) by a small amount.
1.4 Ceteris paribus
The notion of ceteris paribus is also an important foundation of economic analysis.

As noted, because the real world is complicated and messy, it is necessary to simplify
real-world situations into tractable economic models, in order to better analyse them.
Thus, in order to determine the impact of a particular event, economists tend to
examine the impact of one change at a time, holding everything else constant. This is
called ceteris paribus, which roughly means ‘other things equal’.
For instance, suppose we are interested in how a change in price will affect the
quantity demanded of a good. In reality, demand for a good can be affected by a number
of other factors, such as changes in the tastes or income of consumers, or the availability
or price of substitute goods. Therefore, in order to isolate the effect of price upon
quantity demanded, we need to hold everything else constant. This is not to deny that
in the real world multiple changes can occur at a time – they often do. Rather, to fully
understand the relationship between price and demand, it is essential to isolate that
relationship from other events that might also be occurring. For example, a firm might
be interested in the effect of advertising on demand for its product. To understand the
impact of advertising, it is crucial to remove other factors that could affect demand,
otherwise advertising could be attributed too much (or too little) influence, which could
lead to poor decision-making by the firm regarding its next advertising campaign.
1.5 Correlation and causation
Another factor to keep in mind is the difference between correlation and causation.
Correlation refers to a situation in which two or more factors are observed to move
together (or in opposite directions). On the other hand, causation refers to a situation
where a change in one factor brings about, or causes, a change in something else. To
make statements about causation requires an economic theory about how the world
works – without a theoretical framework telling us why one factor causes another to
change, we are just observing a statistical relationship between several variables.
Sometimes when we observe correlation between two variables, A and B, it is
because the movement in one variable causes the other to change. Sometimes, it is
because a third factor causes changes in both A and B (like a tide coming in can cause
two boats to rise in their moorings). Sometimes, there is no connection between the
two variables and it is just by chance that we observed the change in both variables
at the same time. Without a theory about how a change in one variable affects the other,
it is not possible to say which option it is in any particular case.
1.6 Concluding comments
This chapter provided a very brief introduction to some key economic concepts. In the
next chapter we outline some of the mathematical tools that we will use throughout
the book.
Notes
1 By ‘market’, we simply mean a place where buyers and sellers of a particular good or service
meet, such as a traditional bazaar or an online trading site.
2 It is common to hear people refer to the ‘economics’ of a particular situation. This
colloquial statement really means that, given the limited resource available, a choice had
to be made and something (possibly worthwhile) could not be done.
3 Sometimes, economists distinguish between ‘economic costs’ and ‘accounting costs’.
Economic costs is just another term for opportunity costs, and therefore includes explicit
and implicit costs. Accounting costs refers to explicit costs only.
4 We are not suggesting that, in the real word, consumers are always fully rational or that
firms do not sometimes have other objectives. Rather, we adopt this simplifying assumption
because it allows us to analyse the behaviour of economic agents in markets. This type of
analysis will be fairly accurate, provided that on average individual consumers and firms
act more or less in their own interest.
C H A P T E R
2
Key mathematical
tools
2.1 Introduction
Even at an introductory level, the analytical nature of economics requires us

to draw upon some mathematical techniques. This chapter outlines some of
those basic tools. Specifically, we cover: the formula for a straight line;
understanding the slope of a line; simple differentiation; and solving
simultaneous equations. In the next chapter we outline some tools that are
useful for thinking about economic environments in which the interaction
between the different participants is important. All of these techniques will
be useful in dealing with the material covered in this text.
2.2 Equations
Often we are interested in the relationship between two variables, x and y.

Sometimes, it is convenient to express that relationship using an equation. In
this section, we will cover straight lines and briefly discuss equations of curves.
2.2.1 Straight lines

The equation of a straight line can take the form:
y ⫽ mx ⫹ c. (2.1)
In this equation, x and y are the variables whose relationship we are interested
in; m and c are parameters.1
When graphing this equation, we might want to know where the line intersects the
axes. Along the y axis, x takes the value of zero, so we can solve for the y-intercept
by setting x = 0 (in Equation 2.1, the y-intercept is y = c). Similarly, y takes the value
of zero along the x axis, so setting y = 0 yields the x-intercept (in Equation 2.1, the
x-intercept is x = –c/m).
Example. Suppose we are interested in the relationship between the price of a

good (P) and the quantity demanded of that good (Q). We might represent this
relationship using the equation P = 100 – 2Q. Here, Q and P take the places of x
and y respectively. The slope of the line is –2 (see the next section). The line cuts
the P-axis at 100 and the Q-axis at 50.
2.2.2 The slope of a straight line

Note that, in Equation 2.1, if we increase x by 1 unit, y must increase by m units. Thus,
the parameter m determines how a change in x affects a change in y (recall that this
relationship is the focus of marginal analysis – see Chapter 1). The parameter m is
also known as the slope or the gradient of the line; it tells us how steep the line is.
That is, if we graph Equation 2.1 with x and y on the horizontal and vertical axes
respectively, for every 1 unit we move horizontally (along the x-axis) we must also
move m units vertically (along the y-axis). Note that if a change in x results in an
increase in y, the slope is positive. Conversely, if the change in x results in a fall in y,
the slope of the line (by definition) is negative. It is the fact that the slope is constant
that makes this the equation of a straight line.
2.2.3 Determining the equation of a straight line

When we are not explicitly given the equation of a straight line, we can determine that
equation from any two points on the line. First, we need to determine the slope, m.
Recall that the slope is simply the change in y for a one-unit change in x. Therefore,
if we have two points on a straight line, (x1, y1) and (x2, y2), the formula for the slope
m is:2
⌬y y2 − y1
m= = (2.2)
⌬x x2 − x1
In other words, the slope of the line m is the change in y (⌬y) given a change in x (⌬x).
Since we have calculated the slope m, it can be substituted into the general equation
of a straight line, y = mx + c, so as to find c. To do this, we use the fact that the equation
must pass through the point (x1, y1), so we can substitute these values into our equation
to determine c.
Example. Suppose a line passes through (4, 3) and (5, 1). Applying Equation 2.2,
we can find the slope, m = (3 – 1)/(4 – 5) = –2. Recall that the general equation
Key mathematical tools 9
of a straight line is y = mx + c. Substituting in m = –2 and (4, 3) yields 3 = –2 ×

4 + c, so c = 11. Therefore, the equation of the line is y = –2x + 11.
2.2.4 Curves
Some economic relationships are represented by equations that are not a straight line.
For example, we might think that total cost (TC) is related to output (q) in such a way
that TC = q2. Plotting this curve (for q ⭓ 0) does not yield a straight line; rather, the
slope of the curve becomes steeper and steeper. The key thing at this stage is to
remember that the marginal relationship between two variables need not be constant.
When this is the case, we can use differentiation to determine that marginal relationship,
as will be discussed in the following section.
2.3 Differentiation
Sometimes, when we know the equation that connects two variables x and y, we want
to find what happens to y when we change x by a very small amount. In these cases,
we can use differential calculus. Differentiation is similar to finding a slope in that it
shows how y responds to changes in x (in fact, for a straight line, calculating the slope
and differentiating both yield the same answer). The difference is that calculus is
concerned with very small changes in x, whereas slope is usually concerned with whole-
unit changes. Differentiation is a useful tool because economics is often concerned
with marginal changes; in particular, we will make use of it when discussing elasticities
(Chapter 10) and determining marginal revenue for a monopolist (Chapter 13).
2.3.1 A simple rule for differentiation

For our purposes, we need one rule of differentiation. Take a function y(x) = Axn. Here,
y is written as a function of x. A and n are parameters. By differentiating a function
once, we get the first derivative, which we can write as dy/dx or y′(x). The rule of
differentiation that we need is:
dy
y(x ) = Ax n ⇒ = nAx n−1 (2.3)
dx
This rule can also be applied to each individual additive component of a function, as
shown in the example below.
Example. Consider a function P (Q) = 10 – Q2 + 3Q –5. We can apply the rule in

Equation 2.3 to each individual component (10, –Q2 and 3Q–5) separately. Note
that 10 is really 10Q 0, so after differentiation this becomes 0. Applying the rule
to the next two components of the function yields –2Q and –15Q–6 respectively.
Thus, the answer is dP/dQ = –2Q – 15Q –6.
Like slope, the first derivative of a function tells us how steep a curve is at a particular
point. If the first derivative is positive (resp. negative) the curve is rising (resp. falling),
and if the first derivative is zero, then the curve is flat.
2.3.2 Finding minima and maxima

Differentiation is also a useful way of finding the maximum or minimum of a function.
Notice that when a function reaches its maximum or minimum, the slope of the function
is zero (that is, the curve is ‘flat’ at that point). Therefore, by setting the first derivative
equal to zero, we can find the maximum or the minimum of a function. Of course, we
will need to know whether the point we have found is a maximum or a minimum.
There are ways to do this (for instance, by checking the second derivative), but
graphing the function is usually a good check.
Example. Suppose we want to find the maximum of the function P = 100Q – Q2.
Setting the first derivative equal to zero, yields dP/dQ = 100 – 2Q = 0. Solving
this equation yields Q = 50. We can graph the function to confirm that this is a
maximum.
2.4 Elasticity
Another thing we might like to know about the relationship between x and y is how
responsive y is to changes in x. That is, when we increase x by a certain amount, does
y change by a small amount or by a large amount? We can measure the responsiveness
of y to changes in x using the concept of elasticity.
We can calculate elasticity (ε) by dividing the percentage change in y by the
percentage change in x:3
% ⌬y
ε= (2.4)
% ⌬x
This tells us that what the percentage change in y will be, given a 1 per cent change
in x. As you can see, the larger the absolute value of ε is, the more responsive is y to
changes in x; conversely, the smaller the absolute value of ε, the less responsive is y
to changes in x.
We will discuss the methods of calculating elasticity and the economic applications
of elasticity further in Chapter 10.
2.5 Simultaneous equations
In some cases, we will have multiple equations that link our variables. In the simplest
case, two equations link two variables, x and y. If we want both equations to hold, we
will need to find values of x and y that satisfy both equations. One way of solving this
Key mathematical tools 11
‘system’ of equations is to rearrange each equation to have y on its left-hand side, and
then equate the right-hand sides of the equations.
Example. Suppose we are given the following two equations that link P and Q:
P = 1 – 2Q (2.5)
P = 3Q (2.6)
(These two equations could represent demand and supply, for instance.) First, note
that P is on the left-hand side of both equations. Therefore, we can equate the right-
hand side of the equations, giving:
100 – 2Q = 3Q
Solving this gives us Q = 20, which can then be substituted into equation 2.5 or
2.6 to find P = 60.
Mathematics allows us to make explicit assumptions and solve and analyse complicated
models. This chapter outlined some of the mathematical tools that we will use
repeatedly in this text. The next chapter outlines some of the conceptional techniques
used to analyse strategic interaction. Taken together, these tools form the foundation
of how the models in this text are constructed and solved.
Notes
1 Here, ‘parameter’ simply means that the value of m and c are fixed and do not depend upon
x or y. m and c are sometimes called ‘constants’ because their value does not change. For
example, c could equal 2 and m could 5; importantly, this will be the case regardless as to
the value of the variables x and y.
2 In this equation, ⌬ simply means ‘change’, so ⌬y means ‘the change in y’.
3 Here, %⌬ simply means ‘percentage change’.

C H A P T E R
3
Key strategic
tools
3.1 Introduction
As we have previously discussed, economic agents try to do the best they can
for themselves; in other words, they try to maximize their objectives subject
to the constraints that they face. Sometimes, it is sufficient to consider a
consumer or firm’s maximization problem in isolation. But at other times, the
strategic interaction between parties is important. In these situations, the
individual’s choice of action will typically depend on what other parties
choose to do. In this chapter, we outline a few basic tools that are useful in
analysing these situations. We will later look at applications of these tools in
Chapters 5 and 15.
3.2 The essentials of game theory
Strategic interactions between economic agents can be analysed using game

theory. This requires the relevant strategic interactions to be represented in
the form of a game.
For most people, the word ‘game’ brings to mind card- and board-games
or sports. However, in economics, the term has a specific meaning. In
particular, a game has the following elements:
• Two (or more) players.

• A complete description of what actions each player may take.
• A specification of each player’s payoff associated with the actions taken.
For now, we will assume that each participant in the game has full knowledge
of these things. That is, each player knows who the other players are, what
actions each player may take, and the payoffs that each player receives if certain actions
are taken. Note that this does not imply that each player knows what actions are actually
taken by the other players.
As you can see, the economic notion of a game is different but related to the everyday
understanding of what a game is. In fact, the economic definition of a game would
probably include most card- and board-games and sports, but also embraces a wider
variety of situations. Consider the following examples.
Example. Annie and Kat are working on a group project together, and their final
mark depends upon how much effort they jointly put in. Let us say each person
has the choice of putting in effort or not. If both of them put in effort, they will
get a mark of 100 per cent; if only one of them puts in effort, they will get a mark
of 70 per cent; and if neither of them puts in effort they will get a mark of 0 per
cent. This situation has all the characteristics of a game: there are two players
(Annie and Kat), we know what actions each player may take (effort or no effort),
and we know each player’s payoff associated with the actions taken (their marks).
Example. Fran and Maxwell are playing soccer, with Maxwell as the goalkeeper.
Fran can kick the ball into the left or the right side of the goal and, at the same
time, Maxwell can dive to the left or the right. If Fran kicks the ball to the opposite
side that Maxwell dives, she will score a goal. If Maxwell dives to the same side
that Fran kicks the ball, he will save the goal and prevent Fran from scoring. This
has all the characteristics of an economic game: there are two players (Fran and
Maxwell), we know what actions each player may take (left or right), and we know
each player’s payoff associated with the actions taken (the score).
3.3 Simultaneous-move games
In this section, we will consider games in which the players have to choose their actions
simultaneously or without knowledge of what the other player has chosen. We will
consider how to represent and solve these games.
3.3.1 Representing simultaneous-move games: the normal form

It is often convenient to represent a simultaneous-move game using the normal form
of the game. In the normal form, the elements of a game are represented in the form
of a table (or matrix).
To see how this is done, consider the following example. Two firms, A and B are
selling an identical product in the same market. At the beginning of the day, each firm
must choose to set their price high (pH) or low (pL ) without knowledge of what the
other firm has chosen. If both firms set a price of pH, each firms receives a payoff of
$4. If both firms set a price of pL , each firm will get a payoff of $3. If firm A sets
a price of pH and B prices at pL the payoffs are $1 and $5 to A and B, respectively.
Key strategic tools 15
Player 2
Player 1
(4,4) (1,5)
(5,1) (3,3)
FIGURE 3.1 The normal form of a game
On the other hand, if A chooses pL and B chooses pH the payoffs are $5 and $1 to
A and B, respectively.
This information is represented in the normal form in Figure 3.1. Firm A and its
choices are represented by the rows, and firm B and its choices are represented by the
columns. The payoffs associated with these choice combinations are depicted in four
of the boxes of the matrix; firm A’s payoff is the first number in the box and firm B’s
payoff is the second number in the box.1 Together, A and B determine the set of payoffs
that are realized. Suppose A chooses pL and B chooses pH . A’s choice of pL puts us
in the bottom row; B’s choice of pH puts us in the left column. Hence, it is the payoff
in the bottom left box that is realized (that is, $5 to A and $1 to B).
3.3.2 Solving simultaneous-move games

Let us now turn to the question of what the outcome of a simultaneous-move game
will be. We can solve (or predict the outcome of) a game by assuming that each player
is only interested in maximizing his own payoff. In particular, we assume that each
player does not care about the other player’s payoff; that is, he is not interested in
minimizing or maximizing the other player’s payoff.
Dominant strategy equilibrium

Sometimes, it will be possible to solve a game using dominant strategies. A player has
a dominant strategy when the action that gives him the highest payoff does not
depend on what the other player chooses.
For example, consider the game depicted in Figure 3.2. In this game, Player 1’s
dominant strategy is to choose T as this will yield a higher payoff no matter what Player
2 chooses to do. To see why this is so, consider the game from Player 1’s perspective:
Player 2
L R
Player 1
T (4,8) (3,6)
B (2,7) (1,5)
FIGURE 3.2 A game with a dominant strategy equilibrium

• If Player 2 chooses L, Player 1 can choose T for a payoff of 4 or B for a payoff

of 2. Therefore, Player 1 should choose T in this case.
• If Player 2 chooses R, Player 1 can choose T for a payoff of 3 or B for a payoff
of 1. Therefore, Player 1 should choose T.
As can be seen, no matter what Player 2 chooses, Player 1’s best choice is T. Similarly,
it can be shown that Player 2’s dominant strategy is to choose L:
• If Player 1 chooses T, Player 2 can choose L for a payoff of 8 or R for a payoff

of 6. Therefore, Player 2 should choose L.
• If Player 1 chooses B, Player 2 can choose L for a payoff of 7 or R for a payoff
of 5. Therefore, Player 2 should choose L.
When both players have a dominant strategy, the game has a dominant strategy
equilibrium. That is to say, the outcome of the game will be one in which both players
choose their dominant strategy. In the example above, the dominant strategy
equilibrium is (T, L).
Note, the equilibrium of the game is expressed in terms of the players’ strategies.
It is important to express the equilibrium in terms of the strategies adopted, not the
payoffs. While the payoffs help determine the best strategies for each of the players,
as economists we are interested in what the players actually do, and what they would
do in various situations. Another way of thinking of this is that each player needs to
give instructions to a subordinate about what to do, and what to do in any situation
that could arise – here, the instructions would be to play T or L for each player,
respectively. Instructions that stated ‘4’ or ‘8’ might not be understood by a subordinate,
because ultimately these payoffs also rely on the action taken by the other player.
Nash equilibrium
Sometimes, neither player will have a dominant strategy; for example, consider the
game depicted in Figure 3.3. In such cases, we will need another solution concept in
order to solve the game. Because neither player has a dominant strategy, each player’s
choice of action depends on what he or she thinks the other player will do. We say
that an action is Player 1’s best response if that action gives him the highest possible
profit, given Player 2’s choice.
Player 2
L R
Player 1
T (5,2) (7,4)
B (8,3) (6,1)
FIGURE 3.3 A game with two Nash equilibria

To see how this operates in practice, consider the game depicted in Figure 3.3 from
the perspective of Player 1:
• If Player 2 chooses L, Player 1 can choose T for a payoff of 5 or B for a payoff

of 8. Therefore, if Player 2 chooses L, Player 1’s best response is to choose B.
• If Player 2 chooses R, Player 1 can choose T for a payoff of 7 or B for a payoff
of 6. Therefore, if Player 2 chooses R, Player 1’s best response is to choose T.
Let us also consider the game from the perspective of Player 2:
• If Player 1 chooses T, Player 2 can choose L for a payoff of 2 or R for a payoff

of 4. Therefore, if Player 1 chooses T, Player 2’s best response is to choose R.
• If Player 1 chooses B, Player 2 can choose L for a payoff of 3 or R for a payoff
of 1. Therefore, if Player 1 chooses B, Player 2’s best response is to choose L.
A Nash equilibrium exists if each player’s choice of action is their best response to
every other player’s strategy. In the case of two players, this means that Player 1’s
choice of action is his best response to Player 2’s choice and that Player 2’s choice of
action is also her best response to Player 1’s choice.
In the game depicted in Figure 3.3, there are two Nash equilibria: (T, R) and
(B, L).2 Let us check each of these in turn:
• (T, R): If Player 1 chooses T, Player 2’s best response is to choose R. If Player 2
chooses R, Player 1’s best response is to choose T. This is a Nash equilibrium
because both players are choosing their best response given what the other player
has chosen.
• (B, L): If Player 1 chooses B, Player 2’s best response is to choose L. If Player 2
chooses L, Player 1’s best response is to choose B. This is a Nash equilibrium
because both players are choosing their best response given what the other player
has chosen.
A corollary of this is that, in a Nash equilibrium, no player can unilaterally deviate

(that is, switch his choice of action, holding constant the strategies of all other players)
and increase his payoff. This is because every player is already choosing his best
response and making the maximum payoff possible, given the other player’s actions.
It is the fact that no player has an incentive to deviate that makes the outcome
an equilibrium. Note, often a convenient way to check whether an outcome is a Nash
equilibrium is to ensure that no player would like to make a unilateral deviation in
order to make herself better off – often referred to as a profitable unilateral deviation.
To see this, let us check the possibility of profitable unilateral deviation for each of
the feasible outcomes of the game illustrated in Figure 3.3. First, take the outcome
(T, L) – are there any unilateral deviations that make a player better off? Holding
Player 2’s action fixed on L, Player 1 can continue to choose T and get 5 or can make
a unilateral deviation to B and get 8 – Player 1 has a profitable unilateral deviation,
so (T, L) cannot be a Nash equilibrium. Similarly, consider the outcome (B, R).
Holding Player 1’s strategy constant on B, Player 2 would want to deviate and play
L, improving her payoff from 1 to 3. Consequently, (B, R) cannot be a Nash equilibrium.
On the other hand, if we consider either of the two other possible outcomes – (B, L)
and (T, R) – each player has adopted their best response to the other player’s strategy.
As a consequence, there are no profitable deviations, and both outcomes are a Nash
equilibrium.
Finally, it is worth noting several other points regarding Nash equilibria. First, a
dominant strategy equilibrium is always a Nash equilibrium. In a Nash equilibrium all
players have chosen their best response to the strategies of all other players. This is
true in a dominant strategy equilibrium. Consider the game shown in Figure 3.2. Player
1 chooses T – this is his best response to whatever Player 2 is doing. Similarly, Player
2’s choice of L is also a best response. This means that the dominant strategy
equilibrium of (T, L) is also a Nash equilibrium. Note that while all dominant strategy
equilibrium are Nash equilibrium, the reverse is not necessarily true – a Nash
equilibrium need not be a dominant strategy equilibrium. For example, consider the
Nash equilibria in Figure 3.3. The players in this game do not have a dominant strategy,
so no dominant strategy equilibrium exists; there are, however, two Nash equilibria
as outlined above. Second, depending on the possible actions and payoffs, a game could
have one, two or more Nash equilibrium. If a game has only one Nash equilibrium, it
is often referred to as a unique Nash equilibrium.
3.4 Some types of simultaneous-move games
In this section, we describe some of the ‘classic’ simultaneous-move games.
3.4.1 Prisoner’s dilemma

The prisoner’s dilemma is perhaps the most canonical simultaneous-move game. In
this game, two criminals are being separately questioned about a crime and they have
no way of communicating with each other. If both criminals stay silent, there will only
be enough evidence to convict them of a lesser crime with a sentence of one year each.
The police offer each criminal the following deal: if he gives evidence and his partner
stays silent, he will go free but his partner will be imprisoned for three years. However,
if both criminals confess, both will get two years in prison. This game is represented
in Figure 3.4. In this game, remember, payoff represent years in prison, so a lower
number is better.
This game has two important features. First, there is a dominant strategy
equilibrium; the dominant strategy of both parties is to confess and hence the
equilibrium is (Confess, Confess). Second, surplus is not maximized in equilibrium
– that is, there is another outcome that would make both players better off. If both
players confess, each spends two years in prison. A better outcome could be achieved
Player 2
Silent Confess
Player 1
Silent (1,1) (3,0)
Confess (0,3) (2,2)
FIGURE 3.4 The prisoner’s dilemma
if both players stayed silent. However, this is not sustainable as an equilibrium because
each player has an incentive to deviate; given that the other player stays silent, he will
go free if he switches his choice to ‘confess’. It is these two features that allow us to
characterize a game as a prisoner’s dilemma.3
The prisoner’s dilemma has a number of real-world applications, outside the rather
limited scenario outlined above. For example, it can be used to explain why firms in
an industry do not collude and charge high prices to consumers, even though this would
increase their profits (see Figure 3.1 above). It could represent an industry with two
firms (department stores or supermarkets for instance) that choose to advertise,
even though they would be both better off if they could commit not to do so. It has
also been used to explain the arms race during the Cold War; both the USA and the
USSR would have been better off if both parties had chosen to disarm.
3.4.2 Coordination game

Another important game is a coordination game. In one formulation of this game,
two drivers meet on a narrow road and must decide which way to swerve in order to
avoid colliding with each other. If both execute the same swerving manoeuvre, they
will avoid the collision (with a payoff of 1 each); however, if they choose different
manoeuvres, they will crash (yielding a payoff of 0 for each driver). The normal form
of this game is depicted in Figure 3.5. The feature that makes this game a coordination
game is that there are two Nash equilibria, and the parties are trying to coordinate
on which of those two equilibria is realized.
A variant of this is a game called battle of the sexes, in which Jay and Daisy are
trying to organize a date. Jay would prefer to go to the opera, whereas Daisy would
Driver 2
Left Right
Driver 1
Left (1,1) (0,0)
Right (0,0) (1,1)
FIGURE 3.5 The coordination game

Daisy
Opera Boxing
Opera (2,1) (0,0)
Jay
Boxing (0,0) (1,2)
FIGURE 3.6 Battle of the sexes
prefer to go to see a boxing match. However, both would prefer to be together rather
than alone. The normal form of this game is depicted in Figure 3.6, with its associated
payoffs. The distinguishing feature of this game is that the two parties would like to
coordinate, but have opposite preferences as to which of the equilibria is chosen – that
is, Jay prefers (Opera, Opera) whereas Daisy prefers (Boxing, Boxing).
Another variant is the stag hunt, in which two hunters have the choice of hunting
stag or hunting hare. If either chooses to hunt hare, he will be guaranteed to catch a
hare. However, successfully hunting stag requires both hunters, and catching a stag
yields a higher payoff for both hunters. This game is depicted in Figure 3.7. Again,
this game has two Nash equilibria, but the distinguishing feature is that both parties
prefer the same equilibrium (Stag, Stag).
Finally note that the game depicted in Figure 3.3 is also a coordination game with
two equilibria (T, R) and (B, L).
Hunter 2
Stag Hare
Hunter 1
Stag (2,2) (0,1)
Hare (1,0) (1,1)
FIGURE 3.7 Stag hunt
3.5 Sequential games
In this section, we consider games in which one party (the leader) chooses their action
first. The other player (the follower) observes the leader’s choice before selecting their
own action. We will consider how to represent and solve these games.
3.5.1 Representing sequential games: the extensive form

As before, we can represent sequential games using the normal form of the game.
However, it is often convenient to represent sequential games using the extensive
form of the game. This involves drawing out the order of the choices in the form of
a ‘game tree’.
(–5,5)
E (10,10)
(0,20)
FIGURE 3.8 Extensive form of a game
To see how this is done, consider the following example. There is currently one firm
operating in the market (denoted ‘I’ for ‘incumbent’) and there is one other firm that
is contemplating entry (denoted ‘E’ for ‘entrant’). The game proceeds as follows. First,
the Entrant chooses whether or not to enter the market or not. The dot here – or the
node – represents the point at which the Entrant can make a decision, with her possible
choices being ‘enter’ or ‘not enter’. If the Entrant does not enter, the Entrant receives
a payoff of $0 and the Incumbent receives $20. Alternatively, if the Entrant chose to
‘enter’, the incumbent observes this choice, and then decides whether to accommodate
or punish the entrant. If the Entrant enters and the Incumbent accommodates, each firm
makes a profit of $10. Finally, if the Entrant enters and Incumbent punishes, the profits
are –$5 and $5 to the Entrant and the Incumbent, respectively.
This information is represented in the extensive form in Figure 3.8. The left-most
decision point (node) is labelled ‘E’ to signify that the Entrant’s choice is made first.
The branches originating from this node denote the two actions that the entrant
may take. The node labelled ‘I’ and the branches originating from that node denote
the Incumbent’s choice of actions. The choices of the players trace out a ‘path’ along
the game-tree, and the payoffs associated with these choices are shown at the end of
each path.
3.5.2 Solving sequential games

We now turn to the question of what the outcome of these games will be. Again, we
can solve a sequential game by assuming that each player is only interested in
maximizing his own payoff. Again, it should be noted that the players do not care about
each other’s payoffs.
Nash equilibrium
Like simultaneous-move games, we can solve sequential games by determining the
Nash equilibrium of the game. In order to do this, we will need to represent the game
in normal form and identify the best responses of each player, as we did in Section
3.3.2.
As an illustration of this, Figure 3.9 depicts the normal form of the game in
Figure 3.8. Solving for the best responses of each player, we can identify two Nash
equilibria:
Incumbent
Accommodate Punish
Entrant
Enter (10,10) (5,−5)
Not enter (0,20) (0,20)
FIGURE 3.9 The normal form of the game depicted in Figure 3.8
• (Enter, Accommodate): If the Entrant chooses to enter, the Incumbent’s best

response is to accommodate. If the Incumbent chooses to accommodate, the
Entrant’s best response is to enter.
• (Not enter, Punish): If the Incumbent chooses to punish, the Entrant’s best
response is to not enter. If the Entrant chooses to not enter, the Incumbent’s best
response is to punish (the Incumbent cannot do better by choosing an alternative
to punish).
The first of these equilibria has some intuitive appeal. Given the payoffs, the best thing
the Incumbent can do in response to an entry is to accommodate. The entrant, predicting
this accommodation after entry, should therefore enter the market in the first place.
However, the second equilibrium is somewhat less intuitive. In this equilibrium, the
Entrant chooses not to enter, because it anticipates that the Incumbent will punish.
However, the threat of punishment is not credible; we know that if the Entrant actually
does enter the market, the best thing the Incumbent can do in those circumstances is
accommodate. In order to address this, we may wish to find a solution concept that
only identifies credible equilibria.
Subgame perfect equilibrium

The concept of subgame perfection identifies only credible equilibria. To do this, it
breaks down the larger game into smaller mini-games, or ‘subgames’, each of which
comprises one choice by one player.4 Thus, in the game depicted in Figure 3.8, the
Entrant’s choice to enter or not enter is one subgame and the Incumbent’s choice to
accommodate or punish is another subgame.
A subgame perfect equilibrium (SPE) exists where each player’s chosen actions
are a Nash equilibrium in every subgame. That is, we would require the Entrant’s choice
to enter or not enter to be his best response in the circumstances, and we would also
require the Incumbent’s choice to accommodate or punish to be her best response in
the circumstances (that is, if the Entrant did Enter).
Note that the follower’s best response depends on what she observes the leader’s
choice to be. Hence, the leader’s best response depends on what he expects the
follower will do. That is, the leader will try and anticipate the choice of the follower,
before deciding upon his own choice of action. In order to determine the SPE of a
game, we solve the game using backwards induction. This entails starting at the
end of the game, and solving backwards towards the beginning. So, in the game
depicted in Figure 3.8, we would first determine what the Incumbent would do if it
reaches the node marked ‘I’. Comparing the payoffs, we can see that the Incumbent
would choose to accommodate. Now, we can determine what the Entrant will do at
the node ‘E’, knowing that if it chooses to enter, the Incumbent will choose to
accommodate. Again, comparing the payoffs, we can conclude that the Entrant will
enter. Hence, the SPE of this game is (Enter, Accommodate).
Solving backwards (and finding the SPE) reflects the fact that the players themselves
are rational and forward looking. The Entrant, before making its choice, thinks ahead
and tries to anticipate what the Incumbent would do in any given situation. As
economists, the way we try to capture and model this forward thinking is to go to the
end of the game and to work backwards.
Note that the SPE is a Nash equilibrium, but not all Nash equilibria are SPE. This
is because some Nash equilibria are sustained by non-credible threats. For example,
the equilibrium (Not Enter, Punish) is a Nash equilibrium, but not a SPE. This is, in
fact, one of the strengths of using backwards induction to determine the equilibrium
outcome(s) of a game; a SPE cannot be sustained by non-credible threats that would
not actually be employed if the player involved ever had to actually choose.
Also bear in mind that, because subgame perfection requires us to determine the
choice of each player at every possible juncture, the specification of the SPE should
identify the choice made at each node, regardless of whether or not that node is actually
reached.
Example. Jay and Daisy are organizing a date. Jay would prefer to go to the opera,
whereas Daisy would prefer to go to see a boxing match. However, both would
prefer to be together rather than alone. To solve their coordination problem, they
have decided that Jay will go to a venue first, and then call Daisy to inform her
where he is. Following this, Daisy will then choose where she goes. The extensive
form of this game is depicted in Figure 3.10.
Using backwards induction, we first determine what Daisy will do at nodes D1
and D2 . At D1, Daisy will choose to go to the opera; at D2, Daisy will choose to
go to the boxing match. Now, Jay knows that if he chooses the opera, Daisy will
(2,1)
D1
(0,0)
J
D2 (0,0)
(1,2)
FIGURE 3.10 Extensive form of a game in which Jay and Daisy organize a date
also choose the opera; if he chooses the boxing, Daisy will also choose the boxing.
Knowing this, at the node J, Jay will choose the opera as it gives him the higher
payoff.
The specification of the SPE must reflect the choices that would be take at each
of the nodes, even though some of the nodes (namely, D2) are never reached. The
SPE is (Opera; Opera (if Opera), Boxing (if Boxing)); this tells us what is chosen
at each of the nodes, J, D1 and D2 in that order.
The tools outlined in this chapter are particularly useful in thinking about strategic
situations in which the interaction between players (be it workers, firms or governments)
is critical. One of the advantages of game theory is that it can make sometimes very
complicated environments to be analysed relatively simply. Game theory also has the
advantage of making explicit all the assumptions being made. This makes it possible
for others to see what is driving the conclusions of the model.
Notes
1 By convention, the row player’s payoff is given first, and the column player’s payoff is
given second. This is, of course, just a convention; what is crucial, however, is that the
payoffs drawn consistently.
2 There is also a third Nash Equilibrium, a ‘mixed strategy’ equilibrium, in which each player
chooses each possible action with a positive probability. This is a more advanced topic,
so in this book we will not consider mixed strategy equilibria.
3 Indeed, it is possible that a game has one, but not the other, of these features. For example,
the game depicted in Figure 3.2 has a dominant strategy equilibrium, but that equilibrium
is surplus maximizing. Any game that does not satisfy both properties is no a prisoner’s
dilemmas.
4 A subgame can be defined as part of a larger game that looks like a game itself.
PA R T I I
Gains from trade


C H A P T E R
4
Trade and
the PPF
4.1 Introduction
This chapter provides an introduction to why and how people trade with each
other. In particular, we focus on what determines patterns of trade and the
underlying sources of the gains from trade. As we will see, trade makes people
better off because there are gains from exchange and gains from specialization.
4.2 Gains from exchange
A basic proposition of economics is that trade makes people better off. One
of the reasons that trade makes people better off is because it helps allocate
goods and services to those who value them the most.
To understand this, consider a situation in which Baz owns a rarely-used
bicycle, which he values at $10. Chloe, on the other hand, does not own a
bike but would be willing to pay up to $100 to buy a bike. Now, suppose Baz
agrees to sell his bicycle to Chloe for $40. In this case, Baz and Chloe are
both better off: Baz previously owned a bike that he valued at $10, but now
he has $40 instead; Chloe used to have $40, but now owns a bike that she
values at $100. Note that both parties to the transaction are better off; trade
benefited both the buyer and the seller. This is because the trade is voluntary;
neither Baz or Chloe would agree to the trade if it made them worse off.
Specifically, Baz would not have agreed to sell the bicycle if the price was
less than his valuation of $10 and Chloe would not have agreed to buy the
bicycle if the price was greater than her valuation of $100. Similarly, Baz
would not have sold the bicycle to Chloe if his valuation was higher than
Chloe’s (say, $110 instead of $10), as there would be no price that would be
acceptable to both of them.
28 4 Gains from trade
In general terms, trade can only occur if the seller’s valuation of the item (vs) does
not exceed the buyer’s valuation (vb). Given that trade is voluntary, the seller will not
accept a price (p) lower than his valuation and the buyer will not agree to a price higher
than her valuation. Therefore, in order for trade to take place, it must be the case that:
vs ⭐ p ⭐ vb . (4.1)
Because no party will consent to a transaction that makes him or her worse off, trade
will always make people (weakly) better off.1
Two things should be noted here. First, in the above example, the price of $40 was
not the only price that allows for mutually beneficial trade; as you can see, a price of
$50 or $60 would also be mutually beneficial for both parties. In fact, any price that
meets the condition in Equation 4.1 will suffice for trade to occur. The price is
relevant, however, in determining how the gains from trade are split between two
parties. For example, when the price was $40, trade made Baz $30 better off and
Chloe $60 better off; however, if the price was instead set at $50, Baz would be
$40 better off but Chloe would only be $50 better off.2 The exact price at which the
item is traded depends on bargaining between the parties, which we will discuss later
in Chapter 5.
Second, the discussion so far has assumed that money is our medium of exchange.
However, our results hold true even if the parties barter one good for another. For
example, suppose Baz swaps his bicycle for Chloe’s skateboard. Baz would only agree
to the trade if he preferred the skateboard to the bicycle, and Chloe would only
agree if she preferred the bicycle to the skateboard; thus, both parties are made
better off. The role of money is simply to help coordinate (or facilitate) trade between
parties; that is, it allows Baz and Chloe to trade, even if Chloe does not have anything
to trade that Baz wants.
4.3 Gains from specialization
Trade also allows people to take advantage of gains from specialization. In order to
appreciate why this is so, it is first necessary to consider the constraints that individuals
face in production. It will then become clear that parties can be made better off by
specializing in the production of one good and then trading that good in exchange for
other products or services.
4.3.1 The production possibility frontier (PPF)

Sometimes, because resources are limited, we face tradeoffs in terms of what can be
produced. If a particular resource is used to make one good, that resource cannot be
used to produce another good. To understand this better, we can graph the output that
an individual (or a country) can produce for a particular set of resources. This graph
is called a production possibility frontier (PPF).3 The PPF traces out combinations
Trade and the PPF 29
C
Y
PPF
X
X
FIGURE 4.1 The production possibility frontier (PPF) traces out combinations of the quantity of two goods
(X and Y ) that can be produced if all resources are used
of the quantity of two goods that an individual or a country can produce if it uses all
of its resources. An example of a PFF is illustrated in Figure 4.1.
Suppose Australia can use its resources to produce either good X or good Y, or some
combination of both. If all of Australia’s resources are allocated to making Y, Australia
–
will produce Y of Y and zero units of X; similarly, if all of the country’s resources go
–
into X production, X are made of X but zero units of Y. But combinations of X and Y
are also possible – these possible combinations are represented by the line joining
– –
Y and X .
There are several points worth noting about the PPF. First, any point inside or on
the PPF is obtainable in that it is feasible to produce that quantity of X and Y given
the current level of resources and state of technology. Second, any point on the PPF
(e.g. Point A in Figure 4.1) is efficient in the sense that it makes full use of the available
resources, whereas any point inside the PPF (e.g. Point B) is inefficient because it does
not make full use of the available resources. Third, any point outside the PPF (e.g.
Point C) is not feasible, because production of those levels of X and Y would require
more resources (or better technology) than Australia currently has.
As we have discussed, the shape of the PPF depends on the current levels of
resources and technology. Therefore, if either the amount of resources available or the
state of technology changes, so does the shape of the PPF. If there is an increase in
resources or an improvement in technology that boosts the production of both goods,
the PPF will shift outwards from origin along both axes, as shown in Figure 4.2. This
indicates that more of both X and Y can be produced. This might happen if there is an
increase in population (that is, an increase in the available labour) or an improvement
of some technology that is used in the production of both goods.
Sometimes, the increase in resources or technology only affects the production of
one good. For example, if the two goods were bookshelves and cars, an increase in
PPF PPF'
X
FIGURE 4.2 A production possibility frontier for goods X and Y. If there is a shock that boosts the
production of both goods, the PPF will shift outwards from origin along both axes
the availability of high-quality wood would only affect the production of bookshelves;
similarly, an improvement in tyre-making technology would only affect the production
of cars. In such instances, the PPF will rotate or stretch outwards only along the axis
of the affected good, as in Figure 4.3.
It is, of course, possible that there is a decrease in the level of resources or a decline
in the level of technology. In such cases, following the reasoning above, the PPF will
shift or rotate inwards towards origin.
Note, however, that while changes in population and in technology can both shift
the PPF, an increase in population could increase total output without increasing the
PPF PPF'
X
FIGURE 4.3 A production possibility frontier for goods X and Y. If there is a shock that boosts the
production of X only, the PPF will shift outwards from origin along the x-axis only
quantity of output per person. By contrast, increases in technology will increase both
total output and output per person. Output per person (or per worker) is really a measure
of labour productivity, and it is changes in labour productivity that ultimately drive
changes in standards of living (consumption per person or worker). This also helps
explain why large countries such as China can have significantly higher output than
smaller countries, while not necessarily having as high of an average standard of living
per capita.
Finally, we can use the PPF to measure the opportunity cost of producing a particular
unit of a good. Suppose, in Figure 4.4, Australia is currently at a point on the PPF
where it is making some of both goods (say, Point A). Note that, because Australia is
on the PPF, it is fully utilizing its resources. Therefore, in order to increase the
production of X, it is necessary to give up some Y. The amount of Y that must be
foregone depends upon the slope of the PPF. For example, suppose Australia would
like to move from A to A′, increasing the production of X by one unit. In order to do
so, it is necessary to give up y units of Y. However, if Australia were moving from
B to B′, it would have to give up y′ units of Y. As you can see, the slope of the PPF
shows the opportunity cost in terms of Y forgone of getting more X.4
In general, when the output of X is low, the opportunity cost of an additional unit
of X is relatively low. This is because some resources are better suited to making X
and others are better suited to making Y, and as we want more X we switch over the
resources that are best at producing X first. But as the production of X increases,
the resources that are reallocated from Y to X production becoming increasingly less
suited to making X. So, in order to produce an extra unit of X we need to take away
increasingly more resources from the production of Y. This is reflected in the shape
of the PPF – near the vertical axis the slope is relatively flat, indicating that the
A
y A'
y'
B'
PPF
X
1 unit 1 unit
FIGURE 4.4 A production possibility frontier for goods X and Y. A one unit increase in the production
of X is less costly when the output of X is low (e.g. from A to A′) than when the output of X is high
(e.g. from B to B′)
opportunity cost of producing some more X (in terms of Y) is relatively small. But as
more and more X is produced, the opportunity cost of producing X increases, as
indicated by the steeper slope.
4.3.2 Absolute advantage and comparative advantage

Now that we have the means for analysing the production possibilities for one
individual or country, we can also compare the production capabilities of different
individuals or different countries. In particular, we can examine which individual or
country has the advantage in the production of any particular good. In economics, we
refer to two types of advantage:
• We say that Party A has an absolute advantage over Party B in the production
of a good if, for a given amount of resources, A can produce a greater number of
that good than B.
• We say that Party A has a comparative advantage over Party B in the production
of a good if A’s opportunity cost of producing that good is lower than B’s
opportunity cost.
To illustrate the difference between absolute advantage and comparative advantage,

consider the following example.
Example. Broderick takes one hour to make a pepper mill and one hour to make
a salt shaker. Christopher takes four hours to make a pepper mill and two hours
to make a salt shaker. Both Broderick and Christopher work for eight hours a
day. The following table depicts how much of each good they can make in one
working day:
Pepper mills Salt shakers
Broderick 8 8
Christopher 2 4
As you can see, Broderick can produce more pepper mills in a day than
Christopher; Broderick can also produce more salt shakers in a day than
Christopher. Therefore, Broderick has the absolute advantage in the production of
both pepper mills and salt shakers.
Now consider Broderick’s and Christopher’s opportunity cost of producing
pepper mills and salt shakers. If Broderick makes one pepper mill, that takes him
one hour. In that time, he could have instead made one salt shaker. Therefore, in
order to make that pepper mill, he has to give up making one salt shaker, so his
opportunity cost of one pepper mill is one salt shaker. Similarly, if Broderick makes
one salt shaker, he has to give up one pepper mill, so his opportunity cost of one
salt shaker is one pepper mill. On the other hand, if Christopher makes one pepper
mill, it takes him four hours, in which time he could have made two salt shakers;
therefore, his opportunity cost of one pepper mill is two salt shakers. By the same
reasoning, his opportunity cost of one salt shaker is half a pepper mill. The
opportunity cost of each good for Broderick and Christopher is represented in the
following table:
Opp. cost of 1 Opp. cost of 1
pepper mill salt shaker
Broderick 1 salt shaker 1 pepper mill
1
Christopher 2 salt shakers ⁄2 pepper mill
From this table, we can see that Broderick has the comparative advantage in
producing pepper mills, as his opportunity cost of producing pepper mills is
lower. On the other hand, Christopher has the comparative advantage in producing
salt shakers, as his opportunity cost of producing salt shakers is lower than
Broderick’s. Note that Christopher has the comparative advantage in producing
salt shakers, even though he does not have the absolute advantage in producing
anything.
As illustrated in the above example, it may be possible for one person to have the
absolute advantage in the production of both goods. However, as a rule of thumb, it
is not possible for one person to have the comparative advantage in more than one
good. This is because, for each person, the opportunity costs of the two goods will be
the inverse of each other; therefore, if one person has the lower opportunity cost for
one good, he must have a higher opportunity cost for the other good.
4.3.3 Specialization
Now that we have the tools to analyse and compare the production capabilities of
two individuals, we can illustrate how those individuals can gain from trading with
each other.
Suppose Michelle and Rodney spend ten hours a day making shoes. It takes Michelle
two hours to make a left shoe, and half an hour to make a right shoe. Rodney takes
half an hour to make a left shoe and two hours to make a right shoe. The following
table shows the maximum number of left shoes and right shoes each person can make
in one day:
Left shoes Right shoes

Michelle 5 20
Rodney 20 5
Michelle’s and Rodney’s PPFs are depicted in Figure 4.5. Let’s also suppose
that Michelle and Rodney do not like to wear left shoes without right shoes, and vice
versa – that is, they like their shoes to be in pairs.
First, let us assume that there is no trade between the parties. If Michelle wants one
left shoe for every right shoe, she must produce four left shoes and four right shoes.
Similarly, Rodney will have to produce four left shoes and four right shoes in order
to have pairs. This means that, between them, they can produce eight pairs of shoes.
Now let us allow the parties to trade with each other. In this case, if Michelle special-
izes in producing right shoes, she can make 20 right shoes a day; if Rodney specializes
in making left shoes, he can make 20 left shoes a day. Between them, Michelle and
Rodney can make 20 pairs of shoes. Supposing that they trade between them at a rate
of one left shoe for one right shoe, Michelle and Rodney can now each have 10 pairs
of shoes. These consumption points are marked in Figure 4.5; note that they lie outside
each person’s individual PPFs. Here, the gains from trade are manifested in the ability
of each party to consume more than the amount he or she could produce alone.
When the parties specialize in producing the good that they have a comparative
advantage in, total production increases (here from a total of 8 pairs of shoes to 20
pairs of shoes between Michelle and Rodney). Total output increases because trade
allows parties to specialize in producing the good in which they have the lower
opportunity cost. And with more output, both trading parties can potentially be made
better off. In other words, trade create an environment for specialization to be feasible,
increasing the size of the economic pie; this increase in output can potentially be shared
so as to make everyone better off than without trade.
Although we have given a specific example to illustrate the gains from specialization,
the idea is a very general one. That is, trade is beneficial to individuals (and indeed
countries) because it allows them to specialize in industries where they have the
comparative advantage, and trade with others for things that would cost them more to
produce personally. Moreover, this principle holds even if one party has the absolute
advantage in the production of both goods; what matters is the comparative advantages
or opportunity costs of the parties.
Michelle Rodney
Left Left
20
10 10
5
4 4
Right Right
4 10 20 q2 4 5 10 q2
FIGURE 4.5 Michelle and Rodney’s PPFs. When Michelle and Rodney cannot trade, their
consumption points are given by the point A. When Michelle and Rodney can trade with each other,
their consumptions points are given by the point B. Note that point B lies outside each person’s
individual PPF
A basic proposition of economics is that there are gains from trade. First, if trade is
voluntary any party willing to trade must be at least as well-off as they were without
trading. Second, the gains from trade arise due to gains from exchange – reallocating
goods to those who value them most – and from specialization – allowing parties to
specialize in producing the good that they have the lowest opportunity cost. By
allowing the parties to specialize in producing the good that they have a comparative
advantage, total output increases, potentially making all trading parties better off.
Notes
1 By the phrase ‘weakly better off’, we mean that people will either be better off or just as
well off as they were before (in other words, no one is worse off). A party will be just
as well off after trade as they were before if the price was set exactly at their valuation.
2 Again, if price exactly equals either the buyer’s or the seller’s valuation, that individual
would not gain from trade, but they would not be worse off either.
3 Sometimes it is also called a production possibilities curve (PPC).
4 Of course, the PPF need not be concave. If the PPF is a straight line (that is, its slope is
constant), the opportunity cost between X and Y is constant, regardless as to how much
of either product is made.

C H A P T E R
5
Bargaining
5.1 Introduction
Bargaining is an integral part of commerce. For example, a person might

bargain over the price of a car with a salesperson, unions bargain with
employers over wages and working conditions and in some countries it is usual
for bargaining to occur over the price of most goods and services, such as
haggling at a bazaar.
In the last chapter, we found that when two people trade with each other,
there are gains from trade. Bargaining helps determine how those gains are
split between the parties by determining the price at which a good or service
is traded. In this chapter we will build upon the game theory concepts in
Chapter 3 to construct a simple model that roughly represents a bargaining
process. We do this in order to derive some understanding about what the
outcome of bargaining will be. While the material here can be read as a
standalone topic, we strongly recommend that you familiarize yourself with
Chapter 3 before embarking on this chapter.
5.2 Bargaining and surplus
We will begin by setting up a general framework within which bargaining

will take place. Suppose there are two parties to the transaction: the seller
(A) and the buyer (B). The parties are bargaining over the price at which a
good will be sold. The seller values the good at vA and the buyer’s valuation
is vB . In order for voluntary trade to take place, we know that the seller’s
valuation of the good must not exceed the buyer’s valuation (vA ⭐ vB). If the
trade takes place, the gains from trade will be the difference between the two
valuations – that is, vB – vA .
The parties bargain with each other in order to determine the price of the good and
hence the amount of surplus that each party receives; in other words, bargaining helps
determine how the gains from trade are split.
Example. Bill would like to buy a parcel of land from Josie. Josie values the land
at $300,000 and Bill values the land at $400,000. If the sale goes ahead the gains
from trade will be $100,000, but if the sale does not occur there will be no gains
from trade. After extensive bargaining, Bill and Josie come to the agreement that
Josie will sell the land to Bill at a price of $328,000. At this price, Bill’s net surplus
from the transaction is $400,000 – $328,000 = $72,000. Josie’s net surplus from
the transaction is $328,000 – $300,000 = $28,000. The total surplus (that is, Bill’s
surplus plus Josie’s surplus) from the transaction is $100,000 – but it is the price,
determined through bargaining, that determines how the surplus of $100,000 is
split between the two parties.
5.3 Take-it-or-leave-it negotiations
Let us first suppose that only one take-it-or-leave-it offer will be made (this situation
is also called an ‘ultimatum game’). Specifically, assume that the bargaining process
is as follows. The seller (A) makes an offer to the buyer (B) as to the price of the good.
Let us say that A offers a price of p. B can either (a) accept the offer and trade will
occur at A’s suggested price or (b) reject the offer and no trade occurs.1 If B accepts
the offer, she will receive a surplus of vB – p and A will receive a surplus of p – vA .
If B rejects the offer, the trade does not go ahead, so the surplus of both parties is zero.
This sequence of events is depicted in Figure 5.1.
To determine what will happen, we need to solve the game backwards to find the
subgame perfect equilibrium – that is, we need to go to the end of the game and work
our way to the beginning. This is because A is forward thinking and rational, so he
will do his best to anticipate how B will react to any offer he makes. In Figure 5.1, at
B, the buyer will accept the offer if her payoff from doing so (vB – p) is at least as
good as her payoff from rejecting (zero); otherwise, she will reject the offer. Therefore,
at the first stage, the seller can induce the buyer to accept by ensuring that vB – p ⭓ 0
– that is, p ⭐ vB . In other words, the highest price that the seller could charge is
p = vB .2 It remains to be determined whether the seller should make an offer that the
buyer will accept. If the buyer accepts, the seller’s surplus will be p – vA ; if the buyer
( − , − )
Offer B
A
(0,0)
FIGURE 5.1 Negotiation in which A makes a single offer of p

Bargaining 39
rejects, the seller’s surplus will be zero. If the seller sets the highest price that will be
acceptable to the buyer, his payoff will be p – vA = vB – vA . Since we know that the
buyer’s valuation is at least as great as the seller’s (vB ⭓ vA), this implies vB – vA ⭓ 0.
It follows that the seller will prefer to set a price that is acceptable to the buyer.
Therefore, the subgame perfect equilibrium path in this game is as follows: A
makes an offer of p = vB and B accepts the offer. A’s surplus from the transaction is
vB – vA and B’s surplus is zero. As you can see, all the gains from trade in this scenario
accrue to A.
Let us now suppose that it is the buyer who gets to make the offer. Now, B offers
a price p, which A either accepts or rejects. If A accepts, the trade goes ahead and the
payoffs are (p – vA, vB – p) to A and B respectively. If A rejects the offer, both parties
get a payoff of zero. This sequence of events is depicted in Figure 5.2. Solving this
game backwards, the seller will only accept the offer at A if p – vA ⭓ 0. Consequently,
in order to induce the seller to accept, the buyer must offer at least p = vA. It is in the
interest of the buyer to induce the seller to accept because her payoff from acceptance
will be vB – vA, whereas her payoff from rejection will be zero. Therefore, the subgame
perfect equilibrium path in this game is as follows: B makes an offer of the lowest
price that will be accepted – that is p = vA – and A accepts the offer. A’s surplus from
the transaction is zero and B’s surplus is vB – vA . In this case, all the gains from trade
accrue to B.
( − , − )
Offer A
B
(0,0)
FIGURE 5.2 Negotiation in which B makes a single offer of p
In these single-offer games, the person making the offer receives all the gains from
trade. This is because that party is the only party who can make offers and hence has
all the bargaining power. The other party only has the power to accept or reject the
offer, but rejecting the offer means that they will receive zero surplus.
Example. Andrew wants to sell a bottle of scotch whiskey. Andrew (A) values
the bottle at $0. Bonnie (B) is the only potential buyer and she values the bottle
of scotch at $100. Negotiations proceed as follows. Andrew proposes a price (p)
that Bonnie can pay for the bottle. Bonnie then accepts or rejects the offer. If she
accepts, Andrew gets a payoff of p – 0 and Bonnie gets a surplus of 100 – p.
If Bonnie rejects the take-it-or-leave-it offer, bargaining is finished and both
parties get a payoff of 0. (This game is the same as illustrated as in Figure 5.1.)
To determine the outcome of the negotiation, go to the end and solve backwards.
The most Bonnie would be willing to pay for the bottle would be a price of 100.
Anticipating this Andrew will offer the highest price Bonnie will accept, which
is $100 (or $99.99 if he wants to ensure Bonnie is strictly better off from accepting
the offer). Bonnie accepts this offer – she is no worse off accepting this price than
from rejecting the offer and getting 0. All the gains from trade go to Andrew –
his surplus is $100. Bonnie, while she does end up with the scotch, gets net surplus
of 0 ($100 – $100).
5.4 Multiple-offer bargaining
In some circumstances, it will be more realistic to assume that multiple offers can be
made. Thus, one party makes an offer; the other party either accepts, or rejects and
makes a counter-offer, and so on. As an illustration of this, consider the following
sequence of events. First, the seller (A) makes an offer of p1. The buyer (B) can then
either accept or reject the offer. If she accepts, trade occurs at the price p1; if she rejects,
she can make a counter-offer of p2. Finally, the seller can respond by accepting or
rejecting the offer of p2. If he accepts, trade occurs at the price p2; otherwise, the trade
does not go ahead and both parties receive zero surplus. Figure 5.3 depicts the extensive
form of this game.
Again, we need to solve the game backwards to determine what happens in the
subgame perfect equilibrium.
• At A2, the seller will only accept an offer if p2 – vA ⭓ 0. Therefore, in order to

induce the seller to accept, the buyer must offer at least p2 = vA.
• At B2, the buyer will offer p2 = vA because he wants to get the seller to accept
(the buyer’s payoff from acceptance will be vB – vA, whereas her payoff from
rejection will be zero).
• At B1, the buyer will only accept the offer from A if vB – p1 ⭓ vB – vA, or that
p1 ⭐ vA = p2. This is because, if the buyer rejects, she can be certain that the seller
will later accept an offer of p2 = vA. Therefore, in order to induce the seller to accept,
the buyer B must offer at least p1 = p2 = vA.
• At A1, the seller will make an offer of p1 = p2 = vA, because the seller is indifferent
between the buyer accepting or rejecting the offer. The seller’s payoff from
acceptance and rejection will be zero in either case.
( 1− , − 1)
Offer B1
1
A1 ( 2− , − 2)
Offer A2
2
B2
(0,0)
FIGURE 5.3 Negotiation in which A makes an offer of p1 and B makes a counter-offer of p2

Bargaining 41
In this situation, B receives all the gains from trade. This is because she gets to make
the last offer; that is, in the last round, she is able to credibly commit that this is the
last chance to trade, meaning that A’s only possible choices are to accept or reject the
offer received.
In general, when multiple offers or counter-offers can be made, the party who can
make the final offer holds all the bargaining power. Therefore, under the conditions
we have outlined, we would expect the party that makes the final (take-it-or-leave-it)
offer will receive all of the surplus.
5.5 Some caveats
It should be remembered that the extreme results derived in the previous sections relied
on some very strict assumptions. Where these assumptions do not hold, we might
observe different outcomes. Often, in practice, the gains from trade are shared by
the parties, rather than enjoyed by one party alone. There are several reasons why
this may be the case:
1 There may be costs to the bargaining process. These costs may include the cost
of time lost by negotiation or the cost of paying someone to negotiate on your
behalf. In this case, each party knows that the other will be more reluctant to
proceed to the next round of offers and acceptance/rejection, because it will be
costly to do so. This implies that parties will be willing to accept slightly lower
offers in earlier rounds, if it means they can avoid some negotiation costs.
2 The parties may not know when bargaining is going to end. Negotiating parties
typically do not know who will make the last possible offer (that is, the parties
are not exactly sure when the negotiations will be terminated if an agreement is
not reached). This means it may be unclear which party will have the ‘final say’,
so there may be uncertainty as to the distribution of bargaining power.
3 The parties may not know each other’s valuations of the object. In the sections
above, the party with the bargaining power offered a price equal to the other party’s
valuation. However, this will not be possible if the parties do not know each other’s
valuations, which may mean that the offering party is unable to extract all the
surplus.
4 One or both of the parties may have an ‘outside option’. An outside option means
that if negotiations break down, the parties have the option of trading with someone
else. For example, suppose A wants to sell an apple to B. A has an outside option
if, in the event that negotiating with B fails, he can sell the apple to C instead. B
has an outside option if, in the event that negotiating with A fails, she can buy an
apple from D instead. If a party has an outside option and all offers are rejected,
that party will receive a payoff greater than zero. This is because they will receive
some surplus from exercising that outside option. Hence, in order to induce
acceptance, the other party will need to make an offer at least as good as the outside
option.
Example. Suppose Ed wants to buy an apple from Rob. However, Rob also
has the option of selling the apple to Hansen for $5. If Ed proposes a price for
the trade, he will need to offer Rob at least $5 in order for Rob to agree to the
trade.
Sometimes, in practice, there is a breakdown in bargaining. These situations are

also unexplained by the model set out in the previous sections. One reason that this
may occur is if there is asymmetric information between the parties. This may occur
if the parties do not know each other’s valuations of the good. For example, if the
seller overestimates the buyer’s valuation of the good, he may offer a price that is too
high for the buyer to accept. Conversely, if the buyer underestimates the seller’s
valuation of the good, she may offer a price too low for the seller to accept. In these
situations, the two parties may end up not trading, forgoing the potential gains from
trade. A similar point can be made when agreement is only reached after a protracted
(and potentially costly) delay in negotiations, such as with a strike or a lockout. Even
though agreement is eventually reached, some of the potential gains from trade are
lost during negotiations.
In Chapter 4, we learned that there are gains from trade. In this chapter, we discussed
how those gains are divided between the parties. In general, the party with more
bargaining power will receive a greater share of the benefits. In the models we
examined we show that bargaining power can arise if a party has the ability to credibly
make the final offer. However, the distribution of surplus can also be affected by
bargaining costs, the existence of outside options and beliefs about when negotiations
will end. Finally, we found that bargaining may break down in the presence of
asymmetric information, meaning that gains from trade may not be realized.
Several implications arise from these sorts of bargaining models. As noted,
bargaining power tends to arise from some economic characteristic or the nature of
the bargaining process. This means that a party to a negotiation process, such as a
business or a union, might want to manipulate how the negotiations occur in order to
get more of the gains from trade. For example, a business might explicitly cultivate
an outside option in order to strengthen its bargaining position. Similarly, a union might
build up a war-chest of reserves to help members ensure a long strike (lower their costs
of delay, making it more difficult for the other party to outlast them).
Finally, the negotiation process we have considered here involves two parties (and
possible a third party as an outside option). Often there are many potential buyers and
sellers who could trade with one another. In such a market, price – and hence the
distribution of the gains from trade – is determined by the interaction of these many
buyers and sellers in the marketplace. We turn to these sort of markets now.
Bargaining 43
Notes
1 Note that this describes many transactions that we are likely to be familiar with; the seller
sets a price and the buyer either purchases the good at that price or not.
2 Note that, at a price of p = vB , B gets zero surplus and is indifferent between accepting
and rejecting the offer. For convenience, we will assume that if B is indifferent, she accepts
the offer. Alternatively, we could assume that an offer slightly less than vB is made (for
example, vB – ␩ where ␩ is a small positive amount, like 1 cent), which makes accepting
better than rejecting for B. In this case, our analysis remains the same in substance.

PA R T I I I
Market
fundamentals

C H A P T E R
6
Demand
6.1 Introduction
As noted before, we assume that consumers act in their own best interest. This
allows us to examine consumer behaviour under the assumption that a
consumer will maximize the benefit (or utility) he or she receives from
consuming goods and services, subject to their budget constraint.
In this chapter, we examine the choices of individual consumers in
competitive markets. In a competitive market, the actions and choices of
individual consumers cannot affect the price in the market, so they make their
consumption decisions taking prices as given (that is, consumers are price
takers). In this setting, we will examine a consumer’s decision as to how much
of a product to buy, and hence derive demand curves for individual consumers
and for the market.
6.2 Benefit and willingness to pay
A consumer derives some benefit from consuming a particular good or

service. We can measure that benefit by considering the consumer’s
willingness to pay (WTP) for that good or service. For example, to measure
the benefit that a consumer receives from consuming a cup of coffee, we could
ask, ‘What is the highest amount of money that the consumer would be willing
to pay for that cup of coffee?’. As you can see, this should tell us how much
the consumer values the cup of coffee in monetary terms.
When an individual consumes multiple units of a particular good or service,
we might like to distinguish between total benefit and marginal benefit.
Suppose Candice’s willingness to pay for coffee is $4 for the first cup, $3 for
the second cup and $2 for the third cup. Her total benefit (TB) measures the
48 6 Market fundamentals
benefit that she gets from consuming the total number of cups of coffee; that is, her
total benefit for the three cups of coffee is $9 (= 4 + 3 + 2). By contrast, her marginal
benefit (MB) measures how much extra benefit she derives from consuming one extra
cup of coffee; thus, her marginal benefit is $4 for the first cup, $3 for the second cup
and $2 for the third cup of coffee. Marginal benefit is the change in total benefit derived
from consuming one extra unit of the good. If we are given an equation that links total
benefit with the number of units consumed (q), we can find marginal benefit by
differentiating the total benefit function:
⌬TB dTB
MB = = . (6.1)
⌬q dq
In general, we would expect marginal benefit to decline with each additional unit
consumed. In the above example, this would mean that Candice would find the first
cup of coffee outstanding, the second cup very good, the third cup acceptable, and so
on. This decline in MB is referred to as diminishing marginal benefit.
It is also typical to draw marginal benefit as a continuous (smooth) function. Due
to diminishing marginal benefit, the marginal benefit curve would usually be a
downward-sloping line, as seen in Figure 6.1.1
MB
q
FIGURE 6.1 A typical marginal benefit curve
6.3 Individual demand
We can use a consumer’s marginal benefit curve to derive his individual demand curve.
Individual demand is the quantity of a good or service that a consumer is willing
and able to buy at a certain price. Hence, the individual demand curve traces out all
combinations of (a) market price and (b) individual demand at that price, holding
everything else constant.
Demand 49
We now need to determine the quantity of a good that an individual demands at a

certain price; as we will show, a consumer will purchase units of the good up until the
point where P = MB. To see this, note that if P < MB for a unit of the good or service,
a consumer should buy that unit because his willingness to pay exceeds the price. How-
ever, if P > MB for any unit of the good or service, the consumer should not buy that
unit because the price of the good exceeds his willingness to pay. We also know that,
due to diminishing marginal benefit, the MB of earlier units will be greater than the MB
of later units. Since the consumer will buy additional units if P < MB and fewer units
if P > MB, it follows that a consumer should buy units up until the point where P = MB.
Consequently, a consumer’s individual demand curve is given by his MB curve.
Further, because the MB curve is usually slopes downward, so too will the individual
demand curve. A demand curve represents how much a consumer is willing and able
to buy at different prices. Figure 6.2 depicts a typical demand curve. As you can see
the downward slope of the curve means that the higher the market price, the fewer
units a consumer buys (q1, p1); similarly, the lower the market price, the more units a
consumer buys (q2, p2). This negative relationship between price and quantity demanded
is known as the law of demand.
The demand curve is derived by assuming that only price and quantity can change.
If there is a change in the price/quantity, there will be a movement along the demand
curve itself (as in Figure 6.2), with the change from (q1, p1) to (q2, p2), which is called
a change in the quantity demanded. If there is a movement downwards along the
demand curve (from (q1, p1) to (q2, p2) for example), this is called an ‘increase in
the quantity demanded’; if there is a movement up along the demand curve (that is,
from (q2, p2) to (q1, p1)), this is called a ‘decrease in the quantity demanded’.
As noted above, when we derive a demand curve we assume that any other relevant
factors (other than the price of the good itself and the resulting quantity demanded)
are held constant (ceteris paribus). These factors include the income, tastes, expectation
P
p1
p2
D
q
q1 q2
FIGURE 6.2 An individual consumer’s demand curve is given by his or her marginal benefit curve.
A movement along the demand curve is known as a ‘change in the quantity demanded’
D2
D1
q
FIGURE 6.3 A movement of the demand curve itself is called a ‘change in demand’
and the prices of other related goods. If any of these factors change, the demand curve
itself will shift (as in Figure 6.3), with the shift from D1 to D2. The shift of the demand
curve itself is called a change in demand. When the demand curve shifts to the right
(from D1 to D2), this is called an ‘increase in demand’; when there is a shift to the
left (from D2 to D1), this is called a ‘decrease in demand’.
6.4 Market demand
Now that we know that an individual consumer’s demand curve is given by his MB
curve, we can use this to derive the market demand curve. The market demand curve
traces out combinations of (a) market price and (b) quantities that all consumers in a
market are together willing and able to buy at that price. For example, suppose the
market price of apples is $4, and that there are just two consumers in the market. At
this price, Sonia is willing to buy six apples and Elizabeth is willing to buy three apples.
This means that, at $4, the total quantity demanded in the market is nine apples.
Hence, the market demand curve can be derived by adding together the quantity
demanded by each individual consumer at each price. This means that, in our apple
example, we would also need to check how much Sonia and Elizabeth would together
be willing to buy when the price of apples is $2, $3, $5, $6, etc. This means that,
graphically, the market demand curve can be derived by horizontally summing
together the individual demand curves (that is, the individual MB curves) along the
q-axis. Figure 6.4 shows two examples of deriving a market demand curve from
individual demand curves.2
As you can see, the law of demand also holds for the market demand curve. We can
also use the term change in the quantity demanded to refer to shifts along the
market demand curve, and the term change in demand to refer to a shift of the
demand curve itself.
Demand 51
P P
D1 D2 DM D1 D2 DM
Q,q Q,q
q 1 q2 q1 + q2 q1 q2 q1 + q2 q2
FIGURE 6.4 The market demand curve (DM ) can be derived by summing horizontally the individual
demand curves (D1 and D2). In the first graph, both individual demand curves have the same P-intercept.
If this is the case, then for every price p, the quantity demanded by the market will be the sum of
individual consumer demand (that is, q1 + q2). In the second graph, the individual demand curves have
different P-intercepts. Above the price p–, the quantity demanded by individual 2 will be zero; therefore,
in this range, the market demand curve follows the individual demand curve for consumer 1. Below the
price p–, the quantity demanded by both consumers is positive; in this range, the quantity demanded by
the market will be the sum of individual consumer demand (that is, q1 + q2)
We have now derived the individual and market demand curves. A demand curve
answers the question ‘if the consumer faces a certain price, what quantity would they
buy?’, for a range of possible prices. It is only possible to answer this question if the
consumer is a price taker; the question does not make sense if the consumer can, by
his choices, affect the market price. In a competitive market, a demand curve can be
drawn because consumers are price takers.
Notes
1 A continuous marginal benefit curve implies that it is possible to consume partial units of
a good (e.g. how much petrol to buy). Where the good or service in question is consumable
only in whole units (e.g. t-shirts), the continuous marginal benefit curve is usually a close
enough approximation. An obvious exception to this is where the consumer would only
want to buy one unit (e.g. a house, a car or a holiday), in which case it would not be
appropriate to draw a downward-sloping marginal benefit curve.
2 Note, with two or relatively few consumers the market demand curve can have several
kinks in it. However, with a large number of consumers (as will be the case in a competitive
market), individuals will come into and out of the market at different prices, and changes
in the quantity demanded by one consumer will be counterbalanced by other consumers.
As a result, the market demand curve can usually be represented fairly accurately using a
continuous (smooth) function.

C H A P T E R
7
Production
and costs
7.1 Introduction
This chapter focuses on how firms operate. We begin defining what is the short
and long run for a firm’s production process; in the short run the firm has at
least one fixed input of production, whereas in the long run all inputs can be
adjusted if the firm wishes to. Second, we analyse the relationship between a
firm’s inputs and its outputs – that is, its production function – and how this
relationship can change over time. We then examine how a firm’s output is
related to its costs in the short run and in the long run. Finally, we consider
how a firm’s profit is calculated.
7.2 The short run and long run
A firm, using the available technology, converts inputs – labour, machines

(often called capital) and natural resources (typically called land) – into output
that is sold in the marketplace. Typically, a firm will require more than one
input to produce its final output. We define the short run and the long run of
a firm in relation to whether or not any of the factors of production are fixed
(by ‘fixed’, we mean that the level of that input used cannot be changed
regardless of the output produced).
The short run is the period of time during which at least one of the factors
of production is fixed. For example, a firm might require both a factory and
workers to produce its output. The firm might have a lease on the factory that
must be paid, regardless of how much output the firm produces. In that case,
the factory is a fixed input until the lease ends. In the long run, all factors of
production are variable (that is, not fixed). Therefore, when the firm’s lease
of the factory ends, the firm is free to decide whether or not to renew the lease for that
factory.
Note that, in this context, the short run and the long run is not defined in relation
to a set period of time, but rather in relation to how long it takes for all of a firm’s
inputs to become variable. For some industries the short run will be quite short (for
example, a relatively simple business like a catering company). For other industries
the short run could be many years (for example, pharmaceuticals or the manufacture
of aircraft).
7.3 Production
A firm requires inputs or factors of production (labour, capital, land, etc.) in order to
produce its final output (i.e. goods or services). A production function shows the
relationship between quantity of inputs used and the (maximum) quantity of output
produced, given the state of technology. For example, suppose Jonathan owns a factory
that makes umbrellas. For the meantime, let us assume that Jonathan cannot increase
or decrease the size of his factory – that is, we are in the short run. He can, however,
adjust the amount of labour input he uses. If he employs one worker, he can make 60
umbrellas; with two workers, he can make 110 umbrellas; three workers, 150 umbrellas;
four workers, 180 umbrellas. The relationship between inputs (number of workers) and
output (number of umbrellas) comprises the production function. Sometimes a
production function can be represented using an equation. For example, suppose the
level of output of a good, q, depends on the amount of labour used, L. The production
function can be represented by q = f (L). A typical short-run production function looks
like the one in Figure 7.1.
f (L)
FIGURE 7.1 A typical short-run production function, where L represents the amount of labour employed
and q (L) represents the level of output
Production and costs 55
7.3.1 Marginal product

Once we have a production function, we might be interested in how output changes
as we change the quantity of only one of the inputs. The marginal product (MP) of
an input refers to how output responds when there is an increase in the number of that
input used. In the example above, hiring one worker (rather than having no workers
at all) allows 60 umbrellas to be made rather than 0 – the MP of the first worker is 60.
If Jonathan has one worker and hires one additional worker, his output will increase
from 60 to 110; therefore, the MP of the second worker is 110 – 60 = 50. If Jonathan
has three workers and hires one additional worker, his output will increase from 150
to 180; therefore, the MP of the fourth worker is 30 umbrellas.
If we are given the production function as an equation, we can use differentiation
to find the marginal product of an input. Suppose the production function takes the
form q = f (L), where q is the quantity of output and L is the quantity of labour used
as an input. In this case, we can find the marginal product of labour by differentiation
the production function with respect to L:
⌬q dq
MP = = (7.1)
⌬L dL
We might also be interested in how the MP of an input changes as we increase the use
of that input. If the MP becomes progressively smaller, this is called diminishing
marginal product. If the MP becomes larger, this is called increasing marginal
product. In the example above concerning Jonathan’s umbrellas, the marginal products
of the second, third and fourth workers respectively are 50, 40 and 30, indicating
diminishing marginal product; that is, each additional worker contributes less to output
than the worker before.
Again, if we are given the production function as an equation, we can use differen-
tiation to work out how MP changes as we increase the quantity of an input. So, if we
have the production function q = f (L) and we have found the marginal product, we
can differentiate the marginal product with respect to L:
⌬MP dMP
MP′ = = (7.2)
⌬L dL
If MP′ is positive (resp. negative), then we have increasing (resp. diminishing) mar-
ginal product.
Example. Shirley owns a hairdressing salon and employs a number of hairdressers.

The number of haircuts she can produce in one day is given by the production
function q = K1/2L1/2, where Y is the number of haircuts, L is the quantity of labour
and K is the amount of capital employed. Differentiating the production function
with respect to L gives
dq 1 1/ 2 −3/ 2
MP = = K L
dL 4
which is the marginal product of labour. Differentiating the marginal product

gives
dq 1 1/ 2 −3/ 2
MP = = K L
dL 4
which tells us that the marginal product of labour is diminishing – or, in other
words, there are diminishing returns to labour.
Diminishing MP is thought to be very common. First, in the short run there is a

fixed input of some kind which creates a capacity constraint; this will mean that each
additional workers will contribute to output less and less than those first hired. Consider
Jonathan’s umbrella factory. Because workers have to share factory space and
machinery, each additional worker will be able to increase output by less and less.
Note that this implies that diminishing MP is a short-run phenomenon; the capacity
constraint that drives the phenomenon will only exist if at least one output is fixed.
Also note that diminishing MP does not imply or require that the additional workers
are worse than those initially hired (in fact, we assume that every worker is identical);
rather, diminishing MP arises due to the capacity constraint due to the fixed input. Note
here, as at least one factor of production is fixed (the factory), diminishing MP is a
short-run concept.1
7.3.2 Returns to scale

Now let us allow all inputs into the production process to be variable. In our umbrella
manufacturing example, Jonathan can now vary all inputs in production process; he
can choose the factory size as well as the amount of labour utilized. Given all factors
of production are variable, we are in the long run.
With this choice available in the long run, we might also be interested in how the
quantity of output responds when we change the quantity of all of the factors of
production. Returns to scale refers to how the quantity of output changes when there
is a proportional change in the quantity of all inputs. If output increases by the same
proportional change, there are constant returns to scale – for example, if we double
the quantity of all the inputs and output also doubles in quantity. If output increases
by more than proportional increase in all inputs, we have increasing returns to scale.
If output increases by less than the proportional increase in all inputs, there are
decreasing returns to scale.
Example. Consider the production function q = KL, where q is output, and K and
L are capital and labour respectively. Notice that if we double the quantity of K
and L, output increases by a factor of 4. Therefore, we have increasing returns to
scale.
Note, when we are examining the returns to scale properties of a firm, all inputs are
variable; returns to scale is a long-run concept.2
7.4 Short-run costs
In order to use the inputs of production and transform them into outputs, a firm will
have to incur some costs. These costs include wages paid to workers and the cost of
leasing or buying factories and machines. A cost function is an equation that links
the quantity of output with its associated production cost (i.e. TC = f(q), where TC
represents total cost and q represent the quantity of output). Figure 7.2 graphically
represents a typical cost function (or ‘cost curve’), with output on the x-axis and total
cost on the y-axis.
TC
f(q)
FIGURE 7.2 A typical total cost function, where q represents the level of output and TC represents
total cost
Several things should be noted about the curve in Figure 7.2. First, when output is
zero, total cost is positive; this is because, in the short run, some factors of production
are fixed and the firm must pay the costs of these factors regardless of the amount of
output produced. Second, the total cost curve rises as output increases; this represents
the increase in cost as greater quantities of variable inputs are used. Third, the curve
rises at an increasing rate; this captures diminishing MP, as a greater quantity of inputs
is needed to increase output by the same amount as output goes up.
7.4.1 Fixed and variable costs

In the short run, some inputs will be fixed and some inputs will be variable; as a
consequence, a firm will have some fixed costs and some variable costs. Fixed costs
(FC) are costs that do not vary with quantity of output produced. So, when a firm’s
output is zero, all the costs it incurs will be fixed costs:
FC = TC when q = 0
By contrast, variable costs (VC) are those costs that vary with or depend on the
quantity of output produced. All costs that are not fixed costs will be variable costs.
VC = TC – FC
Example. Consider the cost function TC = 100 + 20q + q2, where q is the level
of output produced. In this case, fixed costs are FC = 100 (the cost function
evaluated at q = 0 and variable costs are VC = 20q + q2.
7.4.2 Marginal cost

Sometimes, a firm will be interested in its marginal cost (MC) – that is, the increase
in total cost that arises from an extra unit of production:
⌬TC ⌬VC
MC = =
⌬q ⌬q
When we have total cost expressed as a continuous function, the marginal cost can be
calculated by taking the first derivative of the total cost function with respect to q:
dTC dVC
MC = =
dq dq
Due to diminishing MP, a typical MC curve will eventually be increasing in output;
that is, MC has a positive slope. Returning to our umbrella example above, as each
worker costs the same to hire but produces progressively less than the previous hire
(diminishing MP), the extra cost of producing another unit of output (MC) must go
up. Consequently, in the short run diminishing MP implies increasing MC. This is
illustrated in Figure 7.3.
7.4.3 Average costs

A firm may also be interested in its average costs – that is, its costs per unit. These
costs can be determined as follows, and the typical shape of these curves is illustrated
in Figure 7.3.
• Average fixed cost (AFC) is fixed cost per unit of output:
FC
AFC =
q
Note that the AFC curve is always downward-sloping; the numerator is fixed, but
the denominator increases with output.
• Average variable cost (AVC) is variable cost per unit of output:
VC
AVC =
q
Because AVC is affected by diminishing MP, the AVC curve will eventually be
upward-sloping over output.
• Average total cost (ATC) is total cost per unit of output:
TC
ATC =
q
Note that ATC = AVC + AFC. Therefore, the shape of the ATC curve is determined
by the shape of the AFC curve (which is always declining) and the AVC curve
(which can be initially declining, but will eventually be increasing in output). At
very low levels of output, it is usually the decline in AFC that dominates, but at
higher levels of output, it will be the effect of the upward-sloping AVC that
dominates. Together, this will give the ATC curve a U-shape (i.e. initially decreas-
ing, but eventually increasing with output).
The relationship between MC, ATC and AVC is also important. As a rule, the MC curve
passes through the minimum of ATC and AVC. To see why this must be the case,
consider the following analogy. Suppose a student has an average exam score of 60
per cent, and he sits another exam. We can think of the score from the additional exam
as the marginal exam score. How will the marginal exam score affect his average exam
score? If the student scores above 60 per cent, his average score will increase; if he
scores below 60 per cent, his average score will decrease. By the same logic, if the
marginal cost of a unit of output is higher (resp. lower) than average total cost, it will
have the effect of increasing (resp. decreasing) average total cost. Thus, so long as the
MC curve lies below the ATC curve, it will drag that curve downwards; so long as MC
is above ATC, it will pull that curve upwards. Therefore, as ATC is decreasing when
MC is below it, and increasing when MC is above it, ATC must be at its minimum
when it is intersected by MC. For the same reasons, MC intersects AVC at its minimum.
This is illustrated in Figure 7.3; when drawing cost curves it is important to illustrate
this relationship correctly.
7.5 Long-run costs
7.5.1 Long-run marginal cost

In the long run, all inputs are variable. This implies that all costs are variable in the
long run. Consequently, the marginal cost of increasing output by one unit must take
into account the fact that all inputs can be varied to achieve this increase. Thus, for
Costs
MC
ATC
AVC
AFC
q
FIGURE 7.3 The typical shape of the average total cost curve, the average fixed cost curve, the average
variable cost curve and the marginal cost curve. Note, the MC curve intersects the ATC curve from below
at its minimum. The MC curve also intersects the AVC curves from below the minimum of AVC
the same level of output, long-run marginal cost will be less than or equal to short-run
marginal cost; the extra flexibility in relation to inputs in the long run means that a
firm might be able to increase its output at a lower cost than in the short run. For
example, suppose the most cost-effective way for a car manufacturer to increase its
output is to buy more machinery, but in the short run only labour is variable. Therefore,
if the car manufacturer wishes to increase output in the short run, it must hire more
workers; however, in the long run, it can buy more machinery, which gives a lower
marginal cost in the long run than in the short run.
7.5.2 Long-run average cost

For the same reason long-run average cost can be no greater than short-run average
cost.3 For example, consider a firm that can choose the size of its factory in the long
run. That firm, anticipating a certain output requirement, will choose the factory size
that gives it the lowest average cost for that level of output. If the required output was
smaller or larger, the firm might choose a different factory size to again minimize
average cost. As a result of this, the long-run average cost curve will be the lower
envelope of all of the short-run average cost curves. Figure 7.4 illustrates how to derive
a long-run average cost curve from several short-run average cost curves.
The term economies of scale refers to cost advantages that a firm obtains from
increasing its output. Specifically, if long-run average costs are decreasing with output,
this is called economies of scale. On the other hand, if long-run average costs are
increasing with output, this is known as diseconomies of scale. If long-run average
costs are constant as output expands, by definition there are constant average
costs.4 These three cases are illustrated in Figure 7.5.
Costs
SRAC1
SRAC2
SRAC3
LRAC
FIGURE 7.4 The long-run average cost curve can be obtained by taking the lower envelope of all short-
run average cost curves. In this diagram, the short-run average cost curves are represented by the grey
curves. The long-run average cost curve is the broken black curve that runs beneath them. As more
short-run average cost curves are drawn in, the long-run average cost curve will become smoother
7.5.3 A final note

Even though we have allowed all inputs to be variable in the long run, we have also
implicitly assumed that several things remain constant. First, we have assumed that
input prices are fixed. Second, we have assumed that there have been no changes in
the state of technology that would affect a firm’s ability to produce output. Of course,
in the real world these changes are typical and important, but we have abstracted from
such changes in order to better understand a firm’s costs in both the long run and the
short run.
Costs
LRAC
Constant
average costs
Economies Diseconomies
of scale of scale
FIGURE 7.5 When the LRAC curve is downward-sloping, the firm is experiencing economies of scale;
when the LRAC curve is upward-sloping, there are diseconomies of scale. When the LRAC curve is flat,
there are constant average costs
7.6 Total revenue, total cost and economic profit
In order to derive a firm’s economic profit, we will need to define its total revenue and
total cost. Total revenue (TR) is the amount a firm receives for the sale of its output.
This will be the price at which the firm sells each unit, multiplied by the quantity of
units sold:
TR ⫽ P ⫻ Q (7.3)
Total costs (TC) refers to the economic costs that a firm incurs for producing output,
as we have discussed in this chapter. When we refer to economic costs, we include all
opportunity costs. Recall from Chapter 1 that this includes both the explicit and
implicit costs of production. In this context, an example of an explicit cost is the cost
of buying the inputs of production; an example of an implicit cost might be forgone
earnings (see the following example).
Example. William uses $300,000 of his savings to invest in a new business venture.
If he had kept those savings in his bank account, he would receive interest of 5
per cent per year. Therefore, by investing his savings in the business venture,
William forgoes those interest payments and they are part of the implicit cost of
investing in the business venture.
Economic Profit (␲ ) is the amount that a firm makes, net of the costs that it incurs.
That is:
␲ ⫽ TR ⫺ TC (7.4)
One thing to note is that we have used opportunity costs, not just explicit (accounting)
costs to calculate profit. An important implication of this is that economic profit may,
and typically will, differ from accounting profit – and this affects the way that we
interpret economic profit. For example, if a firm earns zero profit, this means that total
revenue only just covers total cost (the opportunity cost) – in other words, the firm
could earn the same net benefit from undertaking the next best opportunity. It does not
mean that the firm has received no accounting profit. Zero economic profit means that
a person has no incentive to switch from their first choice to the next best alternative.
Example. Abby makes some cookies at a cost of $20 (including the cost of labour)
and sells those cookies for $25. Alternatively, she could have instead put her initial
$20 into a savings account, which would have earned $5 in interest. In this case,
Abby’s accounting profit is $25 – $20 = $5. However, her economic profit must
include the implicit cost of $5 in forgone interest payments – that is, her economic
profit is $25 – $(20 + 5) = $0. Notice that Abby has no incentive to switch from
making cookies to putting her money in the savings account.
To have something to sell in the marketplace, a firm converts inputs into outputs. We
can represent the relationship between inputs and outputs using a production function.
In the short run, at least one input is fixed (like a factory, the size of a restaurant and
its decor, the computer system used by a firm). With at least one fixed input, eventually,
the additional output generated from using an extra input will decline – that is, there
will be diminishing marginal product. This is an important concept as it is the
underlining cause as to why the short-run marginal cost curve is upward sloping.
In the long run, a firm is able to vary all of its inputs (there are no fixed inputs or
fixed costs). This means that in the long run cost associated with particular level of
production should not be more than a firm’s short- run cost for the same level of output.
For example, a firm’s long-run average total cost curve is the lower envelope of the
short-run average total cost curves. In the long run, if a firm’s long-run average cost
curve is downward sloping with output, the firm experiences economics of scale. If
long-run total costs are rising with output there are diseconomies of scale. Finally, a
horizontal long-run average total cost curve indicates that there are constant average
costs.
Notes
1 Short run (and long run) are explicitly in Section 7.2.
2 Moreover, increasing returns to scale is not necessarily incompatible with the short-run
notion of diminishing MP.
3 Note that, in the long run, there is only one average cost. We do not need to separate average
costs into average fixed cost and average variable cost for the simple reason that there are
no fixed costs in the long run.
4 The shape of the long-run average cost curve is determined by returns to scale (discussed
earlier in section 7.3.2). That is, if we have increasing returns to scale (whereby output
more than doubles when all inputs are doubled) this will have the effect of creating
economies of scale (that is, declining long-run average cost). Similarly, if there are
decreasing returns to scale, we will have diseconomies of scale; if there are constant returns
to scale, we will have constant average costs. However, it is important not to conflate
‘economies of scale’ with ‘returns to scale’. The former refers to the relationship between
output and cost; the latter refers to the relationship between inputs and output.

C H A P T E R
8
Supply
8.1 Introduction
In the last chapter we examined a firm’s costs and production. We now use
those concepts to derive an individual firm’s supply function and the market
supply function. In this chapter, we will focus on competitive markets, in
which there are many buyers and sellers, such that no individual buyer or seller
has the power to materially affect the price in the market. As a consequence,
both sellers and buyers in the market are price takers.
8.2 Firm supply
Firm supply is defined as the quantity of output a firm is willing and able to
supply at a certain price. The supply curve traces out all combinations of (a)
market price and (b) quantities that a firm is willing and able to sell at that
price.
Given a certain market price, how much should a firm sell? As we shall
see, a firm should sell up until P = MC. To see why this is true, first note that
the marginal revenue for each unit that the firm sells is the price, P. Now,
consider a firm that sells a quantity such that P > MC for the last unit sold
(and this is true for at least one additional unit). If that firm increases its output
by one unit, it will increase its profit since the additional revenue from selling
that extra unit (P) outweighs the cost of producing that same unit (MC). For
example, if the price of a good is $10 but the marginal cost is $4, the firm
will receive an extra $6 of profit from selling that unit. Next consider a
firm producing where P < MC for the last unit made. If this is the case,
the firm should not have produced and sold its last unit. This is because the
last unit cost the firm more to produce than the price that the firm received
MC
p2
p1
q
q1 q2
FIGURE 8.1 An individual firm’s supply curve is given by its marginal cost curve. A shift along the supply
curve is known as a ‘change in the quantity supplied’
for it; therefore, the firm lowered its profit by producing that unit. For example, if the
price of a good is $10 but the marginal cost is $14, the firm would receive an extra
$4 of profit by not producing and selling that unit of the good.
Generally speaking, this means that a firm should sell an extra unit if P > MC, but
sell one fewer unit if P < MC; thus, at the end of the day, a firm should sell up until
P = MC. Figure 8.1 graphically represents this result. Suppose the initial market
price is represented by p1; in this case, profit is maximized if the competitive firm
sells up to the quantity where p1 = MC (that is, q1). If price, for whatever reason, rises
to p2, the firm will maximize profit by again selling up to the new quantity where
p2 = MC (that is, q2).
What this means is that a firm’s supply curve is given by its MC curve. Because
the MC curve is upward sloping due to diminishing marginal product, there is a positive
relationship between the price of a good and the quantity of that good supplied. This
positive relationship is known as the law of supply.
The firm’s supply curve is derived by assuming that only the price and quantity
supplied of the product can change. If there is a change in the price, there will be a
movement along the supply curve itself (as in Figure 8.1, with the shift from
(q1, p1) to (q2, p2 )), which is called a change in the quantity supplied. If there is
a movement up along the supply curve (that is, from (q1, p1) to (q2, p2)), this is called
an ‘increase in the quantity supplied’. If there is a movement left along the supply
curve (that is, from (q2, p2) to (q1, p1)), this is called a ‘decrease in the quantity
supplied’.
As noted, when we derive a supply function we assume that all other relevant factors,
other than the own price and quantity of the, are held constant (ceteris paribus). These
factors include the cost of inputs, technology and expectations about the future. If any
of these factors change, the supply curve itself will shift (as in Figure 8.2, with the
Supply 67
A change in supply
P
S1
S2
FIGURE 8.2 A movement of a firm’s supply curve itself is called a ‘change in supply’
shift from S1 to S2), which is called a change in supply. When the supply curve shifts
to the right (from S1 to S2), this is called an ‘increase in supply’; when there is a shift
to the left (from S2 to S1), this is called a ‘decrease in supply’.
One caveat is worth noting. When we discuss short-run and long-run supply we will
need to be careful to check whether the firm will want to produce anything at all (that
is q > 0). So far, we have simply determined that if a firm chooses to supply a positive
quantity, they should supply up until the point where P = MC.
8.3 Market supply
Given that an individual firm’s supply curve is given by its MC curve, we can use this
to derive the market supply curve. The market supply curve traces out combinations
of (a) market price and (b) quantities that all firms in a market are together willing and
able to sell at that price. For example, suppose the market price of carrots is $1. At
this price, Jackson is willing to sell five carrots and Jared is willing to sell eight carrots.
This means that, at $1, the total quantity supplied in the market is 13 carrots.
Simply put, the market supply curve can be derived by adding together the quantity
supplied by each individual firm at every price. This means that, in our carrot example,
we would also need to check how much Jackson and Jared would together be willing
to supply when the price of carrots is $2, $3, $4, etc. This means that, graphically, the
market supply curve can be derived by summing together the individual supply curves
(that is, the individual MC curves) horizontally along the q-axis. Figure 8.3 shows
two examples of deriving a market supply curve from individual supply curves.
The law of supply also holds for the market supply curve. We can also use the term
‘change in the quantity supplied’ to refer to movements along the market supply curve,
and the term ‘change in supply’ to refer to a shift of the supply curve itself.
P P
S1 S2 SM S1 S2 SM
p p
Q, q Q, q
q1 q2 q1 + q 2 q2 q1 q2 q1 + q2 q2
FIGURE 8.3 The market supply curve (SM) can be derived by summing horizontally the individual supply
curves (S1 and S2). In the first graph, both individual supply curves have the same P-intercept. If this is
the case, then for every price p, the quantity supplied by the market will be the sum of individual firm
supply (that is, q1 + q2). In the second graph, the individual supply curves have different P-intercepts.
Below the price p–, the quantity supplied by firm 2 will be zero; therefore, in this range, the market
supply curve follows the individual supply curve for firm 1. Above the price p–, the quantity supplied by
both firms is positive; in this range, the quantity supplied by the market will be the sum of individual firm
supply (that is, q1 + q2)
In this chapter, we derived the individual and market supply curves. A supply curve
answers the question ‘if the firm faces a certain price, what quantity would they want
to supply?’, for a range of possible prices. It is only possible to answer this question
if the firm is a price taker; the question does not make sense if the firm can choose the
market price. We can draw a supply curve in the context of a competitive market,
because firms in that environment are price takers.
C H A P T E R
9
Equilibrium and
welfare
9.1 Introduction
In Chapters 6 and 8, we derived the demand and supply curves. Together,

demand and supply determine the price and quantity traded of a good or service
in a market. These market outcomes are the focus of this chapter.
9.2 Market equilibrium
A market is in equilibrium if, at the market price, the quantity demanded by

consumers equals the quantity supplied by firms in the market. The price at
which this occurs is called the market-clearing price (or ‘equilibrium
price’). Figure 9.1 depicts a market in equilibrium at point A. At the
equilibrium price (P*), quantity demanded is equal to quantity supplied (qd =
qs = q*). This point is called an equilibrium because there is no pressure on
price or quantity traded in the market to change. As we shall see, if a market
is not in equilibrium, there will be pressure on price and quantity to move
towards the equilibrium price and equilibrium quantity.
First consider the case where the market price (P1) is above the equilibrium
price, as seen in Figure 9.2. As a consequence, the quantity supplied (Q 1s )
exceeds the quantity demanded (Q 1d ). This difference is called excess supply,
which means that sellers cannot find buyers for all units supplied to the
market. In these circumstances, there will be downward pressure on prices,
as sellers try to bring more consumers into the market; at the same time, the
quantity supplied will fall in response to the decrease in prices. This downward
pressure on prices will continue until the excess of supply is eliminated,
moving the market towards equilibrium.
p* A
D
Q
Q*
FIGURE 9.1 A market in equilibrium. The equilibrium price is P * and the equilibrium quantity
traded is Q *
Excess supply
P1
D
Q
Q1d Q1s
FIGURE 9.2 When market price is above the equilibrium price, there is an excess of supply
in the market
Next consider the case where the market price (P2) is below the equilibrium price,
as seen in Figure 9.3. Here, the quantity demanded (Q 2d ) exceeds the quantity supplied
(Q 2s ), which is called excess demand. This means that sellers do not supply enough
units to meet consumer demand. Now, there will be upward pressure on prices, as
buyers compete for limited units in the market; this increase in prices will increase the
quantity supplied and also decrease quantity demanded. This upward pressure on prices
will continue until the excess of demand is eliminated, moving the market towards
equilibrium.
Equilibrium and welfare 71
p2
Excess demand
D
Q
Q2s Q2d
FIGURE 9.3 When market price is below the equilibrium price, there is an excess of demand
in the market
How quickly does the market move towards equilibrium? Evidence from experi-
ments and real-world markets suggests that, if there are many buyers and sellers, the
adjustment process to the market-clearing price is quite fast.1
9.3 Comparative static analysis
Sometimes, markets are affected by a change or event beyond the direct control of
buyers or sellers in that market. In such cases, we may want to analyse how that change
or event affects the choices of firms and/or consumers in the market, and how those
choices affect market outcomes. For example, we may want to know what will happen
in the market for corn chips if there is a drought, what is the impact on the market for
cars if the cost of steel falls, or how a government campaign promoting healthy living
will affect the market for yoga classes. This type of analysis is often called
comparative static analysis, and essentially involves an examination of how the
market equilibrium is affected by the change or event – that is, a comparison of the
old and the new market equilibria.
To conduct comparative static analysis, first assume that the market in question is
initially in equilibrium. Next, ascertain whether the change or event will affect the
demand curve or the supply curve of the market (or both), and if so how that curve
will shift. Finally, use the demand and supply diagram to analyse the impact of the
change or event; that is, compare prices and quantities traded in the market before and
after the change.
Example. Consider how the market for corn chips, depicted in Figure 9.4, is
affected by an increase in the price of potato chips. The demand for corn chips
S1
P2*
P1*
D1 D2
Q
Q1* Q2*
FIGURE 9.4 The market for corn chips when there is an increase in the price of potato chips
is denoted by D1 and the supply for corn chips is denoted by S1. The market is
initially in equilibrium at (Q1*, P1*). Assuming corn chips and potato chips are
substitutes, an increase in the price of potato chips is likely to increase the demand
for corn chips, shifting the demand curve to the right, to D2. This will cause an
increase in the quantity supplied, as we move along the supply curve S1 to the
new equilibrium. The new equilibrium price and quantity (Q2*, P2*) are higher than
before the change in demand.
Example. Consider how the market for cars, depicted in Figure 9.5, is affected
by an increase in the price of steel. The demand for cars is denoted by D1 and the
supply for cars is denoted by S1. The market is initially in equilibrium at (Q1*, P1*).
Assuming that steel is an input in the production of cars, the MC of making each
P
S2
S1
P2*
P1*
D1
Q
Q2* Q1*
FIGURE 9.5 The market for cars when there is an increase in the price of steel
car will increase, causing the supply curve to shift upwards (or to the left) to S2.
This will cause a decrease in the quantity demanded, as we move along the demand
curve D1 to the new equilibrium. At the new equilibrium (Q2*, P2*), price of cars
is higher but quantity traded is lower than before the change in the price of steel.
As you can see, such demand-and-supply analysis is a relatively simple yet

extremely powerful tool. This structured method of analysis gives us a formal way of
thinking about changes in markets and supporting our intuition conclusions about how
markets work.
9.4 Welfare
Markets are one of the main ways that good and services are produced and distributed.
Of course, consumers and firms will only participate in markets if it is beneficial to
them. We can measure and observe changes in these benefits to these participants using
welfare analysis. In this section, we examine the welfare of consumers and firms.
9.4.1 Consumer surplus

Consumer surplus (CS) is the welfare consumers receive from buying units of a
good or service in the market. We can measure consumer surplus by evaluating the
net value of a good or service to the consumer, as he or she perceives it. That is,
consumer surplus is given by the consumer’s willingness to pay, minus the price
actually paid, for each unit bought.
To provide some intuition for this, suppose Hamish is considering buying a chocolate
bar. His willingness to pay (or marginal benefit) for the chocolate bar is $5.50, but the
price of chocolate bars is $2. If he buys the chocolate bar, he will receive $5.50 of
benefits, minus the price actually paid of $2. Therefore, his surplus from buying the
chocolate bar is $3.50. However, consumer surplus in the market takes into account
every unit of the good or service purchased. Therefore, if Hamish buys multiple
chocolate bars, we would need to add up the surplus from each chocolate bar in order
to get his total consumer surplus.
Recalling that the individual demand curve traces out a consumer’s marginal benefit
or willingness to pay, we can find an individual’s CS by calculating the area between
the individual demand curve and the price line. Similarly, we can find the CS of all
consumers in the market by calculating the area between the market demand curve
and the price line. This area is shown in Figure 9.6.
What happens when price falls? Consider Figure 9.7, in which the market price falls
from P1 to P2. As you can see, the area of consumer surplus has now increased, due
to two additional benefits: on all the units previously consumed, the difference between
MB and price is now larger, meaning that the net benefit from consuming each of these
units has increased; secondly, the lower price now means that more units are purchased,
which also generates an additional benefit to consumers.
CS
P*
D
Q
Q*
FIGURE 9.6 The area of consumer surplus in this market is denoted by the shaded area
9.4.2 Producer surplus

On the other side of the market, producer surplus (PS) is the welfare producers (that
is, firms) receive from selling units of a good or service in the market. Producer surplus
can be measured by considering the net benefit of selling a good or service. That is,
producer surplus is given by the price the producer receives, minus the cost of
production, for each unit of the good or service bought.
Now suppose that Adam is a producer of chocolate bars. As before, the price of
chocolate bars is $2, but the marginal cost to Adam of producing the chocolate bar is
P1
B
C
P2
D
Q
Q1 Q2
FIGURE 9.7 When the market price falls from p1 to p2, the area of consumer surplus increases from A to
A + B + C. The area B represents the increase in consumer surplus that arises from an increase in the
net benefit of previously consumed units. The area C represents the increase in consumer surplus arising
from the consumption of additional units
P*
PS
D
Q
Q*
FIGURE 9.8 The area of producer surplus in this market is denoted by the shaded area
$0.50. Therefore, if Adam sells the chocolate bar, his net benefit is the $2 received,
minus $0.50 in production costs. Therefore, his surplus from selling the chocolate bar
is $1.50. However, producer surplus in the market takes into account every unit of the
good or service sold. Therefore, if Adam sells multiple chocolate bars, we would need
to add up the surplus from each chocolate bar in order to get her total producer surplus.
Remembering that a firm’s supply curve is given by its MC curve, a firm’s PS
can be found by calculating the area between the price line and the firm’s supply
curve. Similarly, we can find the PS of all producers in the market by calculating the
P2
C
B
P1
Q
Q1 Q2
FIGURE 9.9 When the market price increases from P1 to P2, the area of producer surplus increases from
A to A + B + C. The area B represents the increase in producer surplus that arises from an increase in
the net benefit of selling units that would have been sold previously. The area C represents the increase
in producer surplus arising from the sale of additional units
area between the price line and the market supply curve. These areas are shown in
Figure 9.8.
Let us again consider what happens when there is a change in price. Suppose, as in
Figure 9.9, there is an increase in price from P1 to P2. PS increases for two reasons:
on all the units previously sold, the price is now higher, meaning that the net benefit
from selling each of these units has increased; second, the higher price now means
that more units are sold, which also generates some net benefit to producers.
9.4.3 Total surplus

We can also measure the total welfare of all participants in the market. Here, there are
only two types of participants: consumers and producers. In later chapters, we will
allow for other participants in the market (namely, the government). For now, however,
total surplus (TS) is the sum of consumer surplus and producer surplus in the market
equilibrium:
TS ⫽ CS ⫹ PS (9.1)
As CS is the area below the demand curve and above the price line, and PS is the area
below the price line and above the supply curve, it follows that TS is the area between
the demand and supply curves, up to the market equilibrium quantity Q*. This is
illustrated in Figure 9.10.
TS
D
Q
Q*
FIGURE 9.10 The area of total surplus in this market is denoted by the shaded area
9.5 Pareto efficiency
To analyse welfare in a competitive market further, we now introduce the concept of

Pareto efficiency. An outcome is Pareto efficient if it is not possible to make
someone better off without making someone else worse off. Conversely, an outcome
is not Pareto efficient if it is possible to reallocate resources (or do things differently
in the market) and make someone better off without making someone else worse off.
Another way of thinking about Pareto efficiency in this context is that the Pareto
efficient outcome maximizes total surplus.
As it turns out, the outcome in a competitive market is Pareto efficient. Consider
the market equilibrium in Figure 9.1. For all the trades between 0 and Q*, the MB of
the good to the consumer was greater than the MC of production. Hence, the consumer
was willing to pay more than the minimum amount needed to induce the producer to
make and sell the good. Moreover, trading the good yields a net benefit to at least one
of the parties and no one was made worse off. For the last unit traded (the Q*th unit),
MB = MC, but neither the consumer nor the producer were made worse off.2
Suppose, for some reason, fewer than Q* units are traded in the market. At this
quantity, MB > MC. This outcome is not Pareto efficient, because it is possible to
increase the number of units traded in order to make the consumer and/or the producer
better off, without making anyone worse off.
Similarly, suppose more than Q* units are traded in the market, such that, for the
last unit MB < MC. In this case, all units traded beyond Q* made someone worse off:
either the buyer paid more than his MB, the seller received a price less than her MC,
or both. Therefore, this outcome is not Pareto efficient, because total surplus would
be higher if the trade of units in excess of Q* did not take place.
Indeed, at the competitive-market equilibrium, all the potential gains from trade are
exhausted; there are no consumers left in the market with a willingness to pay higher
than any sellers cost of providing an additional unit of output. Moreover, the market
through the price mechanism ensures that the people with the highest value for the
product (those that are willing to pay more than the price) end up with the goods, and
that those firms with the lowest cost are the ones who make the goods (the firms who
have a MC less than the market price). While these actions are completely decentralized,
in the sense that there is no one person coordinating the actions of the many parties
in the market, a competitive market manages to maximize total surplus (that is, reach
a Pareto efficient outcome).
It is important to note that Pareto efficiency has a very strict and specialized
definition. In particular, it does not imply either uniqueness or fairness/equity. That is
to say, in a given market (or economic situation), there could also be more than one
outcome that is Pareto efficient. Further, an outcome that is Pareto efficient is not
automatically the most fair or equitable, or even the most desirable. In order to
determine the ‘best’ market outcome, it may be necessary to weigh up efficiency against
other socially desirable objectives.
In this chapter, we defined and calculated consumer surplus, producer surplus and total
surplus. We found that the outcome in a competitive market has the following
characteristics: (i) via the price mechanism, it allocates goods to consumers who value
them most highly; and (ii) via the price mechanism, it allocates demand for goods to
sellers who can produce at the least cost. We also found that a competitive market
maximizes total surplus and, hence, is Pareto efficient. It follows that a person who
can dictate the price and quantity of a good traded in the market (a ‘social planner’)
cannot achieve an outcome that is more efficient than the free (competitive) market.
Notes
1 Experimental evidence comes from researchers setting up experiments that simulate
economic problems in a laboratory. Experiments have the advantage of being able to control
important variables, better replicating our assumption of ceteris paribus.
2 For convenience, although there is no net gain to either party, we just assume that this
trade goes ahead; this assumption does not affect our welfare analysis of the market.
C H A P T E R
10
Elasticity
10.1 Introduction
In this chapter, we are concerned with measuring how a change in one variable
affects another. For example, we might be interested in how the equilibrium
quantity traded responds to a change in the market price, or how demand
responds to changes in consumer income. One issue with measuring quanti-
tative changes is that different markets use different units of measurement
(litres, kilograms, ounces and so on), each market has its own price level
(a few cents or millions of dollars) and it is even possible that different markets
use different currencies. A way we can compare quantitative changes across
different situations is to look at proportional (or percentage) changes.1
10.2 Measuring elasticity
As discussed in Chapter 2, elasticity measures how responsive one variable

(y) is to changes in another variable (x). That is, when we increase x by a certain
amount, does y change by a small amount or by a large amount? We can
calculate elasticity (ε) by dividing the percentage change in y by the percentage
change in x:2
% ⌬y
ε= . (10.1)
% ⌬x
This tells us that, for 1 per cent change in x, there will be an ε per cent change
in y. As you can see, the larger the absolute value of ε, the more responsive
y is to changes in x; conversely, the smaller the absolute value of ε, the less
responsive y is to changes in x.
Generally, we can calculate the proportional change in a variable by dividing the

change in the variable by the variable itself; that is:
⌬x
(10.2)
x
However, it is not always obvious how to determine what the proportional change in
a particular variable is. For this reason, we have two methods of calculating elasticity:
the point method and the midpoint (arc) method. It will be appropriate to use the point
method for calculating elasticity when we are calculating elasticity at a single point.
The midpoint (arc) method will be appropriate when we are interested in elasticity
moving from one point to another. These methods are discussed in more detail below.
10.2.1 Point method

At times we are interested in elasticity at a particular point. For example, suppose when
the price of a good is P1, the quantity demanded of that good is Q1; what is elasticity
at the point (Q1, P1)? In these cases, we can use the point method of calculating
elasticity. This simply involves recognizing that, for very small changes in the variables,
⌬y/⌬x = dy/dx.
In general terms, the point method can be expressed as follows:
(⌬y ) / y ⌬y x dy x
ε= = ⋅ = ⋅ (10.3)
(⌬x ) / x ⌬x y dx y
Example. Suppose the demand curve for forks is given by Q = 100 – 2P, and the
price of forks is P = 30. What is the elasticity at this price? From the demand
equation, we know that when P = 30, Q = 40. The slope of this line is ⌬Q/⌬P
= –2. (This can also derived by the first derivative of the demand equation, which
gives us dQ/dP = –2.) Substituting these values into the point formula gives:
30
ε = −2 ⋅ = −1.5 (10.4)
40
The interpretation of this is: if price of forks increases by 1 per cent, the quantity
demanded of forks falls by 1.5 per cent.
10.2.2 Midpoint (or arc) method

Sometimes we are interested in elasticity when moving from one point to another. For
example, suppose the price of a good changes from P1 to P2, which causes the quantity
demanded to change from Q1 to Q2. Here, we are moving from one point (Q1, P1) to
another (Q2, P2). However, it is unclear in this situation whether we should measure
the change in price (resp. quantity) as a percentage of P1 or of P2 (resp. Q1 or Q2)?
Elasticity 81
To resolve this ambiguity, sometimes we adopt the midpoint (or arc) method; that
is, when calculating percentage changes, we use the midpoint (or the average) of P1
and P2, and the average of Q1 and Q2,
In general terms, the midpoint (or arc) method can be expressed as follows:
(⌬y ) / y m ⌬y x m (10.5)
ε= = ⋅
(⌬x ) / x m ⌬x y m
x1 + x2 y +y
where x m = and y m = 1 2 .
2 2
Example. When the price of spoons is $10, the quantity demanded is 50 units.
When the price increases to $20, the quantity demanded falls to 30 units. In order
to calculate elasticity, we need to find that average (or r) of price and quantity:
Pm = 15 and Qm = 40. We also need to find the change in price and quantity:
⌬P = 20 – 10 = 10 and ⌬Q = 30 – 50 = –20. Substituting these values into the
midpoint formula gives:
−20 15
ε= ⋅ = −0.75 (10.6)
10 40
The interpretation of this is: if price of spoons increases by 1 per cent, the quantity
demanded of spoons falls by 0.75 per cent.
10.3 Applications
The concept of elasticity has several important applications in economics. We will

discuss some of these now.
10.3.1 Elasticity of demand

One of the most important applications is the elasticity of demand (εd)3 It measures
how sensitive the quantity demanded of a good (Qd ) is to changes in price (P).
Specifically, it tells us the proportional change in quantity demanded of a good, given
a 1 per cent change in its price. By substituting Qd and P into the formulas above, we
can write down the midpoint (arc) method and the point method for calculating
elasticity of demand:
Midpoint method Point method

⌬Qd P m dQd P
εd = ⋅ εd = ⋅
⌬P Qdm dP Qd
Given the law of demand, the elasticity of demand will normally be negative (or at
least non-positive).4 For this reason, some authors find it convenient to drop the minus
sign when reporting the elasticity of demand, treating the negative sign as implicit.
We will not adopt that convention here, but it should be noted that either approach is
fine, so long as it is consistently applied.
Now that we can calculate elasticity of demand, let us consider what our results mean:
• If εd = 0, demand is perfectly inelastic. For a 1 per cent change in price, there

is no change in the quantity demanded; in other words, quantity demanded is not
at all responsive to changes in price. In this special case, the demand curve is
vertical, as shown in Figure 10.1.
• If εd < 0, demand is inelastic. For a 1 per cent change in price, the resulting change
in quantity demanded is less than 1 per cent; in other words, quantity demanded
is not very responsive to changes in price.
• If εd = –1, demand is unit elastic. For a 1 per cent change in price, there is a
1 per cent change in quantity demanded; in other words, the quantity demanded
changes by the same proportion as price.
• If εd < –1, demand is elastic. For a 1 per cent change in price, the change in
quantity demanded is more than 1 per cent; in other words, quantity demanded is
very responsive to changes in price.
• If εd = –∞, demand is perfectly elastic. For a small increase in price, quantity
demanded will drop to zero. Intuitively, this means that if a firm raises its price
at all, its customers will go elsewhere to buy the product. In this special case, the
demand curve is horizontal, as shown in Figure 10.1.
You can see, from the formulas above, that the elasticity of demand depends partly
on the slope of the demand curve (that is, ⌬Qd /⌬P or dQd /dP). However, elasticity of
Perfectly inelastic demand Perfectly elastic demand
P P
D1
D2
Q Q
FIGURE 10.1 When demand is perfectly inelastic, the demand curve is vertical (as seen the left panel).
When demand is perfectly elastic, the demand curve is horizontal (as seen the right panel)
Elasticity 83
demand also depends upon the values of P and Q at the reference point where elasticity
is being measured. For this reason, even when the demand curve is a straight line, the
elasticity of demand changes as we move along the curve.
Example. Consider the demand curve, Qd = 100 – P. Along the entire demand
curve, the slope is constant as dQd /dP = –1. However, as we move along the curve,
elasticity of demand changes. At the point (Qd, P) = (90, 10), elasticity is given
by ε = –1·(10/90) = –1/9 and the demand is inelastic. At (50, 50), ε = –1·(50/50)
= –1, and demand is unit elastic. At (25, 75), ε = –1·(75/25) = –3, and demand
is elastic.
The reason for this is as follows. Because the slope of the demand curve is constant,
the absolute change in quantity demanded is the same for a given change in price as
we move along the curve. Elasticity varies because the proportional change in quantity
(and price) varies depending on the size of quantity (or price) at a particular point. For
instance, when price is low and quantity demanded relatively high, a given change in
quantity is a smaller proportional change, whereas the change in price will be a larger
proportional change, given that price is relatively low.
Therefore, for every linear demand curve, there is an inelastic section (when quantity
is high and price is low); a point that is unit elastic (in the middle of the demand curve);
and an elastic section (when quantity is relatively low and price relatively high). In
fact, on any linear demand curve, the price elasticity of demand ranges from 0 (when
it cuts the Q-axis) to –∞ (when it cuts the P-axis). This is illustrated in Figure 10.2.
This means that care is needed when referring to an ‘elastic’ or an ‘inelastic’ demand
curve; whether a linear demand curve is elastic or inelastic depends upon where on
the demand curve we are.
Elastic
Unit elastic
P2
Inelastic
D
Q
Q2 Q
FIGURE 10.2 When the demand curve is linear, elasticity changes as we move along the curve. The
curve is most elastic near the P-axis and the most inelastic near the Q-axis
Elasticity and revenue

Another thing we can determine from the elasticity of demand is how total revenue in
the market will change as price changes. As we know from the demand curve, the
quantity demanded in the market (Q) depends upon the market price (P). This means
that we can write the quantity demanded as a function of price: Q(P). We also know
that the total revenue in the market is the number of units sold multiplied by the price
(that is, P × Q). Therefore, we can express total revenue as a function of P:
TR(P) = P · Q(P) (10.7)
The size of market revenue can be depicted in a diagram, as shown in Figure 10.3.
We can differentiate this equation with respect to P in order to determine how total
revenue changes in response to a small increase in price:
dTR dQ
=Q+P⋅ (10.8)
dP dP
By rearranging this equation, we can see the how the change in total revenue depends
upon the elasticity of demand:
dTR dQ ⎛ P dQ ⎞
=Q+P⋅ = Q ⎜1 + ⋅ ⎟ (10.9)
dP dP ⎝ Q dP ⎠
= Q(1 + ε d )
Equation 10.9 provides a direct link between the price elasticity of demand and the
change in total revenue. In order for TR to increase with price, the right-hand side
of the equation must be positive. Assuming that Q > 0, this will be true if and only if
Total revenue
D
Q
Q(P)
FIGURE 10.3 Total revenue can be calculated by multiplying price and quantity. In this diagram, the size
of total revenue is denoted by the shaded area
Elasticity 85
ε > –1; that is, if demand is inelastic. On the other hand, if demand is elastic (ε < –1),
TR will fall when the market price rises.
Consider this conclusion in light of Figure 10.2. On the elastic part of the demand
curve (that is, the upper part) the price needs to be lowered in order to increase TR.
On the inelastic part of the demand curve (that is, the lower part) the price needs to
be raised in order to increase TR. What this suggests is that TR is maximized when
demand is unit-elastic, in the middle of the demand curve. Indeed, we can confirm this
by noting that when ε = –1, dTR/dP = 0.
To provide some intuition for this result, recall that the elasticity of demand measures
how responsive the quantity demanded is to changes in price. If demand is elastic, a
1 per cent increase in price will cause a greater than 1 per cent fall in the quantity
demanded. This means that the increase in P is more than offset by the decrease in
Qd, causing TR to fall overall. Conversely, if demand is inelastic, a 1 per cent increase
in price will cause quantity demanded to fall but by less than 1 per cent. Thus, the
increase in P outweighs the decrease in Qd, causing TR to increase overall.
10.3.2 Elasticity of supply

The concept of elasticity can also be applied to supply. Elasticity of supply (εs)
measures how sensitive the quantity supplied of a good (Qs) is to changes in price (P).
That is, what is the proportional change in quantity supplied of a good, given a 1 per
cent change in its price. The midpoint (arc) method and the point method for calculating
elasticity of supply are as follows:

⌬Qs P m dQs P
εs = ⋅ εs = ⋅
⌬P Qsm dP Qs
The elasticity of supply is typically positive, due to the law of supply. Again, we can
characterize the values of elasticity of supply as follows:
• If εs = 0, supply is perfectly inelastic. For a 1 per cent change in price, there is

no change in the quantity supplied; in other words, quantity supplied is not at all
responsive to changes in price. In this special case, the supply curve is vertical,
as shown in Figure 10.4.
• If 0 < εs < 1, supply is inelastic. For a 1 per cent change in price, the change in
quantity supplied is less than 1 per cent; in other words, quantity supplied is not
very responsive to changes in price.
• If εs = 1, supply is unit elastic. For a 1 per cent change in price, there is a 1 per
cent change in quantity supplied; in other words, the quantity supplied changes
by the same proportion as price.
• If εs > 1, supply is elastic. For a 1 per cent change in price, the change in quantity
supplied is more than 1 per cent; in other words, quantity supplied is very
responsive to changes in price.
• If εs = –∞, supply is perfectly elastic. For a small decrease in price, quantity

supplied will drop to zero. Intuitively, this means that if the price of a good falls
below a certain price, firms will stop supplying the product. In this special case,
the supply curve is horizontal, as shown in Figure 10.4.
Perfectly inelastic supply Perfectly elastic supply
P P
S1
S2
Q Q
FIGURE 10.4 When supply is perfectly inelastic, the supply curve is vertical (as seen the left panel).
When supply is perfectly elastic, the supply curve is horizontal (as seen the right panel)
10.3.3 Cross-price elasticity

Sometimes we are interested in the relationship between the quantity demanded of one
good and the price of another related good. This relationship can be examined using
the cross-price elasticity.5 This measures how sensitive the quantity demanded of
a Good A (QA) is to changes in price of Good B (PB).
⌬QA PBm dQA PB

ε AB = ⋅ ε AB = ⋅
⌬PB QAm dPB QA
Example. Suppose that, when the price of teabags is $4 per box, Candice sells
100 litres of milk. If the price of teabags rises to $8 per box, Candice only sells
60 litres of milk. Let us use the midpoint formula to calculate cross-price elasticity.
Here, ⌬QA = –40 and ⌬PB = 4. The average values for price and quantity are
QAm = 80 and PBm = 6. Therefore, cross-price elasticity is εAB = (–40/4)·(6/80)
= –0.75.
Elasticity 87
Cross-price elasticity provides some information about the relationship between the
two products:
• If εAB > 0, an increase in the price of Good B is associated with a rise in the quantity
demanded of Good A. This means that the Good A and Good B are substitutes.
That is, they are goods that are likely to be consumed in place of each other, for
example tea and coffee.
• If εAB < 0, an increase in the price of Good B is associated with a fall in the quantity
demanded of Good A. This means that the Good A and Good B are complements.
That is, they are goods that are likely to be consumed together, for example bacon
and eggs.
• If εAB = 0, an increase in the price of Good B is not associated with any change
in the quantity demanded of Good A. This means that the Good A and Good B
are independent goods. That is, the two goods are completely unrelated, for
example ice cream and chainsaws.
10.3.4 Income elasticity

The demand for a good may also depend, in part, on a consumer’s income. Income
elasticity (␩)6 measures how sensitive the quantity demanded of a good (Q) is to
changes in income (Y). Applying the midpoint and point methods gives us:
⌬Q Y m dQ Y
␩= ⋅ ␩= ⋅
⌬Y Qm dY Q
We can characterize the good, depending upon its income elasticity:
• If ␩ < 0, demand for the good decreases when income rises. This type of good is
called an inferior good. An example of an inferior good might be offal; as income
increases, consumers can afford to eat superior cuts of meat.
• If ␩ = 0, demand for the good does not change when income rises. This type of
good is called a neutral good.
• If 0 < ␩ ⭐ 1, when income rises by 1 per cent, demand for the good increases by
less than 1 per cent. This type of good is called a normal good. An example of
a normal good might be food generally; as income increases, consumers will
consume more food, but the increase is likely to be proportionally less than the
increase in income.
• If ␩ > 1, when income rises by 1 per cent, demand for the good increases by more
than 1 per cent. This type of good is called a luxury good. An example of a luxury
good might be caviar; if income increases by 1 per cent, the consumption of caviar
will likely rise by more than 1 per cent.
Elasticity measures the proportional change in one variable, given a proportional change
in another variable. The concept can be applied to any two variables of interest. Most
commonly, economists will be interested in the elasticity of demand, the elasticity of
supply, cross-price elasticity and income elasticity. However, in some cases we might
be interested in other relationships. For example, a business might like to know how
the quantity demanded for its product changes in response to an advertising campaign;
policy makers might want to measure consumers’ response to a government subsidy.
The concept of elasticity is unit-free, in the sense that it is measured in proportions
and not units (such as pounds, milligrams, dollars, and so on). This allows comparison
across markets and between countries.
Finally, elasticity will depend on a range of factors, including the timeframe
considered. In general, we would expect elasticities to be greater in the long run than
in the short run, as people have time to adapt to the new change. For example, the
elasticity of demand for petrol is likely to be inelastic in the short run, because many
machines rely on petrol to run. However, over a longer term, individuals and firms are
better able to adopt alternative energy sources, making the demand for petrol more
elastic in the long run.
Notes
1 The percentage change is the proportional change in a variable times by 100.
2 Here, %⌬ simply means ‘percentage change’.
3 Also known as ‘price elasticity (of demand)’ or ‘own-price elasticity (of demand)’.
4 There are exceptional cases in which the law of demand does not hold; that is, the quantity
demanded for of a good rises when there is an increase in that good’s price so that the
demand curve has a positive slope in that range. These particular goods are called ‘Giffen
goods’. (A good for which the law of demand holds is called an ‘ordinary good’.) However,
there is very little evidence to support the existence of Giffen goods in the real world, and
we do no focus on them in this text.
5 This is also called ‘cross-price elasticity of demand’.
6 Also known as ‘income elasticity of demand’.
PA R T I V
Types of
markets

C H A P T E R
11
Introduction to
markets
11.1 Introduction to the four types of markets
We have outlined the fundamental tools necessary for a basic understanding

of markets. We will see that alternative market structures result in different
competitive environments and outcomes. An understanding of how different
markets work is essential for both firms competing with each other and for
policy makers wishing to influence market outcomes.
In the following chapters, we will look at four different types of markets:
• Perfectly competitive markets. In these markets, there are many

buyers and sellers. There are low barriers to entry and all producers sell
an identical product. Consequently, firms do not have the market power
to set prices.
• Monopoly markets. Here, there is only one seller and high barriers to
entry. As a result, the single producer has the power to choose the price
that it charges.
• Monopolistically competitive markets. In these markets, there are
many firms who differentiate themselves from each other by selling
slightly different products. As a result, these firms have some scope to
set their own prices. There are, however, low barriers to entry.
• Oligopoly markets. In these markets, there is only a handful of sellers
and high barriers to entry. Depending on the market, there may or may
not be product differentiation. The actions of the each firm in the market
affects other firms, so firms have some power to set prices but their choices
may be dictated by the actions of other firms in the market.
The four types of market are summarized in the table on the page overleaf.
92 11 Types of markets
TABLE 11.1 Types of market structure

Number Barriers Power to Product
of firms to entry set price differentiation
Perfect comp. Many Low No No
Monopoly One High Yes n/a
Mono. Comp. Many Low Yes Yes
Oligopoly Few High Yes Sometimes
C H A P T E R
12
Perfect
competition
12.1 Introduction
In previous chapters, we have referred to competition and competitive markets.

In this chapter, we examine more formally what is meant by these terms. In
particular, we construct a theoretical market, called a perfectly competitive
market, which has the most extreme characteristics in terms of competition.
We also consider economic outcomes in such a market in both the short and
the long run, and revisit the profit and welfare implications in each case.
12.2 Characteristics of perfect competition
Perfectly competitive markets have the following characteristics:
1 Many buyers and sellers. Any one buyer or seller is only a very small
part of the total market in terms of the quantity demanded or supplied.
2 Homogeneous products. All goods or services offered in the market
are identical so that, for a given price, consumers are indifferent as to who
they purchase from. We also assume that all firms have access to the same
technology.
3 Price taker. Given they are trading a homogenous product and the
number of buyers and sellers, no individual has sufficient market power
to influence market prices – that is, every participant is a price taker.
4 Free entry and exit. Firms can freely (that is, costlessly) enter and exit
the market in the long run. In other words, there are no barriers to entry
in the long run.
12.3 Supply in the short run
Recall from Chapter 7 that at least one of a firm’s factors of production is fixed in the
short run. Consequently, that firm will face a fixed cost of production that will be
incurred regardless of its output. In other words, the fixed cost is a sunk cost in the
sense that it cannot be recovered no matter what.1 This means that, in deciding the
level of output to produce in the short run, a firm will ignore its fixed costs.
12.3.1 Firm supply in the short run: the shut-down decision

In Chapter 8, we found that if a firm produces output, its supply curve is given by its
marginal cost curve. However, we also noted that there may be cases in which a firm
will not produce anything at all. If a firm chooses not to produce output in the short
run, we say that the firm shuts down.
Generally speaking, a firm will shut down if the revenue from selling that output
cannot cover the cost of producing it. In the short run, the firm should only take into
account its variable costs, as its fixed costs are sunk. That is to say, the firm will have
to pay for its fixed inputs regardless of whether or not it produces any output, so it
should ignore those costs when deciding whether or not to produce any output. Hence,
we can derive the shut-down condition that a firm will shut down in the short run
if total revenue is less than variable cost:
TR < VC . (12.1)
We can also divide both sides of Equation 12.1 by the level of output (q) to yield the
following condition:
TR VC
< ⇒ p < AVC . (12.2)
q q
This tells us that if price falls below AVC, a firm will shut down; however, if a firm
does produce a positive output, it chooses the level of output in accordance with its
supply curve – that is, its MC curve. One final thing to remember is that the MC curve
intersects the AVC curve at its minimum.2
Therefore, we can rewrite the shut-down rule for a competitive firm given in
Equation 12.2 as:
p < AVCmin (12.3)
The flip side of this is, of course, that a firm will produce positive output if p ⭓ AVCmin .
Therefore, we can say that a firm’s short-run supply curve is traced out by the part
of its MC curve that lies above AVCmin . The short-run supply curve is depicted in
Figure 12.1.
Perfect competition 95
P
SSR
ATC AVC
MC
q
FIGURE 12.1 The short-run supply curve of a firm is traced out by the part of the MC curve that lies
above AVC. In this diagram, it is denoted by the dashed line
12.3.2 Market supply in the short run

In the short run, there is no entry or exit in the competitive market. A firm is prevented
from exiting the market by its fixed costs; if a firm in the market wishes not to produce
anything, it shuts down (but does not exit). Additional firms are prevented from
entering the market because they do not have the necessary fixed inputs to establish
operations. As a consequence, the number of firms in the market is fixed in the short
run.
Following the technique outlined in Chapter 8, we derive the short-run market supply
by horizontal summation of the individual supply curves (that is, the MC curves above
AVCmin ).
12.3.3 Profits and losses

In a competitive market, it is possible for firms to make profits or incur losses in the
short run. Recall from Chapter 7 that profit is simply total revenue minus total cost.
Therefore, if a firm is making positive profit, we know that its total revenue is greater
than its total cost:
␲ > 0 ⇒ TR > TC (12.4)
Dividing both sides of this equation by q gives us the equivalent expression:
TR TC
> ⇒ p > ATC (12.5)
q q
Market Firm
P P
S MC
ATC
p*
ATC*
D
Q q
q*
FIGURE 12.2 A firm in a perfectly competitive market, making a profit. The shaded area represents the
size of that profit
In a nutshell, this tells us that if a firm is making profits, it must be the case that
price is greater than average total cost. Conversely, we can show that if a firm is making
losses (i.e. ␲ < 0), it must be the case that price is less than average total cost
(p < ATC).
Using this information, we can identify in a diagram the area that represents the
firm’s profit or loss. Figures 12.2 and 12.3 depict a firm making a profit and a loss
respectively. In both figures, the market price is given by p*. Recall that the firm is a
price taker, so this price is determined in the market by the forces of supply and demand
(in the left panel), and then the individual firms take that price as given.
Consequently, the quantity supplied by the firm is q*, as determined by the firm’s
supply curve (the MC curve); at that quantity, the firm’s average total cost is ATC*.
The difference between p* and ATC*, multiplied by the quantity supplied (q*),
represents the firm’s profit or loss; in both figures, this is denoted by the shaded areas.
In Figure 12.2, price exceeds average total cost, meaning that the shaded area denotes
the firm’s profits; in Figure 12.3, price is less than average total cost, so the shaded
area represents the firm’s loss.
12.4 Supply in the long run
In the long run, there is free entry and exit in the market. This is because all inputs
are variable; a firm wishing to exit the market is not hindered by having to pay fixed
costs and a firm wishing to enter the market has the time to acquire all the necessary
inputs to start operations. This means that all costs are opportunity costs. Hence, a firm
deciding its level of output in the long run will take into account the costs of all inputs.
As we shall see, a firm will enter or exit the market depending on its (anticipated)
level profit or loss in the market. The market will reach its long run equilibrium when
Market Firm
P P
MC
S
ATC
ATC*
p*
D
Q q
q*
FIGURE 12.3 A firm in a perfectly competitive market, making a loss. The shaded area represents the
size of that loss
the number of firms in the market stabilizes – that is, there is no longer any entry into
or exit from the market. We will argue that this occurs when firms are making zero
profits.
In this section, we will discuss how a perfectly competitive market transitions from
its short-run equilibrium to its long-run equilibrium.
12.4.1 Firm supply: the exit/entry decision

In the long run, there is free exit from a competitive market. Therefore, if a firm chooses
not to produce output in the long run, it can exit the market and incur zero production
costs. A firm will choose to exit the market if its total revenue is less than its total
costs – or, in other words, if its profits are less than zero. From Section 12.3.3, we
know that this occurs when p < ATC. Because MC curve intersects the ATC curve at
its minimum, we can write the exit condition as:
p < ATCmin . (12.6)
In the long run, there is also free entry into competitive markets. This captures the fact
that firms wishing to enter the market have enough time to obtain the necessary fixed
inputs in order to establish operations in the market. A firm will choose to enter the
market if it can make a profit by doing so (␲ > 0). This occurs when p > ATC. Noting
that the MC curve intersects the ATC curve at its minimum, we can write the entry
condition as:
p > ATCmin . (12.7)
The upshot of this is that a firm’s long run supply curve is traced out by the part of
MC curve that lies above ATCmin . The long-run supply curve is depicted in Figure 12.4.
P
MC
ATC AVC
AFC
q
FIGURE 12.4 The long-run supply curve of a firm is traced out by the part of the MC curve that lies
above ATC. In this diagram, it is denoted by the green dashed line
One thing to note, however, is that because all factors are variable in the long run,
the firm will choose the most appropriate mix of inputs for its level of output. This
may entail a different mix of inputs for various levels of output. What this means is
that the firm’s long-run marginal cost curve may not be the same as its short-run
marginal cost curve – and it is, of course, the long-run marginal cost curve that should
be taken into account here. For analytical simplicity, we tend to assume the two curves
are identical; however, it is worth bearing in mind that this is not necessarily the case.
12.4.2 Elimination of profits and losses

We have now shown that firms will enter the market if it would be profitable to do so
and will exit the market when they are sustaining losses. Together, these decisions will
eliminate all profits (and losses) in the market.
• When firms in the market are profitable (p > ATC), firms will want to enter the
market. The entry of more firms into the market will progressively shift the short-
run supply curve to the right, driving the equilibrium price downwards. This shift
is illustrated in Figure 12.5 with a shift in the short-run supply curve from S to S′;
S′ is derived after entry of new firms into the market, horizontally summing the
individual firms’ supply curve.
• When firms in the market are sustaining losses (p < ATC), firms will tend to exit
the market. The exit of firms from the market progressively shifts the short-run
supply curve left, pushing the equilibrium price upwards. This shift is illustrated
in Figure 12.6 with the shift in the short-run supply curve from S to S′ after firms
have left the industry. (Remember, each short-run supply curve is derived for a
fixed number of firms.)
Market Firm
P P
S MC
S'
ATC
D
Q q
FIGURE 12.5 When the market price is above average total cost, firms in the market are making profits.
This will encourage entry into the market, shifting the supply curve right from S to S′. In turn, this will
put downward pressure on market prices
Market Firm
P P
S' MC
S
ATC
D
Q q
FIGURE 12.6 When the market price is below average total cost, firms in the market are making losses.
This will encourage exit from the market, shifting the supply curve left from S to S′. In turn, this will put
upward pressure on market prices
In other words, if price tends to decrease when it is above ATC and increase when
it is below ATC, then it will eventually settle at ATC in the long run. In particular,
because the firm supply curve cuts the ATC at its minimum, the price will settle at
p = ATCmin . Thus, because the relationship between price and average total cost
determines the level of profits, it follows that there will be zero profits in the long
run (␲ = 0). The long-run equilibrium is depicted in Figure 12.7.
Market Firm
P P
S MC
ATC
p*
D
Q q
Q* q*
FIGURE 12.7 In the long run, there are zero profits in a perfectly competitive market. This requires
p = ATCmin . Because there are zero profits, there is no incentive for any firm in the market to exit
the market and there is no incentive for any additional firms to enter the market. In this diagram,
the long-run equilibrium price is p*, the quantity traded in the market is Q * and the output of the
firm is q *
12.5 Market supply in the long run
When deriving the long-run market supply curve, we need to account for the fact that
the market responds to demand via the entry and exit of firms. In fact, we now know
that, in the long run, price will adjust back to the minimum of average total cost, no
matter what the quantity traded in the market is. This tells us that the long-run supply
curve is perfectly elastic at p = ATCmin , as illustrated in Figure 12.8. An industry with
a perfectly elastic long-run industry (or market) supply curve is a constant-cost
industry. Unless otherwise stated, a competitive industry is assumed to be a constant-
cost industry.
This also has another implication. Because a competitive industry minimizes ATC
in the long run, it also minimizes the total cost of production. Moreover, all trade for
which MB > MC occur, meaning that all gains from exchange are realized. This means
that a perfectly competitive industry maximizes the level of total surplus available.
As an example of the dynamics following a change in demand in a constant-cost
industry, consider the industry and a representative firm illustrated in Figure 12.9. This
figure shows the short- and long-run changes for both the firm and at the market level
following an increase in demand from D1 to D2. Note the axes of q and Q for the firm
and market. Market price is determined by the intersection of the short-run market
supply curve S1 and demand D1, as shown in the right-hand panel. Initially the industry
is in its long-run equilibrium, as the short-run supply curve S1 intersects demand D1
at p* = ATCmin . Note that this price p* = ATCmin is the industry long-run supply curve.
In the left-hand panel, the firm has a supply curve of its M C curve above the minimum
Firm Market
P P
MC S
ATC
LRS
p*
D
q Q
q* Q*
FIGURE 12.8 In the long run, free entry and exit means that the price in a constant-cost industry
will always be driven back to ATCmin . The long-run market supply curve LRS is perfectly elastic
at P = ATCmin
of its ATC. It takes the market price of p* as given, and sells the q* units; this is its
profit-maximizing output, but it is making zero profits. There is no incentive for entry
and exit as all firms in the industry are earning zero economic profits. The equilibrium
is also illustrated at the market level.
Assume then, there is a shift of demand to D2. In the short run, there is no entry or
exit, so the new market equilibrium is the intersection of S1 and D2; price rises to P2.
This is now the price faced by firms (it is their new MR curve). In response they increase
their quantity supplied to q2, and earn positive economic profits, as illustrated by the
shaded area.
Finally, consider the new long-run outcome. In the long run, firms are free to entry
or exit; given the opportunity for economic profits, new firms will come into the market.
With entry, the short-run supply curve will shift to the right. With entry, the market
price will be successively determined by the intersection of the short-run market supply
curve and D2. This process of entry and dropping market prices will continue until
there is no further incentive for entry (with a short-run supply curve of S2); this occurs
when economic profits are zero, hence when price has dropped to p* = ATCmin . In the
long-run equilibrium price equals P*, which coincides with the long-run supply curve
LRS, as well as the intersection of S2 and D2. Market output is Q2.
Note that the price has returned to its original level P* again this is the MR curve
for the price-taking firm, and it will supply a quantity of q*, the same quantity it
sold in the original equilibrium outcome with D1. Thus, although the total quantity
supplied in the market has increased (Q1 to Q2), each individual firm still sells q*.
The increase in the market output comes about because there are now more firms in
the market in the long run.
In a constant-cost industry, this process of entry or exit in the long run following a
change in demand ensures that firms earn zero profits and the price is p* = ATCmin .
Firm Market
P P
MC S1
ATC
S2
p1
LRS
p*
D2
D1
q Q
q* q1 Q*
FIGURE 12.9 Following an unanticipated increase in demand, in the short run price rises and firms in
the industry make positive economic profits. However, in the long run, entry forces prices back down to
the p * = ATCmin . Each firm sells q * units and economic profits are zero
Several caveats are worth noting here. When deriving the long-run industry supply
curve in a constant-cost industry, we assume that all firms and potential entrants have
the same costs (and access to technology). Consequently, free entry drives price back
to the minimum of ATC – this is the only way that there will be no incentive for further
entry or exit. However, if potential entrants have higher costs than incumbent firms
(already in the market), the long-run industry supply curve need not be perfectly elastic.
One possible example would be the mining industry; existing mines typically have the
most accessible mineral deposits (with lower associated costs) than potential entrants,
who might have to progressively mine deposits of lower quality or in more difficult
and costly locations (deeper under the surface, under the ocean, and so on).
As an illustration, consider Figure 12.10. Following an increase in demand entry of
new firms will continue to occur until the next potential entrant does not anticipate
making a positive economic profit (that is, it assumes that following its entry P < ATCpe ,
where ATCpe is its expected average total costs of production. As the potential entrants
have higher costs than the existing market participants, price does not have to fall
all the way to the ATCmin of firms in the market to prevent further entry; in the long-
run equilibrium P > ATCmin . If we continue to trace out the long-run supply curve
(accounting for entry), as demand shifts to the right, with each shift higher cost firms
will entry the market, and the long-run supply curve will be upward sloping. An
industry with an upward-sloping long-run supply curve is called an increasing-cost
industry. Moreover, in an increasing-cost industry, firms (with lower production
costs) can earn positive economic profits in the long run.
Finally, consider an alternative scenario, illustrated in Figure 12.11. Suppose that
as output in an industry expands, costs for all firms fall. Consider the software industry.
As it has expanded (essentially from the 1970s until now), there are increasing numbers
of talented programmers and hardware inputs have become cheaper. In this sort of
S2
S1
LRS
P2
P1
D2
D1
Q
Q1 Q2
FIGURE 12.10 In an increasing-cost industry, the long-run industry supply curve is upwards sloping
S1
S2
P1
P2
LRS
D2
D1
Q
Q1 Q2
FIGURE 12.11 In a decreasing-cost industry, the long-run industry supply curve is downward sloping
industry, if demand increases, market output will expand and average costs for all firms
could actually fall. If this is the case, following an increase in demand, as entry will
continue until it is no longer profitable, the new long-run equilibrium price has to be
lower than the initial equilibrium price. In this case, the long-run industry supply curve
is downward sloping – this is a decreasing cost industry.
In this chapter, we set up a stylized type of competition, called a perfectly competitive

market. In such a market, we found that firms in the industry can sustain profits or
losses in the short run, but that profits and losses will be eliminated in the long run.
The driving factor behind this is the assumption of free entry and exit, which allows
firms and hence the supply curve to respond to the market price.
In practice, firms in the real world rarely report zero profits, even in markets that
are very competitive. There are several reasons why this might be the case. First,
it is very unlikely that a real-world market will meet the stringent criteria of perfect
competition. For example, firms in the market might have slightly differentiated
products or there might not be perfect information about prices. Second, it is important
to distinguish between accounting profits and economic profits. Real-world firms tend
to report accounting profits, whereas this model deals with economic profits. Third, it
is possible that real-world markets have not yet ‘settled’ into their long-run equilibrium.
This might occur if there are ongoing demand shocks.
Notes
1 Recall our discussion of opportunity costs versus sunk costs in Chapter 1.
2 See Chapter 7 for why this is the case.
C H A P T E R
13
Monopoly
13.1 Introduction
In the last chapter, we looked at a market that exhibited the most extreme
characteristics of competition. Now we turn to the other end of the spectrum:
here we examine markets with only one seller. A market with one seller is a
monopoly, and that seller is a monopolist. In this chapter, we will examine
how the market power of a monopolist allows it to charge higher prices in
order to increase its profits, and we consider the welfare implications of this
market power. We then analyse how the monopolist might use price
discrimination to further increase its profits. Finally, we look at the regulation
of a natural monopoly, where it is less costly for the whole market to be
serviced by one firm, rather than two or more.
13.2 Characteristics of a monopoly
Monopolies have the following characteristics:
1 One seller and many buyers. There is a single producer of all output
in the market.
2 Price maker. Because the monopolist is the only firm in the market, it
has the market power to determine the price in the market – that is, it is
a price maker.
3 Barriers to entry. Firms that might like to enter the market are prevented
from doing so by barriers to entry. Barriers to entry may exist for a number
of reasons: the monopolist may have access to a natural resource or
technology that is not available to other firms; the monopolist might hold
a patent or a copyright that prevents other firms from selling the same product;
the government might ban entry by other potential seller; or, the monopolist might
simply have a lower cost of production that effectively allows them to prevent
other firms from entering the market.
13.3 The single-price monopolist
In this section, we examine the behaviour of a monopolist who charges the same price
to all of its consumers (also known as a single-price monopolist).
Because the monopolist is the sole producer, it faces all the demand in the market.
In other words, the monopolist faces the downward-sloping market demand curve.
However, the monopolist does not itself have a supply curve; recall from Chapter 8
that the supply curve applies only to competitive firms, as a supply curve is derived
assuming a firm is a price taker.1 Therefore, we cannot apply our demand-supply
framework to determine the price that a monopolist will charge.
Instead, we will determine the monopolist’s price by considering its profit-
maximizing choice. In order to do this, we will first need to understand the monopolist’s
marginal revenue curve.
13.3.1 Marginal revenue

Marginal revenue is the additional revenue that the firm received from selling one
extra unit of a good. Because the monopolist faces a downward- sloping demand curve,
if it increases output by one unit the price will fall by some amount. In other words,
for a monopolist there are two effects at play: (i) the increase in output increases total
revenue, and (ii) the decrease in price decreases total revenue. Either effect could
dominate.
To calculate marginal revenue (MR), we differentiate total revenue equation (TR)
with respect to output (Q). This tells us how total revenue changes as we increase output
by one unit:
dTR
MR = . (13.1)
dq
To illustrate how this is done, consider a demand curve given by the equation
P = a – bQ, where P is price, Q is the quantity demanded, and a and b are positive
parameters (for example, the demand curve could be *P = 100 – 2Q). We can write
the total revenue equation as follows:
TR = P ⫻ Q = (a – bQ) ⫻ Q = aQ – bQ2 . (13.2)

Monopoly 107
MR D
Q
a a
2b b
FIGURE 13.1 When the demand curve is linear, the marginal revenue curve has the same vertical
intercept and twice the slope of the demand curve
Now, we can differentiate the right hand side of Equation 13.2 to obtain marginal
revenue:
dTR
MR = = a − 2bQ. (13.3)
dq
Two things should be noted about the MR equation. First, the MR equation has the
same vertical intercept as the demand curve; that is, they both cross the P-axis at a.
Second, the MR curve has twice the slope of the demand curve; the MR curve has a
slope of –2b whereas the demand curve has a slope of –b. As a result, the MR curve
will intersect the Q-axis at exactly half the quantity of the demand curve intersection;
the MR curve cuts the Q-axis at a/2b and the demand curve cuts the Q-axis at a/b. This
is illustrated in Figure 13.1.
Because we have not specified values for a and b, the equation P = a – bQ could
represent any linear demand curve. Hence, our observations about the relationship
between the demand and MR equations will hold true for any linear demand curve.
What this means is that, for any given linear demand curve, we can obtain the equation
of the MR curve by simply doubling the gradient (or slope) of the demand curve
equation.2
Example. Suppose the demand curve is given by the equation Q = 100 – 4P. In
order to find the equation of the MR curve, we need to rearrange the demand curve
equation so it is expressed with P as the subject: P = 25 – 1/4Q. Now, to obtain
the MR equation, take the slope of the demand curve (–1/4) and double it to obtain
MR = 25 – 1/2Q.
13.3.2 Profit maximization

We can now determine the profit-maximizing price (and quality) for a monopolist.
Profits will be maximized when a monopolist sets marginal revenue equal to marginal
cost. To see why this is the case, recall that profit is the difference between total revenue
and total cost:
␲ = TR – TC (13.4)
Now, to maximize profit, we need to take the first derivative and set that equal to
zero:
d␲ dTR dTC
= − = MR − MC = 0 (13.5)
dQ dQ dQ
Rearranging gives us the profit-maximizing condition:
MR = MC (13.6)
Indeed, if MR > MC the monopolist can increase its profit by selling one extra unit;
that is, it should increase production. On the other hand, if MR < MC the monopolist’s
profit falls from selling the last unit, so it would be better off from not selling that unit;
that is, it should decrease production. Consequently, the profit-maximizing level of
output for the monopolist is the quantity where MR = MC.
Example. Consider a monopolist whose demand curve is given by P = 100 – Q

and a total cost curve of TC = Q2 + 50. We can derive the marginal revenue curve
by doubling the gradient of the demand curve: MR = 100 – 2Q. We can also obtain
the marginal cost curve by differentiating the total cost function: MC = 2Q. Setting
MR = MC yields: 100 – 2Q = 2Q ⇒ Q = 25. We can now substitute Q = 25 back
into the demand equation to find the monopoly price P = 100 – 25 = 75.
The profit-maximizing choice for a monopolist is illustrated in Figure 13.2. The

monopolist faces demand curve (D) and a marginal revenue curve (MR).
The monopolist also has some total cost curve, from which we can derive marginal
cost (MC) and average total cost (ATC). As can be seen in the diagram, the level of
output associated with associated with MR = MC is denoted Q m. The price associated
with this quantity is P m, which is the profit-maximizing price.
We can also use this diagram to show the amount of profit (or loss) made by the
monopolist. Recall that profit can be written as follows:
␲ = TR – TC = P ⫻ Q – ATC ⫻ Q = (P – ATC)Q (13.7)
Thus, at the profit-maximizing outcome (Q m, P m), profit is the area between ATCm
and Pm, multiplied by the quantity Qm. This is represented by the shaded
Monopoly 109
MC
ATC
Pm
MR = MC
MR D
Q
Qm
FIGURE 13.2 A profit maximizing monopolist sets MR = MC, and thus produces quantity Q m. At this
quantity, the market price will be P m
MC Proit
ATC
Pm
ATCm
MR D
Q
Qm
FIGURE 13.3 The monopolist’s profit is given by ␲ = Q (P – ATC ). The area corresponding the
monopolist’s profit is shaded green in this diagram
region in Figure 13.3. This also means that the monopolist makes an average profit of
P m – ATCm on each unit that it sells.
Finally, in the preceding discussion we have framed the monopolist’s choice in terms
of what price it should charge. That is, the monopolist charges the price Pm, which
results in a market quantity of Q m. However, we might equally look at the problem
from the perspective that the monopolist chooses a profit-maximizing level of output,
Q m, which determines the price of Pm in the market. Because the demand equation
determines the relationship between price and quantity, setting price to maximize profit
yields the same outcome if the monopolist chooses the profit-maximizing level of output
(that is, once either price or quantity is determined, the other variable is determined
by the demand curve).
13.4 Welfare under the single-price monopolist
In Chapter 9, we found that welfare is maximized in a competitive market; the market

equilibrium is where MB = MC – the equilibrium quantity traded will be where
demand curve (the MB curve) intersects supply (the MC curve). The quantity traded
at this point is called the ‘competitive quantity’, and is marked Q* in Figure 13.4. By
contrast, a monopolist trades up to the point where MR = MC, selling the ‘monopoly
quantity’, and is marked Q m in Figure 13.4. Note, as MB > MR for all units q > 0, this
means that the monopolist sells less than the competitive market quantity Q*.
MC
Pm
P*
MR D
Q
Qm Q*
FIGURE 13.4 The quantity traded in the market is lower under monopolist than in a competitive market
13.4.1 Consumer and producer surplus

Let us begin by identifying consumer surplus under a monopoly. Recall from Chapter
9 that the size of consumer surplus is given by the area below the demand curve and
above the price line for all units traded. In the case of a monopoly, we will need to
use the monopoly price, P m, rather than the competitive equilibrium price. The area
that represents consumer surplus can be seen in Figure 13.5.
Let us now turn to producer surplus. Recall from Chapter 9 that the size of producer
surplus is given by the area above the marginal cost curve and below the price line
for all units traded. Again, we will need to use the monopoly price, P m. We will also
need to remember that the quantity traded under a monopoly is Q m, so the untraded
units beyond that point to not yield any surplus. Hence, the area of producer surplus
will be truncated at Q m. The area that represents producer surplus can be seen in
Figure 13.5.
Compare these areas with the areas of consumer surplus and producer surplus in a
competitive market (see Figures 9.6 and 9.8). Notice that the area of consumer surplus
under monopoly is smaller than consumer surplus in a competitive market, while the
Monopoly 111
MC
CS
PS
Pm
MR D
Q
Qm
FIGURE 13.5 Consumer surplus and producer surplus under a monopoly
monopolist’s producer surplus is larger than that of a competitive market. By charging

a higher price, the monopolist is converting some consumer surplus into producer
surplus.
13.4.2 Total revenue and deadweight loss

We can now find total surplus under a monopoly by adding together consumer surplus
and producer surplus, as shown in Figure 13.6. Compare this with total surplus in a
perfectly-competitive market (see Figure 9.10). As you can see, the area of total surplus
under a monopoly is smaller than the area of total surplus in a competitive market.
The reason for this is that the quantity traded in the market is lower under a
monopoly (Q m) than under perfect competition (Q*). As a result, the gains from trade
for units between Q m and Q* are not realized; that is, there is a loss of welfare resulting
from these units not being traded. We refer to this lost welfare as deadweight loss
(DWL). The DWL of monopoly is also depicted in Figure 13.6.
Thus, a monopoly is inefficient because it does not maximize total surplus. This DWL
is not caused, per se, by the fact that the monopolist converts consumer surplus into
producer surplus. Instead, DWL is caused by the fact that this conversion is not perfect;
that is, some surplus is lost in the process. By charging a higher price, the monopolist
restricts output below the efficient level, and it is this decrease in output that generates
the DWL. Stated another way, the DWL of monopoly is caused because a monopolist
restricts its output below the efficient quantity (Q m < Q*); as there are consumers in
the market with a higher MB than the MC of producing extra units of the product, total
surplus would increase if these extra units were produced (until output reaches Q*).
Rather than total surplus, the monopolist is concerned with maximizing its profit; it
stops production when MR = MC (at Q m). As the MR curve lies below MB (the demand
curve) for every unit sold except for the very first, it must be the case that MB > MC
MC TS
DWL
Pm
MR D
Q
Qm Q*
FIGURE 13.6 Total surplus and dead weight loss under a monopoly
at the profit-maximizing level of output for the monopolist. This ensures that Q m < Q*
for a monopolist charging a single price to all consumers. As a result, there will be a
DWL from monopoly.
13.5 Price discrimination
Let us now turn to the possibility that the monopolist can engage in price
discrimination. In economics, price discrimination is when a firm charges a different
price to different consumers for the same product, or when a firm sells different versions
(quality or quantity) of the product when the change in price between the versions does
not solely reflect the difference in the cost of production.
Price discrimination allows a monopolist to increase its profit further relative to when
it charges the same price to all consumers. Earlier in this chapter, we saw that a
monopolist charges a higher price than a competitive firm. But when it does this, the
quantity traded in the market falls; in particular, consumers with a lower willingness
to pay drop out of the market. The idea behind price discrimination is to bring these
consumers back into the market by charging them a lower price, while still charging
a higher price to customers with a higher willingness to pay. If a monopolist can charge
a high price to those with a high willingness to pay, while also charging a lower price
to consumers with a lower willingness to pay, it can increase profits. This is the key
intuition behind price discrimination.
In order to engage in price discrimination, a firm will of course need to have the
necessary market power to determine prices in the market. The firm will also need
to be able to prevent arbitrage; that is, prevent consumers who are charged a lower
price from reselling the product to the customers with high willingness to pay. The
firm will also need some information about different customers and their willingness
to pay for the product.
Monopoly 113
The ability of a monopolist to engage in certain types of price discrimination

depends on the information it has and how much it can prevent arbitrage. We now
turn to the three different types of price discrimination.
13.5.1 First-degree price discrimination

A monopolist engages in first-degree price discrimination (also called ‘perfect price
discrimination’) when it charges each consumer his or her exact willingness to pay
(i.e. marginal benefit) for every unit consumed. Consequently, the monopolist extracts
all of the consumer’s surplus and receives all the gains from trade in every transaction.
Example. Suppose it costs the monopolist $1 to produce an ice cream cone. Bonnie
is willing to pay $10 for one ice cream cone, whereas Jen’s willingness to pay is
$5. If the monopolist were to charge a single price, it would charge $10 and make
a profit of $9. However, a monopolist employing first-degree price discrimination
can charge Bonnie $10 for an ice cream cone and charge Jen $5 for an ice cream
cone, making a profit of $13.
Of course, in order to engage in first-degree price discrimination, the monopolist

will need to have perfect information about each consumer’s willingness to pay. It will
also need to be able to prevent arbitrage; in the above example, the monopolist would
need to be able to prevent Jen from buying two ice cream cones at the lower price and
reselling one to Bonnie.
While in the illustration above each consumer purchased at most one unit of the
good, first-degree price discrimination can also apply where each consumer buys
multiple units of the good (that is, consumers have typical downward sloping demand
curves). In this case, the monopolist can simply price each unit at the relevant
consumer’s willingness to pay for that unit of the good. The monopolist will continue
to sell units to the customer provided that their MB ⭓ MC. In fact, the MB curve
now becomes the MR curve for the first-degree price discriminating monopolist.
Consequently, the monopolist will sell units up to the point where MB = MC, meaning
that the efficient quantity is traded (i.e. there is no DWL). That is, as the monopolist
captures all of the gains from trade, it will want to continue to sell more units provided
MB ⭓ MC. However, as noted above, all surplus in the market is producer surplus.
This is illustrated in Figure 13.7.
An alternative way of implementing first-degree price discrimination is by use of a
two-part tariff. Under a two-part tariff, the monopolist charges the consumer two
distinct fees: (i) a fixed fee, F, that does not change with the number of units consumed;
and (ii) a per-unit fee, p, that is paid for each unit of the good consumed. In the most
obvious cases, the fixed fee could be an entry fee, access fee or membership fee; the
per-unit fee would be the price of using the good or service. Examples of this include
a joining fee for a gym with a visit price for each time it is used, or an access fee for
the provision of electricity services and a per-unit fee for the amount of electricity
actually consumed. However, there are a number of less obvious cases, where the fixed
P
PS
P* MC
D = MR
Q
Q*
FIGURE 13.7 When the monopolist engages in first-degree price discrimination, the efficient quantity is
traded in the market. However, all surplus in the market is producer surplus
fee is the price of a core part of the product, and the per-unit price is the price of a
non-durable (or less durable) part of the product. Examples of this include a razor
handle and razor blades; a printer and ink cartridges; a gaming console and games.
To illustrate how a two-part tariff works, consider Figure 13.8. Suppose D represents
the demand curve of a consumer in the market and MC represents the monopolist’s
marginal cost. In order to induce the consumer to purchase up until the point where
MB = MC, the monopolist will need to charge a per-unit price of p = MC. If the
monopolist were not to charge a fixed fee, the consumer would anticipate getting surplus
P
Fixed fee
revenue
Per-unit fee
revenue
MC
D
Q
Q*
FIGURE 13.8 When the monopolist uses a two-part tariff, it charges a fixed fee (F ) equal to the size of
the lighter shaded triangle. It then charges a per-unit fee (p) for each unit consumed equal to MC. The
total revenue from the per-unit fee is represented by the darker shaded area. The total revenue of the
monopolist overall is represented by all of the shaded areas
Monopoly 115
equal to the area of the light grey triangle; this represents how much more the consumer
would have been willing to pay to have access to the product (at a per-unit price of
p). Therefore, the monopolist could select a fixed fee that the consumer has to pay
before consuming the good, equal to the amount of that consumer surplus (F = CS).
Thus, the monopolist extracts all of the consumer’s surplus, converting it to producer
surplus.3
13.5.2 Third-degree price discrimination

A monopolist engages in third-degree price discrimination when it separates
consumers into ‘markets’ and charges a different price in each market. Some examples
of this include: adult and student prices for movie tickets; different prices for haircuts
for men and women; and different prices for software in different countries.
The information and arbitrage requirements that support third-degree price dis-
crimination are less stringent than for first-degree price discrimination. First, the
monopolist needs to have a way of identifying the market that any particular individual
belongs to. For example, a monopoly cinema can identify students from non-students
as the students have an ID card. Alternatively, a monopolist might be able to identify
which type of consumer an individual is from their location (or even the location of
their IP address). This information is critical for a monopolist to be able to engage in
third-degree price discrimination. Second, the monopolist needs to prevent arbitrage
between markets; the monopolist, however, cannot prevent arbitrage within a group.
This means that a monopolist can discriminate between groups, but not within groups
– this means that they charge a single (different) price to each market. For example,
a monopoly cinema charges one price to all students with valid IDs and another price
to adults with no student IDs. The monopolist cannot prevent one student with a valid
Market A Market B
$ $
PmA
PBm
MC MC
MR A DA MRB DB
QA QB
Q m
A Q m
B
FIGURE 13.9 When a monopolist engages in third-degree price discrimination (with constant MCs, it
maximizes profit where MRA = MRB = MC. The price is higher in the market where demand is relatively
inelastic; here PB > PA
student card on-selling her ticket to another student. Similarly, any adult can buy and
resell a ticket to another adult. Consequently, the movie cinema has no scope to prevent
arbitrage (resale) within each market, so they charge one price for a student ticket and
a different price for an adult (or non-student) movie tickets. Consequently, individuals
within a market face the same price (for example, the student price for a ticket at the
movies) and therefore cannot benefit from arbitrage. Second, the monopolist does not
need to know each individual’s demand curve; it only needs to know the demand curve
for each market.
Under third-degree price discrimination, the firm essentially acts as a single-price
monopolist in each market. This just means that it faces a different demand (and hence
marginal revenue) curve in each market. Therefore, assuming that the monopolist faces
a constant marginal cost, MC, it simply solves for the profit-maximizing price/quantity
in each market separately. In other words, in Market A, it should set MRA = MC;
in Market B it should set MRB = MC; in Market C, MRC = MC; etc. To illustrate the
intuition behind third-degree price discrimination further, consider the case of a
monopolist with a constant marginal cost selling to two types of consumers, types
A and B respectively. In each market there is a downward-sloping demand curve and
a corresponding marginal revenue curve, labelled here as MRA and MRB . To maximize
profit, the monopolist will sell at the point where MRA = MRB = MC. For a given level
of output, if MRA > MRB, the firm could increase profit by selling the last unit of output
in market A rather than B; hence, for profit to be maximized, marginal revenues need
to be equated across markets. Second, if MRA = MRB > MC, profit can be increased
by increasing output – an extra unit of output generates more revenue than it increases
cost. The reverse argument applies if MC exceeds MR. Hence, for profit maximization
MC needs to be equal to MR in both markets.4
Finally, note that as the MRs are equal across the different market with third-degree
price discrimination, prices can differ between the two markets. As an example,
consider Figure 13.9, which show a monopolist with a constant MC selling its product
to consumers in two markets. Note, an increase in the quantity sold in market A is
indicated by a rightward movement from the origin, in the usual way. An increase in
the quantity demanded in market B is shown by a leftward movement from the origin.
As noted above, to maximize profit the monopolist equates MRA = MC = MRB; given
the two demand curves, however, the price in market B is higher than the price charged
in market A. The monopolist charges a higher price in the market with demand that is
relatively inelastic, and a lower price in the market with relative elastic demand. For
instance, in the real-world example above, often movie ticket prices for students are
lower than the full adult price – this is because adults are in the market with relatively
inelastic demand compared with the students. Similarly, download prices for computer
software often differ between countries – the price for exactly the same software is
often more expensive in Australia than in the United States. In this case, Australia has
the relatively inelastic demand curve as compared with the US.
Monopoly 117
13.5.3 Second-degree price discrimination

A monopolist engages in second-degree price discrimination when a monopolist
knows that there are different types of consumers in terms of their WTP, but it does
not know the type of any particular individual consumer. For example, in a market
there might be high-value consumers (with a WTP for the product of $100) and
low-value consumers (with a WTP of $50), but the monopolist might not be able to
ascertain which group any person falls into. Note the difference between third-degree
and second-degree price discrimination; with third, the monopolist knows exactly what
type of consumer every individual is (a student, a non-student etc.), but with second-
degree the monopolist only knows the possible types that a consumer could be.
Because the monopolist cannot identify which type any particular consumer is, it
needs to offer different versions of the product (at different prices) so that the consumers
‘self-select’ and reveal their types by the product versions they choose. The idea here
is that the versions of the product will be designed so that the high-value consumer
will buy the expensive product, and the low-value consumer will buy the cheaper
product. In order to effectively do this, the monopolist must design the different versions
so that the high-value consumer is at least as well off buying the expensive product
than buying the cheaper option (and the low-value customer must be at least as well
off buying the cheaper product than not buying at all). If not, the high-value consumer
will opt for the cheaper version, reducing the monopolist’s profit.
The classic example of second-degree price discrimination is airline tickets. An
airline might know that there are two types of travellers: a business traveller, with a
high willingness to pay, and a leisure traveller with a relatively low willingness to pay.
The problem for the airline is that when selling the tickets, they do not know whether
a potential customer is a business or leisure traveller. The airline might wish to sell
two types of ticket – business class (which is expensive) and an economy (cheap) ticket.
In order to get the business traveller to buy the expensive ticket, she must be better
off flying business rather than economy. The airline can ‘encourage’ this choice by
not only increasing the value of the expensive option (business-class lounge, flexible
ticketing arrangements, better food, more comfortable seats) but by making the cheaper
option less attractive (inflexible ticketing arrangements, requirement to book the ticket
a long time in advance, uncomfortable seats, inedible food, and so on). Reducing the
attractiveness of the economy ticket encourages the high-value traveller to opt for
the business-class ticket. It also increases the amount that the airline can charge
for the expensive option, because the cheaper version of the product gives the high-
value consumer less net surplus.
Other examples of second-degree price discrimination abound. The ‘versions’ of the
product sold by the monopolist may be different in quality, quantity or even timing:
for example, a soft drink manufacturer may sell cans of soft drink individually or in
boxes of 24 with the intention that people who drink more soft drink will opt for the
24 pack; similarly, a toothpaste manufacturer might charge a different per-unit price
for a small tube than it does for a larger tube of toothpaste; or, a pub may charge less
for drinks during happy hour than otherwise. Fashion labels typically charge higher
prices for items in a new collection while offer discounts on last season’s stock,
discriminating between those consumers who can and cannot wait.
Finally note that second-degree price discrimination has lower information and
arbitrage requirements than the other two types of price discrimination – all that the
monopolist needs to know is the different types of consumer there are in the market.
Because any individual consumer can purchase any of the versions, there are no
advantages to resale and therefore a monopolist need not be able to prevent arbitrage.
The monopolist will still need to know the demand curve of individuals who belong
to a particular group, but they need not be able to identify which group an individual
consumer belongs to. This is because second-degree price discrimination relies on self-
selection – that is, the consumer will of their own volition buy the version that is
intended for them.
13.5.4 Further comments about price discrimination

Price discrimination is a way that a monopolist can increase its profits. Indeed, the
ability to price discriminate will never decrease a firm’s profits, because the monopolist
always has the option of just charging a single price to all consumers.
As stated above, the ability to employ price discrimination depends upon the firm’s
ability to prevent arbitrage and the information they have about consumers. Because
price discrimination increases profit, it is often beneficial for the monopolist to take
steps to prevent resale or to learn more about their consumers. Examples of steps that
firms have taken to prevent resale include region settings for DVD players and disks
and the non-transferability of airline tickets. Examples of steps taken to learn more
about consumers include online sellers using cookies to track a consumer’s previous
purchases, location, and so on.
In our discussion above, we discussed separately the three types of price discrim-
ination. However, it should be noted that it is possible for a monopolist to employ
different types of price discrimination in combination. For example, public transport
tickets exhibit characteristics of both third- and second-degree price discrimination.
Usually, there are different prices for adults and students, who can be differentiated
from each other based on whether they hold a student card. There are usually discounted
prices at the weekend and consumers are left to decide for themselves when to travel.
13.6 Natural monopoly
A natural monopoly is an industry where a single firm can supply an entire market
at a cost lower than two or more firms. Typically, this occurs where the industry has
a large fixed cost and relatively low marginal costs. Some examples of natural
monopolies are electricity, water and communications networks, where the cost of
building the infrastructure is high, but the marginal cost of delivering the service is
relatively low.
Monopoly 119
ATC
MC
FIGURE 13.10 When a firm has a fixed cost and a constant marginal cost, the average total cost curve
will be downward sloping for all values of Q; this industry will be a natural monopoly
One example of a natural monopoly is when a firm has declining average total costs
for all relevant levels of output required in a market. This might occur if the firm has
a high fixed cost and constant marginal costs, leading to declining average total costs
for all levels of output. This is illustrated in Figure 13.10. In this case, a single firm
will be able to produce a given level of output at a lower average cost than two or
more firms, because a single firm only has to incur the fixed cost once, whereas multiple
firms would have to incur the large fixed cost once each.
A natural monopoly arises from the combination of the level of demand and the
state of technology (or production costs). It is possible that an industry will cease to
be a natural monopoly if the level of demand increases sufficiently such that two firms
could meet the required quantity at a lower cost than could one firm. It is also possible
that a change in the state of technology will alter the costs of production (and hence
the cost function) so that it is cheaper to produce the required output with two or more
firms. For example, the telecommunications industry has undergone significant
technological changes, meaning that some services that were historically a natural
monopoly are not anymore.
13.7 Regulating a natural monopoly
Given it is charging single price (and not engaging in first-degree price discrimination),
a natural monopoly will create a DWL. One way the government may wish to address
this problem is by intervening in the market. In this section, we discuss three options:
government ownership of the monopoly; marginal-cost price regulation and average-
cost price regulation.
13.7.1 Government ownership

A privately owned monopoly will maximize profit, disregarding its impact on overall
welfare. In principle, the government take over the ownership and operation of the
monopoly and charge the welfare-maximizing price (rather than the profit-maximizing
price).
However, this option can be difficult to implement in practice. The government will
need to employ a manager to run the firm; it is unlikely that the government itself
will have the skills or knowledge to run such a business. Moreover, the manager will
typically have personal objectives that conflict with the goal of welfare-maximization
– a manager may wish to maximize their own perks, empire build, and so on, and
without an explicit performance contract, the manager will have little incentive to
minimize costs, which itself could lead to loss in surplus. This could result in DWL
greater than that of a privately owned monopoly. Of course, this is not to say that
government ownership will always be problematic; rather, we should just be mindful
it is not necessarily the best solution in all circumstances, and that care is needed to
mitigate the issues that government ownership itself creates.
13.7.2 Marginal-cost price regulation

The government could also regulate the price that the monopolist can charge. In order
to maximize surplus, the government may wish to mandate that the monopolist charge
the efficient price, P = MC. This would ensure that the efficient quantity is traded in
the market, thereby eliminating any DWL.
However, at this price, the monopolist only just covers its variable costs of
production. As seen in Figure 13.11, the monopolist will make a loss equal to the size
ATC
P* MC
D
Q
Q*
FIGURE 13.11 Under marginal-cost price regulation, the government sets the monopoly price at
P = MC (assuming constant MC for simplicity). However, this means that the monopolist makes a loss
equal to the shaded area (that is, its fixed costs). The government will need to subsidize the monopolist
that amount to prevent them from leaving the market
Monopoly 121
of its fixed costs and may choose to exit the market in order to avoid incurring that
loss. If the government wants the monopolist to stay in the market, it will need to
subsidize the monopolist so that the monopolist can cover its fixed costs. Such a subsidy
may be problematic. It can be politically unpopular to funnel tax revenue to a mon-
opolist given the many other things a government could do with those funds. Moreover,
raising tax revenue generally results in a DWL in another market (as we will see later
in Chapter 16), meaning that there may still be loss of surplus.
13.7.3 Average-cost price regulation

To avoid the need for a subsidy, the government could instead impose a price equal
to average cost, so that the monopolist is able to charge a price that just covers its cost
of production. This type of regulation is illustrated in Figure 13.12. In this case, price
is set equal to the point where the ATC curve intersects the demand curve. At this price,
the quantity sold in the market is Q r. Note, however, that this quantity is less than the
efficient quantity, Q*. What this means is that there is still some DWL, albeit a lower
level than without regulation at all.
13.7.4 Hidden information

As we have seen so far, regulation of a natural monopoly is difficult and the govern-
ment’s ability to improve welfare is not guaranteed. This situation is only worsened
when we consider the fact that the government usually has limited information about
the market, costs and technologies. Furthermore, the monopolist is more likely to be
better information about these factors, but it is not in its interest to reveal this inform-
ation to the government. For example, if the government is proposing to implement
Pr
DWL ATC
P* MC
D
Q
Qr Q*
FIGURE 13.12 Under average-cost price regulation, the government sets the monopoly price at P = ATC.
However, the monopolist will produce less than the efficient quantity, so there is still some dead-weight
loss equal to the size of the shaded area. However, note that regulation decreases the area of dead-
weight loss, relative to no regulation at all
marginal cost pricing, it is in the interest of the monopolist to convince the government
that its marginal cost is high, even if its marginal cost is actually low. That way, it
gets to charge a higher price and make a profit.
In this chapter, we looked at a market in which there were many buyers but only one
seller – that is, a monopoly. We first examined what would happen in the market if
the monopolist was permitted only to charge a single price to all consumers. In such
cases, a profit-maximizing monopolist raises price above marginal costs; consequently,
there is a loss in welfare arising from the lower quantity traded in the market. We then
considered instances in which a monopolist could increase its profits further by
engaging in price discrimination. Finally, we looked at the case of natural monopolies
and how they might be regulated or controlled by government.
Notes
1 The supply curve show the quantity that a firm will supply, given a certain price – this
thought experiment is only valid if a firm assumes that it cannot influence the market price.
However, since the monopolist is a price maker, it does not take price as a given.
2 However, to use this shortcut, we will first need to ensure that the equation of the demand
curve is expressed with P as the subject.
3 Here, after paying the fixed fee F and the per unit price p, the consumer has a net CS = 0;
the consumer is actually indifferent between buying the product and not consuming at all;
hence, with first-degree price discrimination, the monopolist captures all of the surplus
available from trade. What is critical here is that a fixed fee F cannot be more than the
anticipated CS a consumer will accrue once having access to the product. If it is, the
consumer will not buy the product at all.
4 If a monopolist has increasing MCs, the same key relationships between MC and MR in
both markets holds, just care is needed to take into account that any increase in output in
one market increases the marginal cost of production in the other market. We leave this
issue aside here to focus on the key intuition of the problem.
C H A P T E R
14
Monopolistic
competition
14.1 Introduction
Our discussion of markets so far has focused on two extreme cases: first, we
looked at competitive markets, where there were many firms each of whom
have no market power; second, we looked at a monopoly, where there was
just one seller. In reality, most markets do not fall at either end of these
extremes.
In this chapter, we will look at the case of monopolistic competition. In
a monopolistically competitive market, there are many firms selling slightly
differentiated products. Product differentiation means that there is no perfect
alternative to any one product available for sale in the market. Because each
firm has a unique product, each seller has some market power in that they can
raise their price and not have the quantity demanded for their product fall to
zero. This means that firms in monopolistic competition are price makers, not
price takers like competitive firms. For example, restaurants sell different types
of cuisine and have different menus, meaning that each restaurant has some
degree of power to set its own prices. Similarly, convenience stores may be
differentiated by their locations; a convenience store one hour away is not quite
the same as a convenience store just down the street. As a result, each
individual convenience store has some degree of market power and acts like
a mini-monopolist in their local market.
14.2 Characteristics of monopolistic competition
A monopolistically competitive market has the following characteristics:
1 Many buyers and sellers. This implies that no producer has complete control
over the price, because buyers can always switch to other sellers.
2 Product differentiation. Each firm sells a slightly differentiated product. This
includes differences in the good or service itself, as well as differences in the
location of the firm and other factors that may influence the benefit (the convenience
for example) of buying from a particular firm.
3 Free entry and exit. Firms can freely (that is, costlessly) enter and exit the market
in the long run. In other words, there are no barriers to entry in the long run.
14.3 The short run
In a monopolistically competitive market, each firm sells a slightly differentiated

product. Because a firm’s product is slightly different from that of its competitors, it
has some control over the price it charges. In other words, each firm in a mono-
polistically competitive market faces a downward-sloping demand curve; that is, if the
firm raises its price slightly, there will be some drop off in the quantity demanded, but
that quantity demanded does not necessarily fall to zero. This means that a firm in
monopolistic competition is a price maker, not a price taker like a firm in a competitive
market. Consequently, a monopolistically competitive firm sets a profit-maximizing
price (or quantity) in the same way that a monopolist does, as described in detail below.
Consider, for example, the case of the restaurant market in an inner-city area. For
a particular consumer, there is one Thai restaurant that she likes best – it is her favourite
place to eat. If that Thai restaurant raises its prices slightly, this customer might still
choose to eat there, even though there are plenty of other options. That is, this Thai
restaurant can put its prices up and not have the quantity demanded for its product
drop to zero (although it might decrease somewhat). As noted, a competitive firm has
no scope to raise prices – any increase in price will result in the quantity demanded
for its product to fall to zero. This is because consumers can never be induced to pay
over the market price because all goods are identical – all goods are perfect substitutes.
This is not the case in monopolistic competition. While different goods might be similar,
they are not perfect substitutes, and it is this product differentiation that gives
monopolistically competitive firms market power.
Let us now consider the profit-maximizing choice of a firm in monopolistic
competition. In the short run, the number of firms in the market is fixed. This is because
each firm faces a fixed cost of production, which constrains the ability of firms to enter
and exit the market in the short run. From the perspective of an individual firm, this
means that the number of its rivals is fixed in the short run; hence, its short-run demand
curve is derived for a given number of firms currently in the market.
Monopolistic competition 125
MC
ATC
Pn
MR = MC
MR D
q
qn
FIGURE 14.1 A firm in a monopolistically competitive market in the short run. The firm maximizes its
profits by setting MR = MC. Thus, the firm sells q n units of its product, at a price of P n
The decisions of a monopolistically competitive firm are depicted in Figure 14.1.

The firm faces a downward-sloping demand curve, D. From this, we can determine
the marginal revenue curve (MR), which lies below the demand curve.1 The firm also
has some total cost curve, from which we can derive marginal cost (MC) and average
total cost (ATC). The firm maximizes profit by setting MR = MC.2 The intersection of
MR and MC is associated with the quantity qn and the price P n.
Note that this is essentially the same as the problem faced by a monopolist. Indeed,
because the firm sells a unique (i.e. differentiated) product, it acts as a mini-monopolist
in the market. Hence, in the short-run, a firm in a monopolistically competitive industry
can make an economic profit (or loss), depending on whether the price charged is higher
(or lower) than average total cost.
14.4 The long run
In the long run, there is free entry and exit in the market. Like perfect competition,
firms will enter the market if they believe that they can make positive profits by doing
so and firms will leave the market if they are sustaining losses.
14.4.1 Effect of entry and exit on demand curves

If a new firm enters the market, this will affect the demand curves of all incumbent
firms in the market. There will be two distinct but related effects:
• There will be a decrease in demand for the products of incumbent firms, as the
new firm’s product offers consumers an alternative. This will shift the demand
curve to the left.
• The demand curve for the products of incumbent firms will become more
elastic. This represents the fact that if an incumbent firm raises its prices, it is likely
to lose more consumers as there are more alternative products for consumers to
switch to.
For example, suppose a new laundromat opens up in a suburb neighbouring that of

an existing supplier. Some customers may switch from the old laundromat to the new
one because it is closer (the first effect). Moreover, customers that stay with the old
laundromat will now be more sensitive to changes in its price (the second effect); if
the old laundromat were to raise its price by, say, one dollar it would be likely to lose
more customers now that there is a nearby alternative.
Conversely, if a firm exits the market, this will affect the demand curves of all firms
that remain in the market. Namely, there will be an increase in demand for all remaining
firms and the demand curve of those firms will also become less elastic.
14.4.2 Elimination of profits and losses

As in perfect competition, the free entry and exit of firms will eliminate all profits (and
losses) in the market. However, under monopolistic competition, this occurs because
the entry and exit of firms affects the demand curve of all other firms in the market:
• The entry of firms in the market decreases demand for other firms, thus lowering
the price those firms can charge and hence lowers their profits.
• The exit of firms from the market increases demand for other firms, thus increasing
the price those firms can charge and hence increases their profits.
Figure 14.2 illustrates a monopolistically competitive firm in the long run. Notice that
the firm still sets MR = MC in order to maximize its profits. This determines the quantity
traded by the firm (qn) and the price that the firm charges (P n). However, the demand
curve (and hence the MR curve) is determined by the entry and exit of firms in
the market, such that at the quantity traded (qn), price is equal to average total cost
(Pn = ATC) – or, in other words, profit is equal to zero. For both conditions hold
in the long run – that is MR = MC and P = ATC – it must be the case that the ATC is
just tangential to the demand curve at qn (the quantity where MR = MC for the firm).
If the ATC cuts the demand curve, there is some quantity of output for which the firm
can make a positive profit – this is inconsistent with our assumption of free entry
causing zero profits in the long run. Similarly, if ATC always lies above the demand
curve, the firm would make a loss at any level of output, and would leave the industry
in the long run. The only possibility for profit-maximizing by the firm in monopolistic
competition (MR = MC) and zero profits (P = ATC) for a q n > 0, is for the ATC
curve to just touch the demand curve (and not cross it) at q n. Furthermore, note here
that zero profits means that the firm is making a positive mark-up on all units it sells
(P > MC) but these variable profits (that is, revenues minus the variable costs of
production) only end up just covering the firm’s fixed costs. For example, a convenience
Monopolistic competition 127
MC
ATC
Pn
MR = MC
MR D
q
qn
FIGURE 14.2 A firm in a monopolistically competitive market in the long run. Because of the entry
and/or exit of other firms in the market, this firm’s demand curve (and marginal revenue curve) has
shifted such that, at the quantity associated with MR = MC, the firm is making zero profits
store might source a litre bottle of milk for $2 and sell it for $5 (P > MC), but at the
end of the day, all those revenues are exactly equal to all the firm’s costs (variable and
fixed, assuming the firm is in the market).3
14.5 Welfare under monopolistic competition

We have now established that, in the long run, firms in a monopolistically competitive
market trade at the point where Pn = ATC. Note that, at this point, ATC > MC, which
means that average total cost (that is, production costs) is not minimized.4
This also implies that P n > MC. Hence, there is a DWL associated with each firm’s
output, as there are gains from trade that are not realized. This is analogous to the DWL
that arises from a monopoly.
However, there are two additional factors that may impact upon welfare, which
cannot be depicted in the diagram:
• Business stealing. A firm entering the market is only concerned with the amount
of profit that it can make in the market. It does not account for the fact that its
entry takes customers away from incumbent firms. Causing a consumer to switch
between firms does not necessarily increase surplus, but it does mean that the
economy has to bear another firm’s fixed costs of production. This suggests that
the number of firms in the market is too high.
• Product variety. On the other hand, a firm entering the market offers additional
differentiation in the market. Product differentiation can increase consumer surplus,
because a greater variety of products means that the market can better cater to the
individual tastes of different consumers. This increase in consumer surplus does
not factor into a firm’s decision of whether to enter the market, which tends to
suggest that the number of firms in the market is not high enough.
It is unclear which of these two effects will dominate, meaning that there may be too
many or too few firms in a monopolistically competitive industry.
Monopolistic competition sits between the two extremes of perfect competition and
monopoly. A firm in monopolistic competition is a price maker – like a monopolist –
using its market power to set price greater than marginal cost. However, like in a
competitive industry, there is free entry and exit in the long run, driving profits over
the longer term to zero. The welfare implications of monopolistic competition are
somewhat ambiguous. In the long run, because firms set a price greater than marginal
cost and earn zero profits, there will be some DWL generated, and each firm will not
be producing at a level of output that minimizes their costs of production. However,
any welfare analysis really should take into account the fact that there is product
differentiation. This means that, for a given industry, there could be too many firms
(if the business steal effect dominates) or too few (if the product variety effect is more
important).
Notes
1 See Chapter 13 for how to calculate the equation of the marginal revenue curve.
2 See Chapter 13 on why this maximizes profit.
3 We are, admittedly, being somewhat loose with our use of the term fixed cost here as in
the long run there are no fixed costs – we are assuming the firm is producing in the long
run, and given it is in the market it will have some variable and some invariant (fixed)
costs, like the rent it has to pay for its shop. The intention here is to emphasize that even
though a monopolistically competitive firm has a price greater than its marginal cost, it
ends up making zero economic profits.
4 Recall from Chapter 7 that the marginal cost curve cuts the average total cost curve at its
minimum. If ATC > MC, this means that the ATC curve has not yet reached its minimum.
C H A P T E R
15
Oligopoly
15.1 Introduction
We now turn to the case of oligopoly, a market that contains a small number
of firms. Because there are only a handful of key producers in the market, the
decisions of each firm have ramifications for not only itself but also for each
of its competitors. For example, if Dell adjusts its price for one of its laptops,
this will generally have an impact on its competitors, such as HP. Similarly,
if a department store decides to advertise, it might be able to increase its
customer base at the expense of other firms. Given the impact oligopolists have
on one another, a firm’s strategic choice – be it price, its output, whether it
introduces a new product and so on – will typically depend on what other firms
in the market are doing. For instance, a brewer might consider dropping its
price following a price reduction from a major beer manufacturer in the same
market. In the soft-drink market, following the introduction of a new energy
drink by Pepsi, Coca-Cola may choose to respond with its own alternative. In
a similar way, if Samsung introduces a new phone handset, Apple will consider
what it should do regarding a new version of its iPhone. A pharmaceutical
company will consider what new drugs its rivals are trying to develop (and
those that they already have patents for) when considering its own research
and development programme. This strategic interaction between firms is
a key feature of oligopoly; moreover, this sort of strategic interaction is absent
in other market structures we previously studied (perfect competition,
monopoly and monopolistic competition).
We model strategic interaction in oligopoly using game theory. We have
previously discussed the core concepts and tools of game theory in Chapter
3. In this chapter, our goal is to illustrate how game theory tools can be applied
in the context of an oligopoly. In doing so, we will highlight a few key
examples of how firms in an oligopoly strategically interact with each other,
but our examples will by no means be an exhaustive demonstration of what game theory
can tell us in this context. We urge you to familiarize yourself with the contents of
Chapter 3 before embarking on this chapter.
15.2 Characteristics of an oligopoly
Oligopolies have the following characteristics:
1 Few seller and many buyers. Output in the market is produced by a handful
of firms.
2 Price maker. Because there are only a small number of firms in the market, each
firm retains the power to set its own prices.
3 Barriers to entry. Entry into the market is difficult, as there are high barriers to
entry.
4 Product differentiation. Products may be differentiated or not, depending on
the market.
15.3 Simultaneous move games
Often firms will need to make strategic decisions without knowledge of what other
firms in the market have decided to do. For example, the board at Pepsi might meet
and decide on the possible introduction of a new energy drink without knowing what
the board decided at Coke. In other cases, firms choose not to share information with
each other; often the law prohibits the sharing of that information (for example, laws
against collusion prevent firms sharing price information). In such circumstances, firms
make decisions as though their choices were made simultaneously, in the sense that
they do not have knowledge of other firms’ decisions. In such cases, it will be
appropriate to analyse the strategic interaction of those firms as a simultaneous move
game.1
15.3.1 Price wars: an example of a prisoner’s dilemma

In some cases, the ‘game’ faced by the firms in an oligopoly might resemble a prisoner’s
dilemma. As an illustration, consider the pricing game outline in Chapter 3.2 In that
example, two firms faced the choice of setting their price high (pH ) or low (pL ) without
knowledge of what the other firm has chosen. If the firms ‘cooperated’ and set their
prices high, both would receive a profit of $4.3 However, if one firm chooses pL
undercutting its rival who chose pH , the payoffs are $5 and $1 respectively. If both firms
set their prices low, they would each receive a payoff of $3.
The normal form of this game is in represented in Figure 15.1. This game is a
prisoner’s dilemma because (a) it has a dominant strategy equilibrium in which both
Oligopoly 131
Player 2
pH pL
pH (4,4) (1,5)
Player 1
pL (5,1) (3,3)
FIGURE 15.1 Price-setting game in an oligopoly
players choose pL, and (b) this equilibrium does not maximize surplus as both players
would receive greater profit if they both chose pH . Note here that the firms are col-
lectively worse off in the Nash equilibrium than they would be if they could somehow
cooperate (tacitly collude) and both price high. This is not possible however as it is
individually rational for each firm to opt for the low price. This example is a demon-
stration that in oligopoly, the competitive interaction between rivals often makes it
difficult for firms not to undercut one another, reducing industry profits. Moreover,
this suggests that a very competitive outcome (and corresponding level of social
welfare) is possible with relatively few firms in the market.
Another example of a prisoner’s dilemma in an oligopoly is the decision of two rival
firms whether or not to advertise. If neither firm advertises, there will be no effect on
customer purchasing behaviour. If one firm advertises and the other does not, the first
firm may steal some customers from the second. However, if both firms advertise, the
advertising campaigns will cancel each other out and there will be no effect on
customer purchasing behaviour. Suppose the cost of advertising is $100, and if only
one firm advertises, it can steal $200 worth of business from the other firm. In this
case, the normal form of this game is represented in Figure 15.2. Again, the dominant
strategy equilibrium is one in which both players advertise, but both players would be
better off if neither chose to advertise.
Firm 2
NA A
NA (0,0) (–200, 100)

Firm 1
A (100, –200) (–100, –100)
FIGURE 15.2 Advertising game in an oligopoly. In this figure, ‘A’ represents the choice to advertise and
‘DA’ represent the choice to not advertise
15.3.2 Can firms avoid the prisoners’ dilemma?

As discussed earlier, one defining characteristic of a prisoner’s dilemma is that its
equilibrium does not maximize profits (we are now focusing solely on the firms and
their profits, and ignoring consumers). Indeed, in the examples above, both firms would
prefer they both charged a high price or did not advertise, but given their incentive
structure they face, neither firm can credibly commit to doing so.
What if the firms are allowed to talk to each other before they set prices (ignoring
the fact that this sort of communication is illegal in most jurisdictions)? This is often
called pre-game communication. But in the case of a prisoners’ dilemma scenario,
as in the pricing game above, pre-game communication is of little use. To see this,
assume the CEOs of each firm meet the day before they need to decide on their
respective prices. Each CEO might swear they are going to price high – after all, they
both understand the structure of the game, and if both firms do stick to this agreement
both will be better off.
But after leaving the meeting, what are the incentives of either firm. If Player 1
assumes that Player 2 will price high, as agreed, Player 1 is better off reneging on the
agreement (‘cheating’) and pricing low. If, on the other hand, Player 1 thinks Player
2 will cheat, then she is also better off pricing low. The structure of the game determines
the incentives for each player, and the pre-game communication has not altered these
incentives (both firms have a dominant strategy to price at pL ). Anything that is said
pre-game that does not involve a firm pricing low in this case is simply not credible.
With a prisoners’ dilemma pre-game communication is of no value.
The key reason pre-game communication does not help, is it has no commitment
value. To overcome this, firms need to find a way to make a binding commitment
that they cannot break. For example, if firms lobby the government to institute laws
that stipulate minimum prices or restrict advertising, this can provide a binding commit-
ment that prevents firms from lowering their prices or increasing their advertising
(in the examples above), thus avoiding the prisoner’s dilemma. A binding commitment
can help firms avoid a prisoners’ dilemma because, by tying the hands of the firms,
they are unable to choose what they would have otherwise done (price low, advertise
and so on) if they were free to choose.
Another way firms might be able to get out of their prisoners’ dilemma trap is if the
firms compete in the marketplace not just once, but many times. Pepsi and Coke, for
instance, do not just play a price-setting game once; rather, they set prices this week,
then in the next week, then in the following week, and so on. Firms in these situations
are playing a repeated game, where they play the same game multiple times. Import-
antly, this means a firm can choose its actions based on what its rivals did in the past
– that is, a firm can punish (or threaten to punish) a rival that does not cooperate.
Repeated interaction does not, in of itself, automatically guarantee that the firms will
cooperate. Let us first consider whether this will make any difference when the game
is repeated a finite number of times. To illustrate, let us say that the pricing game above
is played twice. In the first round, the firms will think forward and try to anticipate
what will happen in the second round. In the second round, the firms are playing the
Oligopoly 133
game for the last time, with no further rounds to play. Hence, in the second round, the
incentives of the firms are the same as when they are playing the game only once;
regardless of what happened in the first round, the dominant strategy of both firms is
to set price at pL. Thus, in the first round, both firms know that the outcome in the
second round is a fait accompli; no matter what they choose in the first round, they
cannot induce cooperation in the second round. Therefore, the incentives are as though
they were playing a single round of the game, so in the first round both parties have
a dominant strategy of pricing at pL. This analysis can be extended to include any game
that is repeated a finite number of times: firms have no incentive to cooperate in the
last round, therefore firms have no incentive to cooperate in the second-to-last round,
therefore firms have no incentive to cooperate in the round before that, and so on.
As you may have noticed, the reason the firms cannot cooperate is because there is
a definite end to the game. In the last round, the actions of each firm have no effect
on future rounds (as there is none), and this sets off the chain-reaction of non-
cooperation backwards through every period. This suggests that it may be possible for
firms to cooperate if there is no definite end to the game – that is, if the game is repeated
infinitely or when there is no definite end to the game.4
Consider a game with no definite last period (such as a game with an infinite number
of periods), in which two firms can either price high (pH ) or low (pL ) in every period.
There are two classic strategies that firms could adopt (these are by no means the only
possibilities, of course). The first is a trigger strategy. With a trigger strategy, a firm
starts the first period by setting a high price. Then, in the second period, if both firms
set a high price in the first period, a firm again sets a high price. If not – that is, their
rival set a low price in the first period – the trigger is flicked, and the firm now set a
low price in the second period, and in all future periods. In other words, once its rival
‘cheats’ and sets a low price, its rival punishes it by setting a low price in every future
period. This is an extreme version of punishment. If this future cost of punishment is
large enough, a firm might not be tempted to ‘cheat’ a set a low price today – the threat
of punishment allows firms to set high prices (or to ‘tacitly collude’). In many ways,
a trigger strategy is one of the most severe punishments a firm can (credibly) impose
on their rivals – once someone cheats at any point, they are punished forever.
The other classic strategy is known as a tit-for-tat strategy. With a tit-for-tat
strategy, a firm starts by setting a high price in the first period, then in every subsequent
period it sets a price equal to its rival’s price in the previous period. That is, if their
rival priced low in the most recent period, a firm will price low in the period;
alternatively, if its rival set a high price last period, a firm will copy this and set a high
price in the current period. This works as a deterrent because any undercutting to a
low price will be followed by a rival also choosing to price low. Moreover, to get out
of a low-price price war, a firm will have to unilaterally raise its price above its rival
(which will cause a loss in profits). Again, a tit-for-tat strategy can help firms maintain
higher prices. It works if the firms think the future cost of being punished from
‘cheating’ and pricing low are larger than the short-term gains.
The tit-for-tat strategy also has some of the price dynamics that we observe in some
real markets. For example, in some markets we see all firms set a high price for a while,
before one firm undercuts. All firms follow by also reducing their price, and a price
war ensues. The price war (with low prices) might continue for a period of time, before
one firm raises its price again. Having seen this price rise, other firms follow suit and
raise their own prices. This pattern continues, giving a price dynamic of periods of
high prices, followed by price wars, which is then followed by another period of high
prices, and so on.
15.4 Product choice: an application of a coordination game
Oligopolists often need to make a decision about what type of product they should sell
– should a car manufacturer concentrate on high-end luxury vehicles or should it
produce family cars? Should a firm developing a new breakfast cereal aimed at the
health-conscience consumer, or make a sugar-filled cereal designed for kids. At other
times location is important. Two fast-food restaurants might need to decide on the
location for their respective outlets. Sometimes software and hardware developers need
to decide which technology or platform to use. In many of these cases, these choices
by oligopolists can be modelled using a coordination game.
Consider, for example, two firms deciding where to set up a restaurant. Suppose
there are two possible locations, A and B. If the firms set up restaurants in different
locations, each will get all the customers at that location; however, if the firms set up
business in the same location, they will have to share the customers at that location.
The normal form of this game is represented in Figure 15.3. In this game, there are
two Nash equilibria: (A, B) and (B, A). The parties would like to coordinate their
actions to ensure that they do not both set up business at the same location; in this
example the two firms would like to set up their respective restaurants in different
places. In terms of firms choosing their products, this model suggests that in this case
firms prefer to accentuate product differentiation. In terms of our restaurant example,
locating in different areas might allow firms to soften price competition between the
two rivals, allowing them to earn greater profits from their respective captive markets.
This model suggest firms want to locate apart. Note, given that there are multiple
equilibria, while the model informs us of the Nash equilibrium, it does not tell us which
Firm 2
A B
A 1,1 (1, 1)
2 2
Firm 1
B (1,1) 1,1
2 2
FIGURE 15.3 Location-choice game in an oligopoly

Oligopoly 135
Firm 2
X Y
Firm 1 X (30, 50) (0,0)
Y (0,0) (60, 20)
FIGURE 15.4 Platform choice for game developers
firm will locate where. That is, it does not tell us which Nash equilibria will be the
outcome in the market.5 Moreover, firms who find themselves in a coordination game
like this will be very keen to avoid coordination failures, which here would mean that
both firms locate in the same place (either both at A or both at B).
Consider now an alternative product-choice game, in which two firms 1 and 2 are
each developing a new computer game. Each firm simultaneously chooses to use
platform X or platform Y. If both firms choose X, Firm 1 receives a payoff of 30 and
Firm 2 a payoff of 50. If both firms choose platform Y, Firm 1 receives a payoff of 60
and Firm 2 receives 20. If Firm 1 chooses Y and Firm 2 opts for X, both firms earn 0.
Similarly, if Firm 1 opts for X and Firms 2 for Y, both firms earn 0. This coordination
game is illustrated in Figure 15.4.
There are two Nash equilibria in this game: (X, X) and (Y, Y). In this game, the two
developers want to have similar products – they want to have minimal product
differentiation. Here, if one firm is choosing X, the best-response for the other firm
is to also choose X. Similarly, Y is the best response to a rival’s choice of Y. There
could be many reasons for this. For example, having games using the same platform
might increase the size of the market by making the products of both firms compatible
or easier for the consumer to use. More broadly, we see examples of minimal product
differentiation in a variety of markets: sometimes restaurants are all located together;
technology companies often want to use the same platform or technology; new phone
handsets from different companies tend to be very similar in look and functionality;
and political parties often seem to offer voters very similar policies.
Overall, this suggests that sometimes firms in an oligopoly prefer to offer products
that are different from their rivals (product differentiation) but this is not always true;
in many markets firms choose to offer products that are very similar to those of their
rivals (minimal product differentiation). Which case applies depends on the specifics
of the market of interest.
15.4.1 Can firms avoid coordination failure?

As noted, in a coordination game there is the possibility of a coordination failure. For
example, in the restaurant example above, the two firms want to locate apart from one
another, not together. There are several ways in which firms can avoid coordination
failure. First, the parties might engage in pre-game communication, as outlined
above. Take the game illustrated in 15.3. In this case each player might communicate
before choosing their actions and come to an agreement as to what they will do. Note
that, once an agreement is reached, no party has an incentive to deviate from that
agreement, because the agreement involves the firms playing a Nash equilibrium.
Unlike the prisoners’ dilemma above, in which both players have an incentive to renege
on any deal, with a coordination game, as the parties are communicating that they will
play a strategy that is part of a Nash equilibrium, the pre-game communication is
credible.
Another way that firms may coordinate with each other might be through customs
or social norms. To understand how this might help, imagine that the firms are ice
cream trucks that decide where to set up business at the beginning of each day. Over
time, a custom or norm that Truck 1 goes to location A and Truck 2 goes to location
B might develop. In this case, there is an understanding or implicit agreement between
the parties as to how they will act.
15.5 Sequential games
So far we have consider simultaneous games in which firms take their actions without
observing what their rivals have chosen. But in many economic and business situations,
one player (or firm) takes their action and their choice is observed by others before
they themselves choose what to do. For example, one firm, Firm A, might choose to
build a small or a large factory. This choice is observed by Firm B, who then decides
on their own factory size. This sequencing of actions has important implications; it
means that the second-movers take make their choices knowing exactly what the leader
did. Moreover, the leader knows that when they make their choice, the followers will
observe what they have done, and react according. This can have important economic
implications. We model these situations now using sequential games.6 This allows us
to focus on several key game theoretic insights for business strategy, including first-
and second-mover advantages and the value of commitment.
15.5.1 Credible threats

As we saw in Chapter 3, some Nash equilibria are sustain by non-credible threats that
a player would not actually implement if they were ever actually called upon to choose.
Given players are forward-looking and rational, they will do their best to anticipate
what the other players will do in the future given the situation they find themselves
in. We expect rational profit-maximizing players will not react to non-credible threats,
so we are really interested in thinking about equilibria in oligopoly markets that are
credible. To eliminate Nash equilibria that rely on incredible threats, we solve back-
wards, looking for the subgame perfect equilibria. In a subgame perfect equilibrium,
every player’s strategy is a Nash equilibrium in every subgame. What this really means
Oligopoly 137
(–5,5)
E (10,10)
(0,20)
FIGURE 15.5 Entry game in an oligopoly
is; we are only interested in equilibria in which all players would act enact what their
requires in every part of the game, even if that part of the game tree is not reached in
the outcome of the game. As noted in Chapter 3, solving backwards is our modelling
technique to capture the fact that players can look into the future and anticipate what
the other players will do.
In the oligopoly context, focusing on credible Nash equilibria is important. As an
example, let us revisit the entry game first outlined in Chapter 3. In that example, there
was an incumbent firm and a potential entrant. The entrant first chooses whether or
not to enter the market. After observing this decision, the Incumbent decides whether
to accommodate or punish the entrant. If the Entrant does not enter, the Entrant
receives a profit of $0 and the Incumbent receives $20. If the Entrant enters and the
Incumbent accommodates, each firm makes a profit of $10. Finally, if the Entrant enters
and Incumbent punishes, the profits are –$5 and $5 to the Entrant and the Incumbent,
respectively. The extensive form of this game is represented in Figure 15.5.
As we have previously found, this game has two Nash equilibria: (Enter, Accom-
modate) and (Not enter, Punish). However, only the subgame perfect equilibrium
(Enter, Accommodate) is credible. In the equilibrium (Not enter, Punish), the potential
entrant chooses not to enter because the incumbent has threatened punishment;
however, this threat of punishment is not credible because if the potential entrant
actually entered, it would not be in the best interests of the incumbent to choose Punish
– rather, the Incumbent is better off choosing Accommodate. In other words, the threat
of punishment is not credible.
We can see this solving backwards for the subgame perfect equilibria. Considering
what the Incumbent would do first, if the game ever reached that point, the Incumbent
prefers to accommodate; $10 beats $5. Given this, the Entrant expects the Incumbent
to Accommodate if ever they have a choice, so the Entrant’s choice is to not enter the
market and receive a payoff of $0, or enter, and receive a payoff of $10. Thus, we
predict that entry will occur because the Incumbent will Accommodate once the
Entrant has entered the market.
This entry game is, of course, one example. But it makes a general point: using
subgame perfect equilibria is a powerful tool in oligopoly markets when actions are
taken sequentially by firms – it allows us to focus on credible outcomes, which is

appropriate for forward-looking and rational profit-maximizing firms, who will try their
best to anticipate the actions of the other players in the future.
15.5.2 Commitment
Normally we think that having more choices is better, but this is not always true. In
business it might be advantageous for a firm to have its choices limited if this provides
the firm with a strategic advantage.
To illustrate the value of commitment, consider the following (possibly half-true)
story from the Dark Ages. Let us assume that there is a boat of Viking raiders that
land on a beach in a strange land. After landing, the chief of the boat has a choice: the
Vikings can either Burn their boat or Not Burn their boat. Following this decision the
Vikings wander up the beach to the nearest village. At this point the Viking can either
Fight or Not Fight (we are suppressing the actions of the villagers to keep the model
as simple as possible).
The payoffs in the game are as follows. If the Viking raider Burn and Fight they
get 200 and the local villagers get 0. If the Viking chief opts to Burn and Not Fight
the Viking and the locals both get 100. If the Vikings choose Not Burn and then Fight,
the payoffs are 50 and 150 to the Vikings and the villagers respectively. If the choices
are Not Burn and Not Fight both parties get a payoff of 100. The extensive form of
this game is illustrated in Figure 15.6.
(200,0)
B1
(100,100)
A
(50,50)
B2
(100,100)
FIGURE 15.6 Viking invasion game; the Viking chief must decide whether to Burn or Not Burn their boat
What should the Vikings do? To consider their best option, solve backwards. If the
Vikings had opted to Burn their boat, Fight yields 200, while Not Fight yields 100;
Fight is the best option. If, on the other hand, the Viking chief did Not Burn their boat,
Fight yields 50 while Not Fight provides them with 100; Not Fight is the best option
following Not Burn. So, working backwards and anticipating their future credible
Oligopoly 139
actions, the choice is Burn (then Fight) for a payoff of 200, or Not Burn (then Not
Fight) for 100. The Viking chief will Burn the boat.
Burning the boat is the best strategy, even though it reduces the Vikings’ options
later (they cannot sail home if things start going badly during the battle, for example).
However, it does provide the Vikings with credibility that they will fight hard: the
villagers know that without a ship, the Vikings will Fight. It is the destroying of an
option (the option to use the boat) that makes the threat of fighting credible, and this
makes the Vikings better off.
What is the general point of this story? Sometimes having less choices can make a
player better off because it commits them to a certain (difficult) action – this is the
value of commitment. This can be true for firms in strategic situations, like in oligopoly
markets. For instance, if a firm takes an action that it cannot reverse, it has made a
commitment; its rivals will observe this commitment and have to adjust to it
accordingly. This can give a firm a huge advantage.
In the fictitious game outlined here, the Vikings burning their boat was their
commitment – given they had no exit plan once their boat was gone, the villagers would
know the Vikings were going to fight hard, and would most likely surrender more
easily. Often in an oligopoly market, a firm can increase its profit if it can make a bind-
ing commitment. For firms in a strategic environment, a credible commitment might
involve building a large factory or hotel; these are not things that can be easily altered
making them credible commitments. And once a firm has made its commitment, its
rivals will observe their choice and react accordingly (by building a smaller factory,
by exiting the market and so on).7
Note here, commitment has value when it allows a firm to credibly take an action
that it would not otherwise be able to do. A large factory might commit a firm to
produce a large level of output, whereas a firm that could adjust its factory size after
having observed what its rivals did might choose a lower level of output – but in this
case the flexible factory provides no commitment value at all. Commitment requires
that the firm has its hands tied – it cannot take back its choice and decide again. But
this inflexibility – or lack of choice – can make the firm better off in the end, because
in strategic environments its rivals will react.
15.5.3 First-mover advantage

Sometimes in sequential games, it is better to be the leader – that is, there is a first-
mover advantage. As noted above, if the first mover (the leader) can commit to its
action, the follower firms must then adapt their actions to what the leader has done. For
example, it might be advantageous to be the first developer to build a hotel in a new
resort area; alternatively, the leader gets to choose the technological standard that it
prefers. In this section we consider several examples where there is a first-mover
advantage.
Consider another coordination game outlined in Figure 15.7. In this simultaneous
game there are two firms, 1 and 2, who can choose between two types of technologies,
B and N. The payoffs are as outlined in the figure. In this game there are two Nash
Firm 2
B N
B (50,30) (0,0)
Firm 1
N (0,0) (30,50)
FIGURE 15.7 Simultaneous technological choice
equilibrium – (B, B) and (N, N). But noted that the preferences of the two firms are
asymmetric; Firm 1 prefers the (B, B) equilibrium while Firm 2 gets a higher payoff
in the (N, N) equilibrium.
Now augment this game so that Firm 1 gets to make its choice first. Following this,
Firm 2 observes the action of Firm 1 before making its own choice. The payoffs are
the same as in the simultaneous-move game above. This sequential version of the game
is illustrated in Figure 15.8.
(50,30)
F21
(0,0)
F1
(0,0)
F22
(30,50)
FIGURE 15.8 Technological choice game with a first-mover advantage
Solving backwards, let us solve for the subgame perfect equilibrium of this game.
First, assume that Firm 1 chose B. In this case Firm 2 can opt for B and get 30, or N
and get 0. So if Firm 1 chose B, Firm 2 will also choose B. Second, consider what
Firm 2 will do if Firm 1 opted for N. Now Firm 2 can get 50 from N but 0 if it chooses
B; following Firm 1’s choice of N, Firm 2 will also choose N. Working backwards,
when Firm 1 makes its choice of action it will take into account how Firm 2 will
respond. Given Firm 2’s responses, Firm 1 will get a payoff of 50 from opting for B
(as Firm 2 also chooses B) or 30 if it opts for N (as Firm 2 will respond by also opting
Oligopoly 141
Firm 2
E NE
Firm 1 E (–100, –100) (500, 0)
NE (0, 500) (80, 80)
FIGURE 15.9 Natural monopoly entry game
with N). Firm 1’s best option is B. As a consequence, the SPE outcome of the game
will involve Firm 1 choosing B, then Firm 2 choosing B.
In the simultaneous version of game there were two Nash equilibria. In the sequential
version, Firm 1 being the first mover essentially chooses the equilibrium that it prefers;
in a coordination game with asymmetric payoffs, there is a first-mover advantage.
Consider now another market-entry game illustrated in Figure 15.9. In the game two
firms, 1 and 2, simultaneously choose whether to enter a new market (E) or to not
entry (NE). If both firms enter, each gets a payoff of –100. If both firms choose NE,
each firm receives a payoff of 80. If Firm 1 chooses E and 2 opts to NE, the payoffs
are 500 and 0 to firms 1 and 2, respectively. Finally, if 2 enters and 1 does not, Firm
1 gets 0 and Firm 2 get a payoff of 500.
In this game – another coordination game – there are two Nash equilibria: (E, NE)
and (NE, E). Essentially, this market is a natural monopoly, which only has room for
one firm. If both enter, each firm suffers losses. An outcome when both firms do not
enter cannot be an equilibrium – each firm would prefer to deviate to be the sole entrant.
This means that we would expect one of the Nash equilibria to ensue – either Firm 1
or 2 will enter, and the other firm will stay out of the market.
Now augment this game to allow Firm 1 to make its choice first; Firm 2 then observes
this choice before taking its action. In a similar manner to above, let us solve for the
SPE by solving backwards. Here, if Firm 1 entered, Firm 2 will choose not to enter.
On the other hand, if Firm 1 opted for NE, Firm 2 would choose E. Working backwards,
Firm 1 will opt for E, and the SPE outcome will have Firm 1 entering the market and
Firm 2 choosing NE. Once again, being the first mover is advantageous; it allows Firm
1 to enter the market knowing Firm 2 will stay out and NE.
The two previous examples show that there can be a first-mover advantage. This
suggests that in situations like this, if possible, a firm should try to manipulate the
environment they are in to ensure that they are the firm that gets to be the leader.
15.5.4 A second-mover advantage

While the benefits of being a leader into a market are often emphasized, it is sometimes
the firms that are followers that do better. For example, Microsoft did not develop
Firm 2
I W
I (4,4) (2,6)
Firm 1
W (6,2) (1,1)
FIGURE 15.10 Free-riding game
the first operating system, and the iPhone was not the first smartphone. In many
situations, it is actually better to be a follower than a leader. In these games there is a
second-mover advantage.
As an example, consider the following game shown in Figure 15.10. In this game
the two firms can simultaneously choose to invest (I) or to wait (W). If both invest
each firm receives a payoff of 4. If they both choose W the payoff to each firm is 1.
If one firm chooses I and the other firm W, the payoff to the investor is 2 while the
firm that waited is 6. This game represents the situation in which a new product needs
to be developed, and this requires some investment costs. If both firms choose I they
share the development costs. If, on the other hand, only one firm chooses I it incurs
all the development costs while the firm that waited can come in after and free ride
on the other firm’s investment. This is the case in many markets where it is not feasible
for a firm to protect its innovations effectively with a patent or copyright laws.8
There are two Nash equilibria in the game: (I, W) and (W, I). In either equilibrium
one firm enters and the other waits. It is evident that it is better to be the second firm
in the market and waiting allows a firm to free ride on the leader’s investment. In a
market like this, there is a second-mover advantage; if it is possible, a firm would like
to commit to be the second mover. How such a commitment can be made is not
necessarily simple. Perhaps a firm can overcommit to undertaking other research
projects so that it does not have the capability to choose I immediately. If this is credible,
such an action could induce its rival to opt for I. This allows the overcommitted firm
to W, increasing its profit.
Another game in which there can be a second-mover advantage in a zero-sum game,
such as the one illustrated in Figure 15.11.9 A zero-sum game has a constant payoff
in all possible outcomes in the game; in the example here, in each possible outcome
the payoff is zero. Assume that the two firms simultaneously choose their possible
actions of either H or T. If both firms chose the same action (both H or both T), Firm
1 gets a payoff of 1 and Firm 2 gets –1. If the actions do not match (one firm chose
H and the other T), Firm 2 gets 1 and Firm 1 receives –1.
In this game there are is no Nash equilibrium; in any of the possible outcomes there
is always a firm that has a profitable unilateral deviation.10
Oligopoly 143
Firm 2
H T
Firm 1 H (1,–1) (–1,1)
T (–1,1) (1,–1)
FIGURE 15.11 A ‘matching pennies’ game
However, now consider if the game is played sequentially, with Firm 1 moving first
followed by Firm 2 where Firm 2 observes the action taken by 1. Now being the second
mover is very advantageous. Whatever Firm 1 does, Firm 2 can react optimally to
ensure that it gets the dollar.
In a strategic situation like this, the second mover is better off. Consider two
television stations, each with a new blockbuster show. It could be that the station that
holds off announcing when it will screen its big show can trump the station than locked
in its schedule first. (Of course, in situations like this, both firms will try to delay
finalizing their schedules as long as they can.)
In this chapter, we discussed how game theory might be used to analyse strategic
interactions in an oligopoly market. Importantly, the examples we have given are by
no means an exhaustive demonstration of how game theory might be applied in these
situations.
Several key implications for oligopoly markets were discussed. First, oligopolists
might find it difficult to not strongly compete with one another (by pricing low,
advertising and so on). While individually rational, competing hard can potentially
lower total industry profits, subsequently making each firm worse off. If they can, firms
would like to find ways not to compete so hard; one way this can occur is if firms
interact with each other, not once, but may times. With repeated interaction, it is
possible that the threat of future punishment can help firms to (tacitly) cooperate.
Second, when firms take their actions in sequence, profit-maximizing firms will try
to anticipate what their rivals will credible do in the future. Focusing on credible
outcomes requires that we solve the game backwards (looking for the SPE of the game).
In sequential games, there can be either first- or second-mover advantages, depending
on the economic environment firms find themselves in.
Notes
1 See Chapter 3 for more details.
2 A similar game can be devised when firms choose output levels.
3 Sometimes this sort of cooperative behaviour is referred to tacit collusion, meaning that
the firms act cooperatively (here pricing high) without necessarily explicitly communicating
and making an explicit agreement. If firms make an explicit agreement on price (or output),
this is collusion, which is illegal in countries like Australia, the United States and in the
European Union.
4 This might be the case when each firm knows the game will end, but they are never exactly
sure when that will happen. So, in any given period, firms think the game will continue
into the future with some positive probability.
5 We discuss this issue further below in the section on sequential games.
6 See Chapter 3 for more details.
7 It is worth mentioning that while we often think of a commitment by a firm being related
to being a large player in the market, there are times when a firm might want to make a
credible commitment to be small niche firm (with a small factory for example) in order to
commit to soft competition with its rivals.
8 Often there are costs of educating consumers about the benefits of a new product, which
then help any firm that enters and starts producing that good.
9 This game is an example of the classic ‘matching pennies game’.
10 There is actually a mixed strategy equilibrium in this game, when players put a positive
probability on playing more than one action. However, as mentioned in Chapter 3 we do
not focus on mixed strategy equilibria in this book.
PA R T V
Market failures

C H A P T E R
16
Price regulation, taxes
and subsidies
16.1 Introduction
In previous chapters, we analysed various types of market structures and the

different outcomes they produce. For the most part, we assumed that there
was no intervention in the market by the government. In this chapter, we
consider two common types of government intervention: (a) price regulation
and (b) taxes and subsidies. For the purpose of this chapter, we will assume
that the market is efficient in the absence of government intervention (that is,
the market is a competitive market). In this framework, we examine how
government intervention affects market price, the quantity traded and welfare
in the market.
16.2 Price regulation
One way that a government can intervene in the market is by regulating the
price of a good or service. Price regulation typically takes the form of a price
floor, where the government sets a minimum price at which a good or service
can be traded, or a price ceiling, where the government sets a maximum price.
16.2.1 Price floor

When a government puts in place a price floor, it sets a minimum price at
which a good or service may be sold. For example, the Australian government
has historically maintained a price floor in the market for wool. Currently,
there are laws that guarantee minimum wages for certain jobs, which is
essentially a price floor in the market for labour.
148 16 Market failures
P P
P*
D
Q
QD Q* QS
–
FIGURE 16.1 A price floor set at P
Figure 16.1 illustrates the effect of a price floor on market outcomes. In the absence
of any government intervention, the equilibrium price and quantity will prevail in the
market – that is, (Q*, P*). In order for a price floor to affect the market outcome, the
price floor must be set above the equilibrium price, P*. A price floor set below P* is
non-binding, because the equilibrium price is already above the minimum price (the
–
price floor). By contrast, a price floor set above P* (say, at P ) is binding because the
market equilibrium price is less than the minimum price at which the good or service
–
may be sold; hence, the price in the market will need to rise from P* to P in order to
meet the requirements of the price floor.
Raising the price above P* affects the quantity demanded and the quantity supplied
–
in the market. At P , the quantity demanded by consumers is QD, whereas the quantity
supplied by producers is denoted QS. This results in an excess of supply of QS – QD.
For example, suppose the government implements a price floor in the labour market
by setting minimum wage laws. This will result in excess supply of labour (that is, the
amount of labour supplied by workers exceeds the quantity of labour demanded by
firms). There will be unemployment; at the going wage rate, there are workers who
are willing to work who are not being employed by firms. Sometimes, a government
deals with the excess supply by purchasing the excess, QS – QD. In fact, this occurred
in the market for wool in Australia; over time the Australian government accumulated
a large stockpile of wool as a result.
Of course, the change in the market price has welfare implications, as depicted in
Figure 16.2. Consumer surplus decreases for two reasons: first, the price has increased,
meaning that consumers receive less surplus on each unit purchased; second, consumers
buy fewer units overall, meaning that surplus is lost through the decrease in the quantity
traded. On the other hand, producer surplus is affected by two countervailing effects:
first, producers sell fewer units overall, which decreases surplus; however, producers
receive more surplus on each unit sold, as a result of the increase in price. Overall, there
is dead-weight loss in the market. Note here, also, that this is the best possible outcome
Price regulation, taxes and subsidies 149
S CS
PS
DWL
P P
P*
D
Q
QD Q* QS
–
FIGURE 16.2 The welfare effects of a price floor set at P
in terms of welfare resulting from the price floor (that is, it is the smallest possible DWL)
as it is assumed that it is the lowest cost firms that are supplying the product. If this is
not the case, the loss of welfare from the price floor will be even larger.
16.2.2 Price ceiling

When a government puts in place a price ceiling, it sets a maximum price at which a
good or service may be traded.
Figure 16.3 illustrates the effect of a price ceiling on market outcomes. In the absence
of any government intervention, the equilibrium price and quantity will prevail in the
market – that is, (Q*, P). In order for a price ceiling to affect the market outcome, the
price ceiling must be set below the equilibrium price, P*. A price ceiling set above
P* is non-binding, because the equilibrium price is already below that price ceiling.
–
By contrast, a price ceiling set below P* (say, at P ) is binding because the equilibrium
price is greater than the maximum price at which the good or service may be sold;
–
hence, the price in the market will need to fall from P* to P in order to meet the
requirements of the price floor.
Again, pushing the price below P* affects the quantity demanded and the quantity
–
supplied in the market. At P , the quantity demanded by consumers is denoted by QD,
whereas the quantity supplied by producers is denoted QS. This results in an excess
of demand of QD – QS; in other words, at that price there are no enough units supplied
to meet the quantity demanded.
This raises the question of how the existing units of the good should be allocated
to consumers. There are several ways this issue of allocation might be addressed.
• Queuing. One way of rationing a limited number of goods among consumers is

on a first-come, first-served basis, which often results in queues. This way, only
consumers who are willing to wait receive the good.
P*
P P
D
Q
Qs Q* QD
–
FIGURE 16.3 A price ceiling set at P
• Discrimination between consumers. Another way of distributing the good is

for a government official to pick and choose which consumers should receive the
good. However, this may be problematic if the official distributes the goods on
the basis of nepotism rather than who values the good the most.
• Side payments. In either case, there is a possibility that consumers will make
side payments (that is, payments in addition to the price) to the firm or the
government official to guarantee access to the product. This is generally viewed
as underhand or corrupt behaviour, and may be illegal.
Again, the change in the market price has welfare implications, as depicted in
Figure 16.4. Consumer surplus is affected by two countervailing effects: first,
S CS
PS
DWL
P*
P P
D
Q
QS Q* QD
–
FIGURE 16.4 The welfare effects of a price floor set at P
consumers receive more surplus on each unit purchased, as a result of the decrease in
price; however, consumers purchase fewer units overall, meaning that surplus is lost
through the decrease in the quantity traded. Producer surplus decreases for two reasons:
first, that producers receive less surplus on each unit sold due to the lower price; second,
producers sell fewer units overall, meaning that surplus is lost through the decrease in
the quantity traded. Overall, there is dead-weight loss in the market. Note again, we
have assumed that the consumers with the highest MB of the good are the ones who
actually get to buy it (that is, the consumers with MBs at the very top-left of the demand
curve. If other consumers who do not value the product as highly (but still have a
–
MB > P ) receive the product instead, the resulting DWL will be even larger.
16.2.3 Price controls in the long run

Supply and demand change over time, which may affect the equilibrium price and
quantity. This means that a price ceiling or price floor that was non-binding may become
binding, or vice versa.
Typically, supply and demand will be more elastic in the long run, as participants
in the market adjust to market conditions. This can mean that shortages or excess of
a product may be worsened in the long run. For example, consider rent-control laws
that set a price ceiling on residential leases. In the short term, this may only generate
a small excess of demand. However, in the long run, participants in the market will
respond to the price control. Landlords know that earnings from their rental properties
are capped by the price ceiling, and so may prefer to invest elsewhere. Moreover, the
price control makes renting cheap relative to, say, making mortgage payments on a
house, so more people will prefer to rent rather than buy a house. The combination of
the increase in demand and decrease in supply of rental properties will worsen the issue
of excess of demand over the longer term.
16.3 Taxes and subsidies
We now turn to the effect of taxes and subsidies on market outcomes. We will restrict
our analysis to taxes and subsidies on consumption and production, but it is important
to be aware that other types of taxes (such as property and income taxes) and other
types of subsidies (such as export or employment subsidies) also exist.
16.3.1 Taxes
A tax is a compulsory payment made to the government. In this section, we will
consider per-unit taxes (or ‘specific taxes’), where the tax for each unit is a fixed
amount – that is, for every unit, a tax of t must be paid to the government. This can
be distinguished from an ad valorum tax, where the amount of the tax is a fixed
percentage of the price.
Tax on consumption
Let us first examine the case where the tax must be paid by consumers. For every unit
purchased, consumers must pay the market price P to the producer as well as a tax of
t to the government. Hence, the total amount paid by consumers is P + t.
The effect of a tax upon demand is illustrated in Figure 16.5. Before the tax is
introduced, the demand curve and the supply curve (D0 and S0) yield the equilibrium
(Q*, P*). Once a tax of t is introduced, consumers must factor the tax into their
purchasing decisions. Now, consumers will only buy an additional unit of the good
if the total amount paid (P + t) does not exceed their marginal benefit (MB); that is to
say, the maximum price P that consumers would be willing to pay for any unit
is P = MB – t, as P + t = MB. The maximum price that consumers are willing to
pay is reduced by the size of the tax; in other words, the demand curve shifts vertically
downwards by the size of the tax, to D1.
S0
P*
D1 D0
Q
Q*
FIGURE 16.5 A tax on consumers causes the demand curve to shift downwards by the size of the tax
This creates a new equilibrium at the intersection of S0 and D1, as seen in Figure
16.6. At this new equilibrium, the quantity traded in the market is Qt, and the price in
the market (received by producers) is Pt . However, the total amount paid by consumers
is Pt + t, as they must pay the market price, plus the tax of t to the government.
Tax on production
Let us now turn to taxes paid by producers. For every unit sold, producers must pay
a tax of t to the government. Hence, the total amount that a producer receives for selling
a unit of the good is P – t, which represents the price received minus the payment of
the tax.
The effect of a tax upon demand is illustrated in Figure 16.7. Before the tax is
introduced, the demand curve and the supply curve (D0 and S0) yield the equilibrium
S0
Pt + t
Pt
D1 D0
Q
Qt
FIGURE 16.6 A tax on consumers creates a new market price and quantity at (Qt , Pt ). However, the total
amount paid by consumers is Pt + t
P S1
S0
P*
D0
Q
Q*
FIGURE 16.7 A tax on producers causes the supply curve to shift upwards by the size of the tax
(Q*, P*). Once a tax of t is introduced, producers must factor that tax into their
production choices. Now, producers should only sell an additional unit of the good if
the amount they receive covers their marginal cost; that is to say, the minimum price
that producers should be willing to accept is P = MC + t. Hence, the minimum price
that producers are willing to accept is increased by the size of the tax; in other words,
the supply curve shifts vertically upwards by the size of the tax, to S1.
This create a new equilibrium at the intersection of S1 and D0, as seen in Figure
16.8. At this new equilibrium, the quantity traded in the market is Qt , and the price in
the market (paid by consumers) is Pt . However, the net amount received by producers
is the market price minus the tax of – that is, Pt – t.
P S1
S0
Pt
Pt – t
D0
Q
Qt
FIGURE 16.8 A tax on producers creates a new market price and quantity at (Qt , Pt ). However, the net
amount received by producers is Pt – t
Effects of tax on welfare

The effect of a tax is to drive a wedge between the price paid by consumers (PC ) and
the amount received by producers (PP). This will have implications for welfare, as
illustrated in Figure 16.9. Both consumer surplus and producer surplus will decrease
as a result of the tax, for two reasons: first, fewer units are traded in the market; second,
on each unit, consumers pay a higher price and producers receive a lower price relative
to a market without a tax.
However, as a result of the tax, the government now receives some surplus in the
form of tax revenue. The total revenue received by the government is denoted by the
S0 CS
PS
GR
DWL
PC
PP
D0
Q
*
Qt Q
FIGURE 16.9 The welfare effects of a tax

size of the tax (t), multiplied by the number of units taxed (Qt ). This is represented by
the area GR in Figure 16.9. The surplus to the government must also be factored into
total surplus; hence, total surplus is given by the sum of consumers surplus, producer
surplus and government revenue (TS = CS + PS + GR).
As a result, the deadweight loss from taxation (DWL) is illustrated in Figure 16.9.
This DWL is caused by reduction in the quantity traded in the market (from Q* to Qt ).
The larger this reduction, the greater the DWL. Hence, the size of the DWL depends
on how elastic the demand and supply curves are. The more elastic the demand and/or
supply curves, the greater the effect of the tax on the quantity traded in the market,
which will result in a greater DWL.
Incidence of tax
The concept of tax incidence analyses how the burden of the tax is distributed between
consumers and producers. The legal incidence of the tax refers to who is legally
responsible for paying the tax. It answers the question: is this tax levied on consumers
or on producers? By contrast, the economic incidence of the tax refers to who, as
a matter of fact, actually bears the burden of the tax.
As a general rule, the legal incidence of the tax has no bearing on the economic
incidence of the tax. Rather, the economic incidence of the tax is determined solely
by the relative elasticities of the demand and supply curves.1 Thus, if the demand curve
is elastic relative to the supply curve, producers will bear a greater share of the tax
burden; conversely, if the supply curve is elastic relative to the demand curve,
consumers will bear a greater share of the tax burden. Usually, the economic incidence
of the tax is shared between both parties, but there may be exceptional cases if either
curve is perfectly elastic or perfectly inelastic. We discuss this issue further now.
Example. Consider the market for tomatoes depicted in Figure 16.10, where the
market price is P*. Suppose a per-unit tax of t is levied on consumers, shifting
the demand curve downwards. Now, the total price paid by consumers (Pt + t) and
the price received by producers (Pt ). In this case, the legal incidence of the tax is
on consumers, because it is they who are legally responsible for paying the tax.
However, the economic incidence of the tax is split between consumers and
producers, as consumers pay a higher price than they did without the tax and
producers receive less than they did per unit sold than they did without the tax.
The loss of consumer surplus is denoted by the area A, and the loss of producer
surplus is denoted by the area B.
Example. Consider the following example. Demand in a market is given by

P = 24 – qd , where P is market price and qd is quantity demanded. Supply is given
by P = 2qs , where qs is the quantity supplied. The market equilibrium price
and quantity traded are P* = 16 and q* = 8. Now assume that a per-unit tax of
$3 is imposed on consumers (the legal incidence is on consumers in the market).
If producers receive a price of Pt following the introduction of the tax, consumers
will pay Pt + 3 in total for each unit they buy. Consequently, consumers will adjust
S0
CS
Pt + t
A
P*
B
Pt
PS
D1 D0
Q
Qt Q*
FIGURE 16.10 A tax on consumers in the market for tomatoes. The legal incidence of the tax is borne
by consumers, but the economic incidence of the tax is shared. The price that consumers pay increases
from P * to Pt + t, whereas producers receive Pt following the introduction of the tax, rather than P *
their willingness to pay such that Pt + 3 = 24 – qd , or that Pt = 21 – qd . The supply

curve is unchanged: Pt = 2q. To find the new equilibrium after the imposition of
the tax, we equate new demand curve (with the tax) and the supply curve, so that
21 – q = 2q; the new quantity traded is qt = 7. From this, the suppliers will receive
a price of Pt = 14, and consumers will pay a total price (to the producers and to
the government) of Pc = 14 + 3 = 17. In this example, consumers pay an extra
$1 per unit and suppliers receive $2 less per unit.
Now consider that the legal incidence of the tax is now on suppliers. After the
tax, the demand curve will be unchanged; Pc = 24 – q, where Pc is the price
consumers pay in the market after the tax has been imposed. Suppliers however
will take into account that they have to pay $3 to the government for each unit
they sell; the supply curve will now be Pc – 3 = 2qs , or that Pc = 2qs + 3. Again,
to find the market equilibrium, equate demand with the new supply curve; the new
equilibrium with the tax has a quantity traded of qt = 7, with a consumer price
Pc = 17 and a supplier price of Ps = Pc – 3 = 14. This outcome is the same as when
the tax was legally imposed on the consumption side of the market – that is the
economic incidence of the tax is the same regardless as to whether consumers or
producers are legally required to pay the tax. Finally, note in this example that the
DWL = 1⁄2(8 – 7)(3) = $1.5.
16.3.2 Further analysis

Let us now explore a little further some of the issues highlighted in the sections
above.
First, we know that the incidence of the tax is invariant to which side of the market
legally pays for the tax. Rather, it is the relative elasticities of supply and demand that
P S1
S0
P* + t
P*
D0
Q
Q*
FIGURE 16.11 If demand is perfectly inelastic, when a tax of t per unit is instituted, consumers pay for
all of the tax; following the introduction of the tax consumers pay P * + t whereas suppliers continue to
receive P *
determine how much of a tax is paid for by consumers and producers; if demand
is relatively more inelastic (demand is relatively steep), consumers will pay more of
the tax (Pc – P* will be larger than P – Ps, where Pc is the full price consumers pay
and Ps *is the net price received by suppliers). Conversely, if supply is relatively
inelastic (relatively steep), suppliers will bear relatively more of the incidence of the
tax (P* – Ps will be larger than Pc – P*).
If one side of the market is perfectly inelastic, that side will pay for all of the tax.
Consider Figure 16.11. Here, demand is perfectly inelastic, so consumers end up paying
for all of the tax, and suppliers continue to receive P* per unit. The same idea applies
if supply was perfectly inelastic and demand was downward sloping – suppliers would
bear all of the incidence of the tax. A similar idea applies if one side of the market is
perfectly elastic – the other side of the market bears all of the incidence of the tax.
Consider in Figure 16.12 the market with a perfectly elastic supply curve and a
downward sloping demand curve. The tax on suppliers shifts the supply curve up by
the size of the tax to St . Consumers now pay Pc = P* + t and suppliers continue to
receive a price of Ps , after passing the tax t onto the government.
When both demand and supply are downward and upward sloping, respectively, then
both sides pay for some of the tax, the precise incidence depending on the relative
elasticity of supply and demand.
Second, the DWL generated by a tax depends on the elasticity of supply and
demand. A DWL is caused by a reduction in the quantity traded (from Q* to Qt ). The
larger this reduction, the bigger the DWL. For a given tax t, the reduction in quantity
is going to be larger the more responsive market participants are, because the wedge
between the D and S curves must equal to the size of the tax. The flatter the curves,
the greater the reduction in quantity required to achieve the necessary gap between the
two curves. For example, consider two markets with the same demand curve that both
P* + t S1
t
P* S0
D0
Q
Qt Q*
FIGURE 16.12 When a tax of t per unit is implemented and supply is perfectly elastic, consumers pay
for all of the tax; following the introduction of the tax, consumers pay P * + t whereas suppliers continue
to receive P *
Relatively elastic supply Relatively inelastic supply

P P
DWL S2 DWL
S1
t
D D
Q Q
FIGURE 16.13 The deadweight loss generated by a tax depends on the responsiveness of supply and
demand. In each panel the demand curve and the tax implemented are the same. There is a larger DWL
generated in the market with relative elastic supply (the left-hand panel) as compared with the DWL in
the market in which supply is relatively inelastic
experience the implementation of a per-unit tax t. As in Figure 16.13, the two markets
have different supply curves, the first being relatively elastic and the second relatively
inelastic. The more elastic the supply curve, the greater the reduction in quantity caused
by the tax and the larger the DWL generated. (The same logic applies to the elasticity
of demand.) This means that, other things equal, a greater DWL is caused by a tax in
a market with relatively responsive (elastic) supply and demand curves.
Third, consider the impact on tax revenue and DWL as a tax is increased. With a
zero tax there is no DWL or tax revenue. If a small tax is levied, it causes a wedge
between D and S, reducing the quantity traded. This creates a DWL. As the tax is
increased, the wedge becomes larger and the DWL increases. This is true up to when
the tax is equal to the difference between the MB of the first unit consumed and the
MC of the first unit made – that is, the difference between the height of the D and S
curves at Q = 0. Once a tax is this large, there is no trade and all the potential gains
from trade are lost (the DWL is equal to the area between the D and S curves between
0 and Q*).
Now consider the relationship between the size of the per-unit tax and tax revenue.
As the tax increases, tax revenue first rises, then falls. The logic is very similar to why
total expenditure in a market can rise or fall with a price increase. To see this, start
with a tax of t = 0. With no tax, tax revenue is zero. If a small tax is implemented, tax
revenue is the area of the rectangle equal to the size of the tax t times by the quantity
traded Qt . Implementing a small tax from a base of t = 0 generates some tax revenue,
so tax revenue increases with a tax increase. This is true for further small tax increases;
the tax revenue rectangle increases as the proportional increase in the height of the
rectangle outweighs the reduction in the length (given by Qt ) . But as the tax gets larger,
the proportional decrease in quantity begins to outweigh the proportional increase in
the tax, and total tax revenue starts to fall. This continues up until the tax totally crowds
out the market (with a tax equal to the difference between MB and MC at Q = 0). At
this level of taxation, quantity traded is zero and there will be no tax revenue.
This observation that tax increases and subsequently falls is often referred to as the
Laffer curve, who used this idea to argue for a decrease in income taxation rates to
President Reagan in the US in the early 1980s. This idea was that income rates were
so high that if the government reduced them, people would respond to this incentive
by working more, decreasing the DWL from taxation and increasing government
revenue. Income tax rates were decreased in the US, however tax revenues fell. This
suggests that while some individuals may have increased the amount they worked, this
was not true for the economy overall. It is also consistent with the idea that the pre-
existing levels of taxation were not as prohibitively high as argued by Laffer and others.
16.3.3 Subsidies
A subsidy is a payment made by the government to an individual or firm, and can be
thought of as a negative tax. For this reason, much of the analysis concerning taxes
can be modified to describe the effects of a subsidy.
Subsidies for consumption

In the case of a subsidy for consumption, the government makes a payment of $s to
consumers for every unit of the good purchased. That is, for every unit consumed,
consumers must pay the market price P to the producer, but will receive a subsidy of
s from the government. Hence, the total amount paid by consumers is P – s.
S0
Ps
P*
Ps – s
D0 D1
Q
Q* Qs
FIGURE 16.14 A subsidy for consumption creates a new market price and quantity at (Qs , Ps ). The total
amount paid by consumers is Ps – s
The effect of this subsidy will be to shift the demand curve upwards by the size of
the subsidy, s. This is depicted in Figure 16.14 with the movement of the demand curve
from D0 to D1. As a result, the new quantity traded in the market is Qs and the price
in the market (received by producers) is Ps . However, the net amount paid by consumers
is Ps – s, as after they pay the market price to producers, they also receive the subsidy
of s from the government.
Subsidies for production

Alternatively, the government may subsidize production, by making a payment of $s
to producers for every unit of the good sold. As a result, the producer will receive the
price P from consumers, as well as a subsidy of s from the government. Thus, the total
amount received by producers is P + s.
As a result of this subsidy, the supply curve will shift downwards by the size of the
subsidy, s. This is depicted in Figure 16.15 by a movement of the supply curve from
S0 to S1. As a consequence, the new quantity traded in the market is Qs and the price
in the market (paid by consumers) is Ps . However, the total amount received by
producers is Ps + s, which includes the subsidy.
Economic incidence of a subsidy

Like with a tax, the economic incidence of a subsidy does not depend on the legal
incidence of the subsidy (that is, whether the government technically pay the subsidy
to consumers or producers). As before, the economic incidence depends on the relative
elasticities of demand and supply. With a subsidy, it is the relatively inelastic side of
the market that enjoys more of the benefits in terms the price they pay or receive. For
example, if demand is perfectly inelastic and the law of supply holds, the resulting
consumer price falls by the size of the subsidy, and the final price (including the subsidy
S0
S1
Ps + s
P*
Ps
D0
Q
Q* Qs
FIGURE 16.15 A subsidy for production creates a new market price and quantity at (Qs , Ps ). The total
amount received by producers is Ps + s
received) by producers remains unchanged. Of course, this distribution of the gains

from the subsidy is invariant to the legal arrangements for the subsidy.
Effects of subsidies on welfare

The effect of a subsidy is increase the quantity of a good traded in the market, as
depicted in Figure 16.16. Moreover, the price paid by consumers (PC) is now less
than the price received by producers (PP) , as the government subsidy makes up the
difference. As a result, both consumer surplus and producer surplus increase for two
reasons: first, more units are traded in the market; second, on each unit, consumers
pay a lower price and producers receive a higher price relative to a market without a
subsidy. The areas of consumer and producer surplus are depicted in Figure 16.15.
S0 CS
PS
PP
P*
Pc
D0
Q
Q* Qs
FIGURE 16.16 Consumer surplus and producer surplus as a result of a subsidy

S0 GR (negative)
DWL
PP
P*
Pc
D0
Q
Q* Qs
FIGURE 16.17 Government revenue and deadweight loss as a result of a subsidy
However, as a result of the subsidy, the government must now make payments to
consumers and/or producers, which represents negative government revenue – or in
other words, negative surplus. The size of this negative surplus is denoted by the size
of the subsidy (s), multiplied by the number of units subsidized (Qs ). This is represented
by the area GR in Figure 16.17. This negative surplus must also be factored into total
surplus; hence, total surplus is given by TS = CS + PS – GR. As a result, there is DWL
arising from the subsidy, as depicted in Figure 16.17. This DWL is caused by an increase
in the quantity traded in the market, beyond the efficient level (from Q* to Qs ).
An alternative (parallel) intuition for this DWL is as follows. Each extra unit
produced between Q* and Q s has a MB indicated by the demand curve (D0), whereas
the MC of production is given by the supply curve (S0). Consequently, as MB < MC
for each of these units, total surplus must fall because the subsidy encourages too much
production of the good (beyond the surplus-maximizing quantity Q*).

Government intervention can affect the price and quantity of goods traded in the market.
In this chapter, we found that if a market is initially efficient, the introduction of a price
control, tax or subsidy makes the market less efficient (that is, it results in deadweight
loss). This is because the intervention moves the quantity traded in the market away
from the efficient quantity. In coming chapters, however, we will explore the possibility
that, if the market is not efficient to begin with, government intervention affects the
quantity of goods traded into the market so as to make the market more efficient.
Notes
1 A corollary of this is that taxes on consumers and taxes on producers will have identical
welfare effects.
C H A P T E R
17
Externalities
17.1 Introduction
As we have seen in previous chapters, competitive markets are usually Pareto

efficient. This is because all mutually beneficial trades1 occur, maximizing the
gains from trade and hence total surplus. However, there are some situations
where the market outcome will not be efficient; these are called market
failures.
This chapter examines one type of market failure: externalities. An
externality is a cost or benefit that accrues to a person who is not directly
involved in an economic activity or transaction. The presence of an externality
means that the market outcome may not be efficient, because the market does
not take into account these external costs and benefits of producing or
consuming a product. This means that a competitive market can end up
producing too much of a product (with a negative externality) or too little
(when there are external benefits).
We begin with a discussion of the nature of and the market failure associated
with externalities, and then turn to several possible solutions to this market
failure.
17.2 External costs and benefits
An externality is a cost or benefit of an economic activity that accrues to a

person not directly involved in that activity.2 These costs or benefits are also
known as ‘external costs’ or ‘external benefits’.
• A positive externality occurs when the economic activity results in

external benefits for a third party. For example, if a student decides to
pursue further education, there may be benefits to society as a whole from having
better-educated citizens, in addition to the individual benefits that are enjoyed by
the student themselves.
• A negative externality occurs when the economic activity results in external
costs for a third party. For example, if a factory manufacturing wrenches pollutes
a river in the course of production, this is a cost borne by people wishing to use
the river downstream rather than by the manufacturer or the end consumers of the
wrenches.
The following subsections examine the effect of positive externalities and negative
externalities in turn. To illustrate a positive and a negative externality we first consider
a positive consumption externality and a negative production externality.
17.2.1 Positive consumption externalities

Consumers derive benefits from consuming goods. However, in the presence of a
positive externality, the consumption or production of the good also has external
benefits for a third party. Hence, the benefit to society as a whole (the ‘social benefit’)
must include both the consumer’s benefit and the external benefit.
Formally, the marginal benefit to society of an additional unit of the good is known
as the marginal social benefit (MSB). It is made up of two components: the marginal
private benefit (MPB) that is enjoyed by the consumer and the marginal external
benefit (MEB) that accrues to a third party:
MSB = MPB + MEB . (17.1)
Figure 17.1 represents the relationship between the MPB and the MSB. Note that,
in the presence of a positive consumption externality, the marginal social benefit is
higher than the marginal private benefit – the difference between the MSB and the MPB
is the size of the externality, for any given unit. In Figure 17.1, the increasing gap
between the MPB and the MSB indicates that the positive externality is increasing with
output. This need not be the case – there could be a constant positive externality per
unit of the good consumed, in which case MPB and MSB would be parallel.
Alternatively, a diminishing gap between MSB and MPB would represent a declining
positive externality as more of the good is consumed. Finally, if there is no positive
consumption externality, MEB = 0 and MSB = MPB.
17.2.2 Negative production externalities

Similarly, producers incur costs from producing goods. When a negative production
externality is present, the consumption or production of the good also has external costs
for a third party. Hence, the cost to society as a whole (the ‘social cost’) must include
both the producer’s cost and the external cost.
Externalities 165
MPB MSB
Q
FIGURE 17.1 The relationship between MPB and MSB in the presence of a positive externality
Formally, the marginal cost to society of an additional unit of the good is known as
the marginal social cost (MSC). It is made up of two components: the marginal
private cost (MPC) that is incurred by the producer and the marginal external cost
(MEC) that is incurred by a third party:
MSC = MPC + MEC . (17.2)
Figure 17.2 represents the relationship between the MPC and the MSC. Note that
the presence of the negative externality means that the marginal social cost is higher
than the marginal private cost. The size of the negative externality is the difference
between the MSB and the MPB for a given level of output. Note in Figure 17.2, the
increasing gap between the MSC and the MPC indicates the negative externality is
increasing with the level of output produced in the market. Of course, the size of the
externality need not be increasing with output – for example, there might be a constant
negative externality incurred for every unit of output in which case the MSC and the
MPC would be parallel. Finally, if there is no negative externality, then MEC = 0 and
MSC = MPC.
17.3 The problem with externalities
When there are no externalities, all the benefits of consumption are enjoyed by
consumers, as depicted by the demand curve. Conversely, all the costs of production
are incurred by producers, as depicted by the supply curve. As a result, the market
equilibrium accounts for all costs and benefits from production and consumption.
Externalities are a source of market failure because they represent external costs or
benefits that are not accounted for by the market. Because consumers only account
MSC
MPC
FIGURE 17.2 The relationship between MPC and MSC in the presence of a negative externality
for their own private benefits and producers only account for their own private
costs, the market equilibrium is determined by the demand (MPB) and supply (MPC)
curves:
MPB = MPC . (17.3)
However, from the perspective of society as a whole, any external costs and benefits
associated with the consumption or production of the good should also be taken into
account when determining what is ‘optimal’ or efficient. Hence the socially optimal
equilibrium is determined by the intersection of the marginal social benefit and
marginal social cost curves:
MSB = MSC . (17.4)
17.3.1 Positive consumption externalities

In this way, the external benefits or costs resulting from the externality mean that the
market equilibrium is not the same as the socially optimal equilibrium. Figure 17.3
compares the equilibrium quantity QM and the socially optimal quantity Q* in the
presence of a positive consumption externality. In this case, the MSB curve is higher
than the MPB curve due to the external benefit. However, the MSC curve and the MPC
curve are the same because there is no negative externality.
At the market equilibrium, there is under-production relative to the socially optimal
level of output, Q*. In particular, note that the units between QM and Q* are not traded
in the market because consumers and producers have no private incentive to do so;
for these units, the MPC exceeds the MPB. However, from the viewpoint of society
as a whole, it would be desirable for these units to be traded, because their MSC is
less than their MSB.
Externalities 167
MPC, MSC
P*
PM
MPB MSB
Q
QM Q*
FIGURE 17.3 The market equilibrium and the socially optimal outcome in the presence of a positive
externality. The area representing deadweight loss is shaded
Because these socially beneficial trades do not go ahead, there is deadweight loss
associated with the market equilibrium. The area of deadweight loss is shaded grey in
Figure 17.3.
17.3.2 Negative production externalities

Similarly, a negative externality can also mean that the market equilibrium is not the
same as the socially optimal equilibrium. Figure 17.4 compares the market equilibrium
MSC
MPC
P*
PM
MPB, MSB
Q
Q* QM
FIGURE 17.4 The market equilibrium and the socially optimal outcome in the presence of a negative
externality. The area representing deadweight loss is shaded
quantity QM and the socially optimal quantity Q* in the presence of a negative

production externality. In this case, the MSC curve is higher than the MPC curve due
to the external cost. However, the MSB curve and the MPB curve are the same because
there is no positive externality.
Now, at the market equilibrium, there is over-production relative to the socially
optimal level of output, Q*. The units between Q* and QM are traded in the market
because, for these units, the MPB of consumers exceeds the MPC of producers.
However, from the viewpoint of society as a whole, it would be desirable for these
trades not to go ahead, because their MSC is greater than their MSB.
Because the amount traded in the market is greater than the socially optimal amount,
there is deadweight loss associated with the market equilibrium. The area of deadweight
loss is shaded grey in Figure 17.4.
17.3.3 Positive production and negative consumption externalities

Thus far we have concentrated on two types of externalities: positive consumption
and negative production externalities. In many ways, it does not really matter on
which side of the market the externality can be attributed to – what is more important
is that the external cost or benefit drives a wedge between private and social marginal
benefits or costs. As a result, the market outcome will not necessarily coincide
with the socially efficient outcome. For completeness, however, we briefly outline the
two other types of externality – a negative consumption externality and a positive
production externality.
Figure 17.5 illustrates a negative consumption externality, which could represent
the consumption of a product like cigarettes that causes a cost to incurred to bystanders
via passive smoke. In this case, the MSB is less than the MPB by the size of the
externality; this is represented by the vertical downwards shift of the MSB from
MPC, MSC
PM
P*
MSB MPB
Q
Q* QM
FIGURE 17.5 The market equilibrium and the socially optimal outcome in the presence of a negative
consumption externality. The area representing deadweight loss is shaded
Externalities 169
MPC
MSC
PM
P*
MPB, MSB
Q
QM Q*
FIGURE 17.6 The market equilibrium and the socially optimal outcome in the presence of a positive
production externality. The area representing deadweight loss is shaded
the MPB by the size of the negative consumption externality at every level of output
(MSB = MPB – MEC, where MEC is the size of the external cost). In this case market
output is too high (QM > Q*), and a DWL results, as illustrated by the shaded area on
the figure; for every unit between Q* and QM the cost to society for each unit (MPC)
exceeds the extra benefits generated (MSB).
A positive production externality is illustrated in Figure 17.6; an example of this
situation could be research and development. While the investing firm could get a
private return from their efforts, other firms (not even necessarily in the same industry)
might also enjoy some benefits from this R&D. These ‘spillovers’, as they are often
referred to, are externalities because the investing firm only considers it private benefits
(and costs) when making its R&D choice. With the positive production externality
shown in Figure 17.6, the MSC is lower than the MPC by the size of the positive
externality. Consequently, the market output, where MPB = MPC, resulting is an
equilibrium quantity QM less than the surplus-maximing outcome Q*. The DWL is the
shaded area – for every unit between QM and Q* the MPB > MSC; the DWL indicates
the surplus forgone in the market equilibrium relative to the efficient outcome.
17.4 Solutions to externalities
A number of solutions exist to correct the deadweight loss arising from an externality.
In this section, we consider three solutions: (a) the Coase Theorem; (b) taxes and
subsidies; and (c) standards and regulations.
17.4.1 Private-market solutions: the Coase Theorem

One way that an externality may be corrected is via private bargaining between the
involved parties. In this way, the market participants and the third parties affected by
the externality can ‘renegotiate’ the market outcome, such that the socially optimal
outcome is implemented and the deadweight loss is eliminated. To understand how
this works, consider the following example.
Example. A beekeeper and an almond grower have farms next to each other. The
beekeeper’s bees provide a positive externality for the almond grower by
pollinating his almond flowers, which increases almond production. If the
beekeeper increases her number of hives, even more flowers will be pollinated.
This suggests that if the beekeeper decides how many hives to keep based on her
interests alone, this number may be fewer than is socially efficient because the
almond grower also benefits from each additional hive. However, this can be
addressed via private negotiation between the parties. The parties may come to an
agreement that the almond grower will pay the beekeeper to increase her number
of hives. Note that this agreement is viable so long as the MSB (that is, the
beekeeper’s MPB plus the almond grower’s MEB) exceeds the beekeeper’s
marginal cost of maintaining the additional hive, which is the same condition
necessary for the extra hive to increase total surplus.
The idea that private bargaining can implement the socially efficient outcome is
articulated in the Coase Theoreom. This theorem is formally stated below.
Coase Theoreom. Provided property rights have been clearly assigned and there are
no transaction costs, bargaining will lead to an efficient (i.e. socially optimal) outcome,
regardless of the initial allocation of property rights.
The Coase Theorem depends crucially on property rights. In this context, the
‘property right’ referred to is the right to decide whether and to what extent the
economic activity causing the externality goes ahead. Thus, the Coase Theorem is
essentially an application of the gains from trade principle, where the parties are simply
trading the property right rather than goods or services. In the example above, the
property right being traded is the right to determine whether or not the beekeeper
maintains an additional hive. This right was initially held by the beekeeper, but could
be sold to the almond grower.
Importantly, the attainment of the socially optimal outcome does not depend on how
property rights are allocated, but merely the fact that they have been allocated.
Nevertheless, the allocation of property rights will have implications for how the gains
from trade are distributed between the parties.
Example. Alex owns a factory that generates some black smoke as a byproduct
of its production. The smoke drifts over to Bob’s laundromat; however, Bob
requires clean air in order to properly wash clothes. At present, Alex makes $300
Externalities 171
from operating the factory, whereas Bob could make $500 profit from washing
clothes. In this situation, the socially efficient outcome is that the factory ceases
production, so that the laundromat can operate (because 500 > 300). Provided that
there are no transaction costs and that property rights are clearly defined, the parties
can negotiate to achieve this outcome, regardless of who initially holds the property
right.
• Suppose that Alex initially holds the right to decide whether the factory
operates. Alex values this right at $300, since that is the profit that she can
make from deciding that the factory will operate. Bob, on the other hand,
values the right at $500 because that is the profit he can make from deciding
that the factory will not operate. Consequently, Alex will sell Bob the right
to decide for a price between $300 and $500, and the factory will cease to
operate.
• Suppose that Bob initially holds the right to decide whether the factory
operates. Bob values the right at $500 because that is the profit he can make
from stopping factory operations and running his laundromat. On the other
hand, Alex values the right at $300, since that is the profit that she can make
from operating the factory. Because Bob values the right more highly than
Alex does, he will not sell the right to her. Hence, Bob will decide that the
factory should cease to operate.
Note that the factory ceases to operate, whether the property right is allocated to
Alex or to Bob. Nevertheless, the allocation of property rights affects how surplus
is distributed between the parties.
• When Alex initially holds the property right, Bob must buy the right from
her at a price of $p (300 ⭐ p ⭐ 500). Thus, Alex’s surplus after the trade is
$p and Bob’s surplus is $500 – $p.
• When Bob initially holds the property right, the property right is not traded.
As a result, Alex’s surplus is $0 and Bob’s surplus is $500.
While the Coase Theoreom offers an important insight into how externalities can
be addressed by private negotiations, it relies on certain conditions. Thus, the Coase
Theorem may fail in the real world if these conditions are not met, for example:
• Property rights not defined. The importance of property rights was discussed
above. If initial property rights are not properly defined, the parties will not have
a ‘starting position’ from which to begin their negotiations. That is, if it is not
clear who initially owns the property right, it will not be possible for parties to
trade that right.
• Transaction costs. The Coase Theorem assumes that there are no transaction
or bargaining costs. However, in the real world, there are often both implicit and
explicit costs in negotiating, executing and enforcing an agreement. If these costs
are too high, it can prevent the parties from negotiating or trading at all, because
the gains to be made do not outweigh the costs. This is more likely to be a problem
if the effect of the externality is dispersed among many parties, such that the relative
benefit to any one individual from negotiating for a better outcome is small
relative to the costs.
• Identity of parties unknown. If the parties are unable to identify each other, it
will not be possible for them to engage in negotiations. For example, if a factory
owner is unable to identify who is affected by its pollution or the affected parties
are unable to discover the source of the pollution, the parties will not be able to
negotiate with each other. Again, this is more likely to be a problem if there is a
large number of parties.
17.5 Government solutions to externalities
Governments may intervene where private markets are unable to correct market
failures on their own. Two typical government interventions are: (i) taxes or subsidies;
and (ii) a regulation (or standard).
17.5.1 Taxes and subsidies

As discussed earlier in the chapter, the presence of an externality causes a divergence
between the market equilibrium quantity (QM) and the socially optimal quantity (Q*).
From Chapter 16, we also know that a tax or a subsidy can influence the quantity traded
in the market. Thus, a tax or subsidy can be used to ‘correct’ the market failure caused
by an externality, by reducing or increasing the quantity traded in the market to the
socially optimal level.3
Positive externality
In the presence of a positive externality, the quantity traded in the market (QM) is less
than the socially optimal quantity (Q*). To raise the quantity traded in the market to
Q*, governments can grant a subsidy to either consumers or producers, creating an
incentive for the market participants to increase the quantity traded.
Figure 17.7 depicts the effect of subsidizing consumers or producers. When the
subsidy is given to consumers, the demand curve (the MPB curve) shifts up by the
size of the subsidy per unit, increasing the quantity traded in the market. Similarly,
when the subsidy is given to producers, the supply curve (the MPC curve) shifts down
by the size of the per-unit subsidy, increasing the quantity traded in the market.
The trick, of course, is to set the size of the subsidy just right so that the socially
optimal quantity is achieved. As you can see from Figure 17.7, this means that the size
of the subsidy (s) should be equal to the size of the externality (that is, the difference
between MPB and MSB) at the socially optimal quantity. Because the size of the subsidy
is equal to the size of the externality, the subsidy causes market participants to
‘internalize’ the externality. This implements the socially optimal outcome and hence
eliminates any deadweight loss.
Externalities 173
Subsidy to consumers Subsidy to producers

P P
MPB + s
MPC, MSC MPC, MSC
MPB – s
P* P*
PM PM
s s
MSB MSB
MPB MPB
Q
QM Q* QM Q*
FIGURE 17.7 A subsidy granted to the consumer and to the producer, used to address a positive
externality
Negative externality
When there is a negative externality, the quantity traded in the market (QM) is greater
than the socially optimal quantity (Q*). To lower the quantity traded in the market to
Q*, governments can impose a tax on either consumers or producers, creating an
incentive for the market participants to decrease the quantity traded.
Figure 17.8 depicts the effect of taxing consumers or producers in the presence of
a negative production externality. When the tax is imposed on consumers, the demand
curve (the MPB curve) shifts down by the size of the tax, decreasing the quantity traded
in the market. Similarly, when the tax is imposed on producers, the supply curve (the
MPC curve) shifts up by the size of the tax, decreasing the quantity traded in the market.
Tax on consumers Tax on producers

P P
MSC MSC
MPC + t
MPC MPC
P* P*
PM t PM t
MPB MPB
MPC – t MSB MSB
Q Q
M
Q* Q M Q* Q
FIGURE 17.8 A tax imposed on the consumer and on the producer, used to address a negative
externality
Again, the size of the tax needs to calibrated so as to achieve the socially efficient
quantity. As depicted in Figure 17.8, this means that the size of the tax (t) should be
equal to the size of the externality (that is, the difference between MPC and MSC) at
the socially optimal quantity. Because the size of the subsidy is equal to the size of
the externality at the efficient level, the subsidy causes market participants to
‘internalize’ the externality – they act as if they take into account the externality, even
though they are actually considering only their private benefits and costs, which now
include the Pigovian tax. This implements the socially optimal outcome and hence
eliminates any deadweight loss.4
17.5.2 Quantity regulation

In the previous section, we discussed how a government could use taxes and subsidies
to indirectly implement the socially optimal outcome. However, as an alternative, the
government could simply regulate the quantity traded in the market directly by
mandating that a certain quantity (specifically, the efficient quantity) be produced.
Example. Suppose that, in equilibrium, the quantity of vuvuzelas traded is 500.

However, because the use of vuvuzelas produce a negative externality (noise pollu-
tion), the socially optimal quantity of vuvuzelas is actually 100. The government
could implement the socially optimal outcome by mandating that no more than
100 vuvuzelas are produced.
One way of implementing quantity restrictions is by requiring a licence to produce

(or consume) a unit of output and by limiting the number of licences issued. If the
licence must be obtained by the party creating the externality, the licence can be also
be thought of as a licence to create the externality. For example, suppose that the
manufacturing of cars is associated with some level of pollution. Because the two go
hand in hand, a licence to manufacture a certain number of cars can equally be thought
of as a licence to emit a certain level of pollution. Similarly, in the vuvuzelas example
above, the government could regulate the quantity in the market by issuing licences
to either producers or consumers (even though the externality is generated by
consumption, the government can regulate output on the other side of the market with
the same economic effects).
17.5.3 Taxes and subsidies versus regulation control

From our discussion above, we can see that taxes and subsidies are aimed at
implementing the socially optimal outcome by influencing the price in the market,
whereas regulations and tradeable permits aim to do so by influencing the quantity
traded. This gives rise to the question: which of these two approaches is better?
As it happens, this will depend on the circumstances of the case, but there are some
factors worth considering. One advantage of regulation is that is creates certainty
about the level of output, that is, the level of the pollution. This is particularly valuable
Externalities 175
when the government is unsure about how market participants will react to a tax
or subsidy.
On the other hand, a tax has several advantages. First, a tax is always a price to
pollute, and firms will have an incentive to avoid the tax if they can. If a new pollution-
reducing technology becomes available, for example, a tax gives firms an incentive to
adopt the technology so as to avoid paying the tax. A regulation does not necessarily
give the firms an incentive to continue to reduce emissions beyond what is required
by the regulation.
Second, some firms can reduce emissions more cheaply than other firms. For
example, more modern factories might be able to reduce emissions relatively more
easily than older establishments. Similarly, the technology in some industries might
lend itself to reducing pollution emissions more than in other industries. A tax, by
setting a price for polluting, allows for unequal reductions of emissions across firms
and industries. This, in turn, reduces the cost of achieving the required reduction in
emissions, as a greater share is undertaken by the firms that can do so at lower cost.
A regulation, on the other hand, might require all firms make the same reductions in
emissions. This means that the cost of reducing emissions is not done in the least-cost
way; low-cost firms do too little and firms for which reducing emissions is costly do
too much.5
17.5.4 Tradeable permits

As noted, a regulation has the advantage that it gives certainty about the level of the
pollution that will be emitted. A tax on emissions has the advantage of providing
incentives for firms to reduce output in order to reduce pollution, as they face a cost
when they do so. A tradeable market is an attempt to capture both of these advantages
while addressing an externality. Tradeable permits are a special type of licence that
may be transferred between parties; thus, consumers (resp. producers) may trade with
each other for the right to consume (resp. produce) units of output. Essentially, this
system creates a market for pollution. If a firm holds a permit, the opportunity cost of
using it is that it cannot sell it. If it sells it, on the other hand, the opportunity cost
is that it cannot use it. What the tradeable permits market creates is an incentive for a
firm with a relatively low value for the permit to trade it with another firm that values
it more. Remember, a permit is essentially a licence to pollute – so that means that
firms with a low value of polluting (maybe they make a low-value product) sell their
rights to pollute (their permits) to others which value the right to pollute more highly
(such as a firm that makes a high-value product). Similarly, a firm that can reduce its
emissions relatively cheaply will be willing to sell its permit to a firm that has a higher
cost of reducing emissions. This market for permits means that, regardless as to the
initial distribution of permit, the firms will trade the permits so that the efficient outcome
is achieved.
Example. Suppose the government issues 100 permits to produce a vuvuzela: 50

permits are given to Firm A and 50 permits are given to Firm B. Due to their
different production costs, Firm A can make $1 profit per vuvuzela, Firm B can
make $5 profit per vuvuzela. Therefore, Firm A values each permit at $1 and Firm
B values each permit at $5. As a result, Firm A will sell all its permits to Firm B
at a price between $1 and $5. At the conclusion of trade, Firm B will hold 100
permits, and surplus will be maximized.
Example. Consider the case when the government wishes to allocate 600 permits
to pollute between two firms, A and B. Firm A can make $5 profit from each permit.
Firm B can make $10 profit from the first 400 permits, then $2 profit from every
subsequent permit used thereafter. The government initially allocated 300 permits
to each firm, but they are both allowed to trade the permits between each other.
Firm B can make $10 for each permit for the first 400 permits, but it only has 300.
B will negotiate and offer to buy an additional 100 permits from A, for a price
somewhere between $5 and $10. Firm A will be willing to sell too, as it can make
at least $5 per permit from the trade. But this is where trading stops. After B has
400 permits, any subsequent permits are only worth $2 to it, and this is not going
to be enough to induce Firm A to sell. So, even though we started with an
allocation of 300 for each firm, Firms A and B end up with 200 and 400 permits
respectively. This outcome, moreover, is the outcome that maximizes total surplus.
This final allocation produces the highest level of profit given that there are 600
permits available in total. A similar process would occur if Firm A was allocated
all of the permits initially and B got none; B would buy 400 permits from A for
a price between $5 and $10 and surplus will be maximized. Note also, that if the
permits were initially allocated by a competitive auction the efficient outcome
would also be achieved.
Again, in terms of dealing with an externality a government could opt to set a tax
(a price to pollute) or it could set a quantity of permits and allow the price of the permits
to be set in the market for permits. Notably, either option could achieve the same
(efficient) outcome. For example, if the government implements a tradeable permits
scheme with 100 permits and the price of a permit (to produce one unit of output)
trades for $20, the government could alternatively get the same outcome instituting a
tax of $20 per unit of output. Under either system, there is an extra $20 cost of
producing output – either that cost is the cost of the permit or it is the cost of the tax.
Thus, for every level of tradeable permits there is an equivalent tax that could
implement the same outcome.
Externalities are a source of market failure because they represent costs and benefits
that are not accounted for by the market. As a result, the market equilibrium that results
from the private decisions of consumers and firms will not necessarily be the socially
optimal outcome. There are a number of solutions to this problem. The Coase Theorem
Externalities 177
postulates that, if property rights are defined and there are no transaction costs, parties
will privately negotiate to implement the socially optimal outcome. Failing this, there
are also a number of government policies that can be used to correct externalities; these
include taxes, subsidies, standards and tradeable permits.
Notes
1 By ‘mutually beneficial trades’, we mean trades that can benefit both the buyer and the
seller.
2 Put another way, externalities can be thought of as ‘spill-over effects’ from economic
activity.
3 These taxes or subsidies are also caused ‘Pigovian taxes’ or ‘Pigovian subsidies’, after
Arthur Pigou who developed the concept of externalities.
4 Note, this is an example of a tax that does not cause a DWL.
5 As an example, say the government wants to reduce total emissions by 500 units due to a
negative externality, and there are two firms in the market. It costs the first firm $5 to reduce
emissions by one unit; on the other hand, it costs the second firm $12 a unit of emissions
reduction. Hence, it costs society $7 more for each unit reduction in emissions undertaken
by the second firm rather than the first firm. If the government’s regulation stipulated both
firms must reduce emissions by 250 units, the target of a total reduction of 500 units will
not be achieved at lowest cost.

C H A P T E R
18
Public goods and
common resources
18.1 Introduction
So far in this book, we have examined goods or services that can be considered
private goods. In this chapter, we examine different types of goods: public
goods and common resources. Because of their nature, public goods and
common resources can also be a source of market failure.
18.2 Public goods
A public good is a good or service that is non-excludable and non-rival.
1 Non-excludable. A good is non-excludable if the owner or provider of

the good cannot stop people from consuming it (and receiving the benefit
from doing so). For example, a person living in a country with a national
defence service cannot be excluded from enjoying the benefit of that
service. Similarly, a person living in a region that maintains a clean
environment or good biodiversity cannot be excluded from enjoying the
benefit of those goods.
2 Non-rivalrous. A good is non-rivalrous if one person’s consumption of
a good does not interfere with another person’s ability to consume the
same unit of the good. For example, one person using a street light to find
their way home does not prevent another person from using that same
streetlight.
By contrast, a private good is one that is excludable (the owner or provider

of the good can prevent others from enjoying it) and rivalrous (one person’s
consumption of the good prevents another person’s consumption of the good).
For example, a donut is a private good because it is possible to refuse to sell or give
someone a donut (excludable) and because the same donut can only be consumed once
(rivalrous).
18.2.1 The marginal benefit curve for a public good

With a private good, the market demand curve is obtained by the horizontal summation
of all the individual marginal benefit (MB) curves along the Q- axis. This is because
private goods are rivalrous, so if several individuals wish to consume the good, they
will each need to separately buy their own units. Thus, the total quantity consumed in
the market is the sum of each individual’s consumption.
However, the society’s total M B curve for a public good is obtained by the vertical
summation of all the individual MB curves along the P -axis.1
This is because public goods are non-rivalrous, so if several individuals wish to
consume a good, they can share units of that good. For example, suppose Aliya values
the construction of a lighthouse at $30 and Victoria values the construction of a light-
house at $50. Because the lighthouse can guide both of their ships, Aliya and Victoria
can ‘share’ the same lighthouse; that is, they do not need one each. Together, Aliya
and Victoria would be willing to pay $80 for the lighthouse. Therefore, the market’s
total willingness to pay is given by the sum of each individual’s valuation of the good.
Example. The local council is thinking about setting aside more space for parks.
Wilma’s marginal benefit for additional parks is MBW = 10 – q and Gerard’s
marginal benefit for additional parks is MBG = 20 – 2q, where q is the number of
additional parks. The MB curves are depicted in Figure 18.1. For simplicity, let
us assume that Wilma and Gerard are the only two residents who will make use
of the additional parks. To obtain the total MB curve (that is, the market demand
curve), we need to vertically sum Wilma’s MB curve with Gerard’s MB curve.
The total MB curve is also depicted in Figure 18.1 and is given by the equation
MBT = 30 – 3q.
18.2.2 The problem with public goods

As we have previously established, the efficient quantity of a good is the quantity at
which MC = MB. In the case of a public good, this means the quantity where the
marginal cost of providing the good is equal to society’s total M B for the good.
Example. The local council is thinking about setting aside more space for parks.
Wilma’s marginal benefit for additional parks is MBW = 10 – q and Gerard’s
marginal benefit for additional parks is MBG = 20 – 2q, where q is the number of
additional parks. The total willingness to pay for an additional park is given by
the equation MBT = 30 – 3q. Suppose the council’s cost of building a new park is
$21 per park. To solve for the efficient or socially optimal number of parks, we
need to solve MC = MBT . This gives 21 = 30 – 3q or q = 3.
Public goods and common resources 181
30
MBT
20
MBG
10
MBW
Q
10
FIGURE 18.1 The market demand curve (MBT ) for a public good is obtained by the vertical summation of
each individual’s marginal benefit curve
The market failure associated with public goods arises from two sources:
• Because the good is non-rivalrous, each individual consumer undervalues the good
relative to its total MB to society. As a result, it might be the case that no individual
consumer would purchase any units of the good, even though the collective
willingness to pay exceeds MC. In the example above, when the marginal cost of
parks is $21, neither Wilma nor Gerard would consume parks if they had to bear
the cost individually. However, because they can share the parks (and also the
burden of paying for the parks), together it is socially optimal that they consume
three parks.
• Because the good is non-excludable, consumers cannot be excluded from enjoying
the public good even if they have not paid for it (this is called ‘the free-rider effect’).
This makes it difficult for private firms to enforce payment and hence make profits
from the sale of a public good; as a result, there is little to no provision of public
goods by the private sector.
The upshot of this is that there tends to be under-provision of public goods in the free
market relative to the efficient quantity. As a result, public goods are often provided
by the government, who can enforce payment for the public good through the taxation
system. Indeed, governments in the real world tend to be responsible for providing
national defence services, street lighting, parks, environmental quality and so on.
18.3 Common resources and the Tragedy of the Commons
Common resources are often thought of as ‘partial’ public goods because they are
non-excludable, but rivalrous. Examples of common resource goods might be: fishing
stocks on the high seas; mineral deposits where there is no effective government
oversight; or use of a road or a bridge that the government cannot, or chooses not to,
restrict the use of.
The market failure arising from common resources is over-exploitation. This problem
is frequently referred to as the Tragedy of the Commons and is often described as
follows. Suppose the residents of a village share a field where each may graze cattle.
The field is non-excludable because no villager can be prevented from using the field
to graze cattle, but rivalrous because grazing by one villager depletes the amount of
grass available for the next villager’s cattle. Consequently, when a villager uses the
field for grazing, she reaps all of the benefits of doing so (her cattle are fed) but the
cost is shared among all the villagers (the stock of grass is depleted). As a result, each
individual villager uses the field for grazing more than she would if she alone bore the
entire cost of doing so, leading to overgrazing.
One way to address the over-exploitation of common resources is to create
enforceable property rights over the good. In the case of the communal grazing field,
this could take the form of land ownership or a permit to graze. By doing so, the good
ceases to be non-excludable, making it much easier to prevent overuse of the resource.
Similarly, driving on a congested bridge increases the cost of other users. If property
rights are allocated, the owner of the bridge could introduce a toll that effectively makes
drivers ‘internalize’ the congestion cost their road use imposes on others.
Because of their non-rivalrous nature, the ‘market demand’ curve for public goods is
given by the vertical sum of each individual’s MB curve. As a result, each individual
consumer undervalues the public good relative to the willingness to pay of society as
a whole. Moreover, their non-excludable nature means that there tends to be under-
provision of public goods by private firms.
Consequently, public goods tend to be provided by the government. Notwithstanding
this, it is important not to conflate the economic concept of a ‘public good’ with the
more common ‘publicly-owned good’. Indeed, many publicly-owned goods are not
public goods. For example, it is not unusual for services such as water supply, health
care and telecommunications to be provided by the government. However, individuals
can easily be excluded from accessing these goods, so they are not public goods.
Common resource goods are those that are non-excludable, but rivalrous. They often
suffer from over-exploitation, which may be mitigated by the allocation of property
rights.
Public goods and common resources 183
Notes
1 Note, sometimes society’s total MB for a public good is referred to as the ‘market demand’
curve for the public good; at times we use this terminology too. However, as discussed
below, the MB curve for a public good is not really a demand curve in the traditional sense
because it is non-excludable.

C H A P T E R
19
The Theory of
Second Best
19.1 Introduction
We have now looked at several sources of market failure, including monopoly

markets, externalities and public goods. These failures cause the market
outcome to deviate from the efficient or surplus-maximizing outcome, reducing
total surplus. We have also seen that government intervention (in the form of
taxes, subsidies, quantity-control, etc.) can address these market failures and
increase welfare in the market. However, in some cases, markets may be
affected by more than one source of market failure – for example, a market
may be affected by more than one externality. In this chapter, we discuss why,
in the presence of multiple market failures, it may be preferable not to address
a single market failure on its own.
19.2 Understanding the Theory of Second Best
In some instances, markets may be affected by more than one market failure.
The Theory of Second Best posits that if there is a market failure that not
or cannot be corrected, actions to correct other market failures may have the
effect of decreasing total surplus overall.
To understand how the Theory of Second Best works, consider the following
example.
Example. Suppose the market for tractors is a monopoly. The market

demand curve for tractors is given by the equation P = 10 – Q and
the marginal cost of producing tractors is $2. However, suppose the
production of tractors is also accompanied by a negative externality of
$4 per tractor.
The socially optimal outcome is given by setting the marginal social benefit
equal to marginal social cost (that is, 10 – Q = 2 + 4), which yields an efficient
quantity of Q* = 4.
Consider the outcome in the monopoly market. In this case the profit-
maximizing monopolist sets marginal revenue equal to marginal cost, disregarding
the negative externality (that is, 10 – 2Q = 2), which yields a monopoly quantity
of Qm = 4; this quantity is actually the socially efficient outcome as Qm = Q* = 4.
Thus, in this example, the monopolist maximizes total welfare and there is no DWL
in the monopoly market.
Now, suppose the government intervenes in the market to remove the monopoly
power of the firm, so that the market becomes perfectly competitive, but it does
nothing about the negative externality. The competitive market outcome after the
policy intervention can be determined by setting marginal benefit equal to marginal
cost, disregarding the negative externality (that is, 10 – Q = 2), which yields a
competitive level of output of Q c = 8. Notice that in this example, the competitive
market produces too much output relative to the socially efficient outcome –
Q c > Q*, so a DWL results. Moreover, the outcome is worse than the monopoly
market outcome without government intervention; here, the removal of one
distortion (the monopoly power) without simultaneously addressing the other (the
negative externality) reduces surplus, rather than making things better. Indeed,
surplus is higher if the government does nothing, rather than partially correcting
the market failures in this scenario.
The intuition underlying the Theory of Second Best is that multiple market failures
can counteract one another. In the above example, the monopoly power of the seller
meant it would tend to underproduce relative to the socially optimal outcome, whereas
the presence of a negative externality meant that the market outcome would tend to
be too high. Overall, the combination of these market failures produced a higher surplus
than when one or the other distortions are removed on their own. This idea potentially
applies whenever there is more than one market failure present in a market.
Hence, before intervening in a market, a policy maker needs to be well informed
about the state of the market and the possible existence of other distortions before trying
to correct or remove a market failure. Moreover, it suggests that good policy requires
caution and careful implementation.
PA R T V I
International
trade

C H A P T E R
20
International
trade
20.1 Introduction
In Chapter 4, we showed that trade between individuals can be mutually

beneficial for all parties involved. This is also true at an international level:
there can be gains from trade if countries produce goods in which they have
a comparative advantage, and then trade those goods with other countries.
In this chapter, we extend our analysis of international trade further, to
examine what determines whether a country is an exporter or importer of a
particular product. We also examine the welfare effects of international trade,
as well as the effects of government policies with respect to trade, including
tariffs and quotas on imports.
20.2 The welfare effects of international trade
In our analysis, we will focus on the effects of international trade upon a single
country. This requires us to distinguish between the country’s market for a
good (‘the domestic market’) and the international market for the same good
(‘the world market’). We will assume that the country is a small country,
such that market outcomes in the domestic market have no effect on the world
market’s outcomes – that is, changes in prices and the quantity traded in the
domestic market do not affect world prices.1
20.2.1 Welfare under autarky

Let us first suppose that the country does not trade with the world market (often
called autarky). In Chapters 6–9, we derived the demand and supply curves
and hence determined the market equilibrium. Now, we will assume that these
190 20 International trade
Sd CS
PS
Pd
Dd
Q
Qd
FIGURE 20.1 The market equilibrium for a country in autarky is given by the intersection of the domestic
demand curve and the domestic supply curve. In this figure, this is denoted by (Qd , Pd )
curves refer to domestic demand (that is, the demand of consumers within a country)
and domestic supply (that is, the supply of producers within a country). Thus, the
intersection of the domestic demand (Dd ) and supply (Sd ) curves denote the market
equilibrium (Qd , Pd ) for the country in autarky. Figure 20.1 depicts the market out-
comes of a country in autarky, as well as consumer and producer surplus.
20.2.2 Welfare with international trade

Now, suppose the country opens up to international trade; that is, domestic consumers
and producers can now trade with consumers and producers in the world market.
Because of the small country assumption, the price at which the good is traded in the
world market (Pw ) is determined independently of the domestic market equilibrium.
This means that, from the perspective of the single country, world demand and world
supply are perfectly elastic at Pw: a domestic consumer can buy a unit of the good
from foreign producers at the price Pw , and a domestic producer can sell a unit of the
good to international consumers at the price Pw .
The relationship between the domestic equilibrium price (Pd ) and the world price
(Pw ) will determine whether the country is an exporter or an importer of the good.
Indeed, the relationship between Pd and Pw is an indicator of comparative advantage:
if the Pd < Pw , the country has a comparative advantage in producing that good; if
Pd > Pw , the country has a comparative disadvantage.
An exporting country
Suppose Pd < Pw , as depicted in Figure 20.2. If the country opens up to international
trade, the effective demand curve faced by domestic producers will change from Dd
(the domestic demand curve under autarky) to the kinked demand curve depicted in
International trade 191
Sd
Exports
Pw
Pd
Dd
Q
QD Qd Qs
FIGURE 20.2 A country that is open to international trade is an exporter if the domestic equilibrium price
(Pd ) is lower than the world price (Pw ). The effective demand curve for an exporting country is traced out
by the dashed green line
Figure 20.2. The construction of the effective demand curve can be explained as
follows. If a producer wishes to sell a unit of the good, she will try to do so for the
highest price possible. For the first QD units, the highest price is offered by the
domestic consumers. However, if more than QD units are required, foreign consumers
now offer the highest price, Pw .
The equilibrium price in the presence of international trade is given by the
intersection of the domestic supply curve and the effective demand curve; that is,
the price in the domestic market, with international trade, will be Pw .2 As the price
moves from the domestic equilibrium price (Pd ) to the world price (Pw ), illustrated in
Figure 20.2, the domestic quantity supplied will increase to QS and the domestic
quantity demanded will decrease to QD; the quantity supplied by domestic producers
now exceeds the quantity demanded by domestic consumers. The difference in these
quantities is sold to the world market; thus the country is an exporter of the good.
The welfare effects of international trade for an exporting country are depicted in
Figure 20.3. Because the price rises to Pw , domestic consumers are made worse off
for two reasons: first, the increase in price means that less surplus is received on each
unit purchased; second, consumers buy fewer units overall, meaning that surplus is
lost through the decrease in the quantity purchased. On the other hand, domestic
producers are made better off for two reasons: first, the increase in price means more
surplus is received on each unit sold; second, as they are now able to sell to the world
market, producers sell more units overall.
On the whole, total welfare increases. The loss of consumer surplus (A) is more than
offset by the increase in producer surplus (A + B). Comparing Figures 20.1 and 20.3,
there is additional surplus (B) with international trade that was not previously available
under autarky. As total surplus is higher with international trade, autarky cannot be
pareto efficient. To see this, note that a change from autarky to international trade
Sd CS
PS
Exports
Pw
B
A
Pd
Dd
Q
QD Qs
FIGURE 20.3 Consumer surplus and producer surplus for an exporting country
generates sufficient extra surplus for producers such that they can (at least theoretically)
compensate consumers fully for their loss in surplus (A) and still be better off by B.
An importing country
Now suppose Pd > Pw , as depicted in Figure 20.4. If the country opens up to
international trade, the effective supply curve faced by domestic consumers will
change from Sd (the supply demand curve under autarky) to the kinked supply curve
depicted in Figure 20.4. The construction of the effective supply curve can be explained
as follows. If a consumer wishes to buy a unit of the good, he will try to do so for the
lowest price possible. For the first QS units, the lowest price is offered by the domestic
Sd
Pd
Pw
Imports
Dd
Q
Qs Qd QD
FIGURE 20.4 A country that is open to international trade is an importer if the domestic equilibrium price
(Pd ) is higher than the world price (Pw ). The effective supply curve for an importing country is traced out
by the dashed green line
producers. However, if more than QS units are required, foreign producers now offer
the lowest price, Pw .
The equilibrium price in the presence of international trade is given by the
intersection of the domestic demand curve and the effective supply curve; that is, the
price in the domestic market, with international trade, will be Pw .3 As the price moves
from the domestic equilibrium price (Pd ) to the world price (Pw ), there will be changes
in the domestic quantity demanded and the domestic quantity supplied. As seen in
Figure 20.4, the domestic quantity supplied will decrease to QS and the domestic
quantity demanded will increase to QD. The quantity demanded by domestic consumers
now exceeds the quantity supplied by domestic producers. The difference in these
quantities will be supplied by the world market; thus the country is an importer of
the good.
The welfare effects of international trade for an importing country are depicted in
Figure 20.5. Because the price falls to Pw , domestic producers are made worse off for
two reasons: first, the decrease in price means that less surplus is received on each unit
sold; second, producers sell fewer units overall, meaning that surplus is lost through
the decrease in the quantity sold. On the other hand, domestic consumers are made
better off for two reasons: first, the decrease in price means that more surplus is received
on each unit purchased; second, as they are now able to buy from the world market,
consumers buy more units overall.
On the whole, total welfare increases. The loss of producer surplus (A) is more than
offset by the increase in consumer surplus (A + B). Comparing Figures 20.1 and
20.5, we can see that there is additional surplus from international trade (B in the
figure) that was not previously available under autarky. Given total surplus is larger
with international trade, autarky cannot be a Pareto efficient outcome. Indeed, it is
(theoretically) possible for consumers to compensate producers for all of their loss
of surplus (A) given a move from autarky to international trade, and still be better
off by B.
Sd CS
PS
Pd
A
B
Pw
Imports
Dd
Q
Qs QD
FIGURE 20.5 Consumer surplus and producer surplus for an importing country
20.3 Barriers to trade
As discussed above, local producers may be negatively affected by international trade

if the price of imports (that is, the world price) is lower than the autarky equilibrium
price. Governments sometimes try to protect local industries by setting up trade
barriers that make it more difficult to import goods into the country. In this section,
we discuss two governmental policies designed to act as barriers to trade: tariffs and
quotas.
20.3.1 Tariffs
A tariff is a tax for importing a good into a country, which increases the cost of
importing the good. For example, a tariff might require a foreign firm to pay $10 to
the government for every guitar it brings into the country to sell. In this section, we
examine the welfare effects of a tariff on domestic consumers, domestic producers and
the economy as a whole.
To begin, suppose that the world price of a good is below the autarky equilibrium
price, so the country is an importer of the good.4 Now, suppose the government
implements a tariff that requires foreign firms to pay t to the government for every
unit imported. As depicted in Figure 20.6, the effect of this tariff is to raise the price
of imports in the domestic market by the size of the tariff (t) from Pw to Pw + t.5
The change in the price of imports will affect the shape of the effective supply curve,
because domestic producers are now ‘competing’ with the higher price. There will
be an increase in the domestic quantity supplied from QS to Q1S. There will also be
a decrease in domestic quantity demanded from QD to Q1D. Intuitively, this is because
the tariff makes foreign producers less competitive in the domestic market, giving
Sd
PW+E Pw + t
PW Pw
Dd
Q
QS Q1S QD1 QD
FIGURE 20.6 The effect of a tariff on domestic market outcomes for an importing country. The effective
supply curve is traced out by the dashed green line
Sd CS
PS
GR
DWL
a h b
PW + t Pw + t
C A G B
PW Pw
g f e c
Dd
Q
QS Q1S Q1D QD
FIGURE 20.7 The welfare effects of a tariff
domestic firms a greater share of sales. The quantity of imports decreases from
QD – QS to Q1D – Q1S.
The welfare effects of the tariff are depicted in Figure 20.7. As a result of the tariff
and hence the increase in price, domestic consumers are made worse off for two
reasons: first, the increase in price means that less surplus is received on each unit
purchased; second, consumers buy fewer units overall, meaning that surplus is lost
through the decrease in the quantity purchased. Specifically, CS falls by the area
C+A+G+B on the figure. On the other hand, domestic producers are made better
off for two reasons: first, the increase in price means more surplus is received on
each unit sold; second, producers sell more units overall. Specifically, PS increases
by area C. Moreover, the tariff creates government surplus as the government now
receives revenue from foreign producers who import the good. Government tariff
revenue is t · [Q1D – Q1S] or area G on the figure.
Importantly, the increase in producer surplus and government surplus (tariff revenue)
does not fully offset the decrease in consumer surplus. The resulting deadweight loss
of the tariff is depicted in Figure 20.7 as areas A + B. The triangle of deadweight loss
next to the demand curve (B) arises from the decrease in the quantity demanded from
QD to Q1D and thus is often referred to as the deadweight loss from underconsumption.
The DWL from arises because for each unit between Q1D and QD there is a consumer
with a higher MB than the MC of acquiring the good (which is Pw ), but the tariff puts
a wedge between this price and the price consumers have to pay, which is Pw + t. As
a consequence, total surplus would increase if consumption was increased to QD.
The triangle of deadweight loss next to the supply curve (A) arises from the increase
in the quantity supplied from QS to Q1S, and thus is often referred to as the deadweight
loss from overproduction. This DWL from overproduction arises because between
QS and Q1S, the economy could have the good at a marginal cost of Pw – the country
can have as much of the good as it wants at the going world price. But instead of doing
this, the tariff raises the price enjoyed by domestic producers, increasing domestic
output from QS to Q1S, even though the MC of production is higher for domestic firms
than Pw over this range of output. This means that the economy is not minimizing the
cost of obtaining the good, reducing total surplus by the additional costs incurred over
and above Pw between QS and Q1S. These costs are captured by the DWL from
overproduction.
Together, the DWLs from underconsumption and overproduction make up the total
deadweight loss from the tariff. The total DWL from the tariff is A + B.
20.3.2 Quotas
A quota is a legally enforced limit on the number of goods that may be imported into
the country. For example, a government might stipulate that no more than 100 cars
may be imported into the country. In this section, we examine the welfare effects of
a quota on domestic consumers, domestic producers and the economy as a whole.
To begin, suppose that the world price of a good is below the autarky equilibrium
price, so the country is an importer of the good.6 Now, suppose the government imposes
–
a quota of Q . As depicted in Figure 20.8, the introduction of a quota will create an
additional kink in the effective supply curve, for the following reasons. As discussed
above, for the first QS units, the lowest price is offered by domestic producers; for units
beyond QS, the lowest price is offered by foreign producers. However, the imposition
–
of the quota means that no more than Q units of the good may be imported, so after
–
QS + Q units, consumers must return to domestic producers to buy the good. In other
words, the next segment of the effective supply curve is the portion of the domestic
–
supply curve above Pw . Therefore, beyond QS + Q units, the effective supply curve is
made up of the remainder of the domestic supply curve, shifted right by the size of
the quota.
Sd
SQ
Pw
Dd
Q
Q0S Q0S+Q Q1D Q0D
Q
FIGURE 20.8 The effect of a quota on domestic market outcomes for an importing country. The effective
supply curve is traced out by the dashed green line
P P
CS
PS
Sd Sd Proits to
foreign irms
SQ SQ
P P
Pw Pw
Dd Dd
Q Q
D D
Q Q1 Q Q S S
Q
1 Q QD
1 QD
FIGURE 20.9 The left panel depicts the welfare effects of a quota. Domestic producer surplus is accrued
on those units sold by domestic producers (the first few and last few units). The intermediate units are
imported and hence foreign firms make profits on these units. The right panel depicts an equivalent way
of showing the welfare effects of a quota. The relevant areas are equivalent because the domestic
supply curve is parallel to the last section of the effective supply curve
Hence, the domestic market outcome will determined by the intersection of the
domestic demand curve and the new effective supply curve (SQ–). The price of the good
–
will increase from Pw to P . The domestic quantity demanded will be Q1D; to meet this
–
demand, Q units will be imported, with domestic producers supplying the balance.
Let us now consider the welfare implications of a quota. As it happens, the effect
of a quota on the welfare of domestic consumers and domestic producers echoes the
effects of a tariff, as depicted in Figure 20.9. Domestic consumers are made worse off
for two reasons: first, the increase in price means that less surplus is received on each
unit purchased; second, consumers buy fewer units overall, meaning that surplus is
lost through the decrease in the quantity purchased. Domestic producers are made better
off for two reasons: first, the increase in price means more surplus is received on each
unit sold; second, producers sell more units overall. As with a tariff, the import quota
creates a DWL because it puts a wedge between Pw and the price the good is traded
for in the domestic market. Because domestic price is higher, there is a DWL from
underconsumption and from overproduction. The intuition is the same as with a tariff,
as discussed in Section 20.3.1.
However, under a quota system, the government does not receive any revenue
because foreign firms do not make any payments to the government for importing the
good. Instead, foreign firms make additional profits from units imported and sold in
the country because they can now sell those goods at a higher price. The size of these
additional profits is shown in Figure 20.9.
20.3.3 Tariffs vs. quotas

As we now know, a tariff acts as a barrier to international trade by increasing the price
of imports, whereas a quota places a limit on the quantity of imports. However, for a
given tariff, there is an equivalent quota that will yield the same price, quantities and
level of imports traded in the domestic market. Conversely, for every given quota, there
is an equivalent tariff that gives rise to the outcomes in the domestic market.
The key difference between tariffs and quotas is who receives the benefit from the
increase in the price of imports. As discussed above, when a tariff is in place this benefit
accrues to the government in the form of tariff revenue. By contrast, when a quota is
in place, the benefit is enjoyed by the importer, who can buy a good on the international
–
market for Pw and sell it on the domestic market for P . In this case, however, the
government can capture this extra surplus by requiring foreign firms to purchase an
–
import licence at a price of P – Pw per unit imported. An equivalent result would result
if the government allocated the licences through a competitive auction process; the
–
price for a licence to import a good would be (approximately) P – Pw per unit. As a
consequence, the government would capture the whole of the foreign firms’ surplus,
making the quota scheme equivalent to a tariff. Importantly, however, the two
deadweight loss triangles remain.
20.4 Arguments against free trade
In our analysis above, we found that international trade increased total surplus overall
and that the imposition of governmental barriers to trade resulted in deadweight loss.
Despite this, several arguments are often made against free trade.
• Infant industries. It is sometimes argued that protection is necessary in the short

term to help a domestic industry develop. Once the industry has had a chance to
establish itself, the barriers to trade can be removed allowing the industry to
successfully compete on the world market. However, evidence from real markets
suggests that this strategy has not been successful; for example, tariffs on
manufactured goods like cars remained in place for decades, often with limited
success of generating an internationally competitive industry.
• Strategic trade policy. Another argument is that trade protection can allow a
country to manipulate the international trading environment in its own interest.
Realistically, however, most countries will not be large enough to unilaterally affect
the world market. Second, in order to implement this strategy, governments would
need to perfectly understand the behaviour of private market firms and other
governments around the world, which is near impossible.
• Anti-dumping measures. Dumping occurs when foreign firms sell their goods
in a country at a price below cost, with the intent of forcing local producers out
of business. This is done so that the foreign producer can capture all of the domestic
market. In principle, certain types of trade protection may be effective at preventing
foreign firms from dumping goods into a country, either by restricting the quantity
of imports or by setting a minimum price at which those goods can be sold.
• Environmental standards. Sometimes, it is contended that domestic firms face
higher costs of production because they have to comply with higher environmental
standards than do foreign firms. While this may be true, it is not clear that
imposing trade barriers is the best way to deal with this issue, particularly when
it is possible to address environmental externalities more directly.
• Employment protection. Another argument made for trade protection is that
barriers are needed to promote employment in the protected industries. While trade
barriers do have this effect, they also divert employment away from other sectors
in the economy, particularly those where the country might genuinely have a
comparative advantage.
In this chapter, we analysed the effect of international trade and barriers to trade on
market outcomes and welfare. We found that free trade increases welfare, while
restrictions on free trade tend to decrease total surplus. We also examined the strength
of some arguments in favour of trade protection.
Notes
1 In many markets, this is a reasonable assumption for a country like Australia. Given its
size, Australia’s contribution to world trade is often dwarfed by the USA, Europe and Asia.
2 Alternatively, the Pw line becomes the perfectly elastic demand and supply curves faced
by domestic producers and consumers. This means that any market participant can buy
and sell as much as they want at P – w. Consequently, domestic producers will not accept
a lower price than Pw per unit, and domestic consumers cannot be charged any more than
this price. Thus, with international trade the equilibrium price will be Pw .
3 With international trade, domestic consumers can buy as much as they want to at Pw .
Similarly, domestic producers can always opt to sell into the world market at Pw , so they
will not accept a lower price than this. Hence, the equilibrium price in the domestic market
allowing for international trade will be Pw .
4 Of course, if the country is an exporter of the good, a tariff will have no effect.
5 This is akin to the effect of imposing a tax on producers where the supply curve is perfectly
elastic.
6 Again, if the country is an exporter of the good, a quota will have no effect on limiting the
quantity of imports.

PA R T V I I
Review

C H A P T E R
21
Questions and
answers
Questions
Chapter 1
1. What is opportunity cost?
2. Opportunity cost only includes explicit costs? Discuss.
3. Which of the following activities involve an opportunity cost for a
consumer?
a. Having a meal tomorrow night that costs $20.
b. Taking time next weekend to go for a walk.
c. Sleeping tonight.
d. Starting a new university degree next year.
e. All of the above.
4. Which of the following is not a sunk cost.
a. The tax a firm needs to pay given it sold one of its properties.
b. An advertising campaign from last year.
c. The non-refundable deposit already paid to secure a new factory.
d. Time spent by senior management at an upcoming corporate
retreat.
5. On a Saturday night, Suzanne went to see a band play some music. If
Suzanne had not gone to see the band she would have gone to see a movie.
Her third preference was to do some study for her new graduate diploma.
Which statement is true?
a. The opportunity cost of seeing the band is missing the movie.
b. Because the movie is still running at the cinema, the opportunity cost
of seeing the band was not studying.
204 21 Review
c. The opportunity cost of seeing the band is the benefit of going to the movie,
net of the benefit of studying.
d. With the above information it is not possible to determine the opportunity cost
of going to the band.
e. The opportunity cost of seeing the band is the benefit of seeing the movie and
studying.
6. Give an example of two variables that move together, but that are otherwise
unrelated. Relate your answer to correlation and causation.
Chapter 2
1. A line passes through a point (40, 80), where q equals 40 and P is 80. It also passes
through a point (180, 10). What is the equation of this line?
2. Differentiate C (q) = 100 + 10q + 0.5q2
3. Demand is given by q = 100 – 2P. What is the price elasticity of demand when
q = 20?
4. Demand is given my P = 120 – 2q, where P is price and q is the quantity
demanded. Supply is given my P = 2q. Solve these simultaneous equations.
5. Profit for a firm ␲ is given by ␲ (q) = 80q – 2q2, where q is the number of units
produced and sold. Find the number of units that maximizes profit.
Chapter 3
1. In a Nash equilibrium:
a. All players adopt their best strategies, given the strategies adopted by all other
players.
b. Each player is doing the best they can, given the strategies of every other player
and the possible payoffs.
c. There are no possible unilateral profitable deviations for any player in the
game.
d. All of the above.
e. Total surplus is maximized.
2. Two supermarkets, Coles and Woolworths simultaneously choose the price of a
litre of milk for the coming week. The options for each supermarket are HIGH
and LOW. If both choose to price HIGH the payoffs are 15 to each firm. If Coles
chooses LOW and Woolworth opts for HIGH, the payoffs are 4 to Coles and 8 to
Woolworths. If Coles chooses HIGH and Woolworths LOW, the payoffs are 8
and 4 to Coles and Woolworths, respectively. Finally, if both firms choose LOW,
each gets 7. Considering the Nash equilibrium of the game, which statement is
true? (The first strategy in the parentheses is Coles strategy; the second strategy
is Woolworths.)
Questions and answers 205
a. The Nash equilibrium is (HIGH, HIGH); this game is not a prisoner’s dilemma.
b. The Nash equilibrium is (LOW, HIGH); this game is not a prisoner’s dilemma.
c. The Nash equilibrium is (HIGH, LOW); this game is a prisoner’s dilemma.
d. The Nash equilibrium is (LOW, LOW); this game is a prisoner’s dilemma.
e. The Nash equilibrium is (HIGH, HIGH); this game is a prisoner’s dilemma.
3. Coke and Pepsi are in a market together in which they can each make the following
sequence of moves. First, Coke decides whether to launch a SPORTS drink or an
ENERGY drink. Second, Pepsi observes Cokes choice then chooses SPORTS and
ENERGY. If ENERGY is chosen by both firms, Coke gets 20 and Pepsi 60. If
Coke chooses ENERGY and Pepsi SPORTS the payoffs are 50 and 30 to Coke
and Pepsi, respectively. If Coke opts for SPORTS and Pepsi opts for ENERGY
the payoffs are 50 and 30 to Coke and Pepsi. Finally, if both choose SPORTS the
payoffs are 30 to Coke and 50 to Pepsi. In the subgame perfect (or credible)
equilibrium, what do we observe each firm do?
a. Coke opts for ENERGY and Pepsi SPORTS.
b. Coke opts for SPORTS and Pepsi ENERGY
c. Coke opts for SPORTS and Pepsi SPORTS.
d. Coke opts for ENERGY and Pepsi ENERGY.
e. None of above.
4. In a dominant strategy equilibrium:
a. Each player reduces the surplus of the other player with the strategy that they
adopt.
b. Total surplus (the sum of the payoffs to every player) is not maximized.
c. The best strategy for each player does not depend on what the strategies of
the other players in the game.
d. Total surplus is maximized.
e. None of the above.
5. On a peaceful Sunday afternoon, Vlad and Guillaume each simultaneously decide
whether to go to Bondi (B) or Potts Point (PP). If both choose to go to Bondi they
get –5 each. If both go to Potts Point they each get a payoff of –1. If Vlad chooses
Bondi and Guillaume Potts Point the payoffs are 3, 2 to Vlad and Guillaume,
respectively. If Vlad goes to Potts Point and Guillaume Bondi, the payoffs are 0
to each party. What are the Nash equilibria of this game?
6. Australia and Japan are negotiating a free trade agreement. At the final stages of
these negotiations there are only two stages remaining. First, Japan can agree to
include beef IN the free trade agreement, or to leave beef OUT. Following Japan’s
choice of IN or OUT regarding beef, this choice is observed by Australia and they
can choose to either AGREE or to NOT agree with the proposal. The payoffs
are as follows. If Japan chooses IN and Australia chooses AGREE the payoffs are
(90, 100), where the first payoff is for Japan and the second is Australia’s payoff.
If Japan chooses IN and Australia NOT the payoffs are (20, 20). If Japan opts for
OUT and Australia still AGREEs with the proposal, the payoffs are (100, 50).
206 21 Review
Finally, if Japan chooses OUT and Australia chooses NOT to agree the payoffs
are (20, 20). What are the actions observed in the subgame perfect (or credible)
equilibrium outcome in this game?
7. Australia and Japan are negotiating a free trade agreement. At the final stages of
these negotiations there are only two stages remaining. First, Japan can agree to
include beef IN the free trade agreement, or to leave beef OUT. Following Japan’s
choice of IN or OUT regarding beef, this choice is observed by Australia and they
can choose to either AGREE or to NOT agree with the proposal. In the case that
Australia chooses NOT, it will be free to sign up to a free trade agreement with
Korea (with the payoffs for Australia detailed below). The payoffs are as follows.
In Japan chooses IN and Australia chooses Agree the payoffs are (90, 100), where
the first payoff is for Japan and the second is Australia’s payoff. If Japan chooses
IN and Australia NOT the payoffs are (20, 60). If Japan opts for OUT and
Australia still AGREEs with the proposal, the payoffs are (100, 50). Finally, if
Japan chooses OUT and Australia chooses NOT to agree the payoffs are (20, 60).
What are actions observed in the subgame perfect (or credible) equilibrium
outcome in this game?
8. Martin and Jim, two legendary guitarists, must simultaneously choose to play ‘On’
the beat, or ‘Off’ the beat. If both guitarists play On the beat, each gets a surplus
of 10. If Martin plays On and Jim plays Off, the payoffs are 50 to each player. On
the other hand, if Martin plays Off and Jim On, the payoffs are 40 to each player.
Finally, if they both get 5 if both decide to play Off. What are the Nash equilibria
in the game?
9. Malcolm and Angus are two other guitarists who have to simultaneously decide
to play ‘Lead’ or ‘Rhythm’. If both play Lead, the payoffs are 2 to each player.
If both opt for ‘Rhythm’ each player gets 10. If Angus plays Lead and Malcolm
Rhythm, the payoffs are 100 to Angus and 120 to Malcolm. If Angus plays Rhythm
while Malcolm opts for Lead, the payoffs are 80 to both players. What are the
Nash equilibria of the game?
Chapter 4
1. Draw the production possibility curve for an economy that can produce either
tractors or bicycles.
a. Show a point in which the economy is not producing efficiently.
b. Show a point that is unattainable given the current state of technology.
c. Show a point where the economy is producing efficiently.
d. Relate shifts along the production possibility curve to the concept of
opportunity cost.
2. There are two countries, Australia and the Rest of the World, that can produce
either good A or good B. Australia is more productive at producing both goods.
Which statement is not true?
a. Australia has an absolute advantage in producing both goods.

b. The benefits of trade accrue to the country that does not have the absolute
advantage in the traded good.
c. Comparative advantage determines the patterns of production between the
countries.
d. The opportunity cost of producing good A is good B.
3. Bob can produce either good x or good y. In one hour he can produce 4 units of
good x. Alternatively he can produce 2 units of good y in that hour. Suzie can
produce either 6 units of good x in an hour or 2 units of good y in an hour.
a. What is absolute advantage? Who has the absolute advantage in producing x
and which party has the absolute advantage of producing good y?
b. What is comparative advantage? Which party has the comparative advantage
of producing good x and good y?
c. If each party has 10 hours in which they can work, draw the production
possibilities frontier for each party. What does the slope of each production
possibilities frontier represent?
d. What is the maximum price that Bob is willing to pay for a unit of good x?
What is the minimum price that Suzie would be willing to sell a unit of good
x for? What is the minimum price that Bob would be willing to sell a unit of
good y for? What is the maximum price that Suzie would be willing to pay
for a unit of good y?
4. Now reinterpret question 3 with Bob and Suzie: rather than both parties producing
goods x and y, instead consider the situation in which Bob and Suzie undertake
tasks x and y at a workplace. Which party should do task x? Which party should
perform task y? What is different about this case than if we are considering two
goods (or services) being traded in the market?
5. Australia can produce two goods – coffee and TVs. Consider two ways in which
the Australian economy can grow: through an increase in population, and through
technological progress. Illustrate both of these changes on a PPF. What is the
advantage of the second type of growth over the former?
6. Take an economy with two workers, Romi and Debra. There are two goods that
can be made: Toys and Balls. Each person works an 8-hour day. Romi can produce
a toy in 20 minutes and a ball in 2 hours. Debra can make 1 toy in 4 hours. It takes
Debra 2 hours to make one ball. Initially, both Romi and Debra spend half of their
time making toys and half of their time making balls. Each person consumes what
they make.
Now consider the following trade. Romi spends six hours making toys, and the
remainder of her time making balls. Debra spends all her time making balls. They
then trade – Romi trades 3 toys for 1 ball from Debra. Show that: total production
of both goods increases following trade; and, both are better off with trade than
without it. Explain your answer.
208 21 Review
Chapter 5
1. Consider the following bargaining situation between a potential car buyer, called
Davo, and a car dealer, Shazza. Shazza has a Falcon for sale that Davo values at
$20,000. Shazza values the car at $0. The bargaining scenario is as follows. Shazza
makes a proposal to Davo that he buy the Falcon for a particular price. If Davo
accepts Shazza’s proposal, trade takes place at the suggested price. If Davo refuses
Shazza’s offer, negotiations immediately break down between the two parties.
Instead, Davo walks across the road and begins negotiating with another car dealer,
Maggie. Maggie has a Corolla for sale that Davo values at $4,000; Maggie values
the Corolla at $0. Negotiations between Maggie and Davo proceed as follows:
Davo makes an offer to buy the car for a particular price. If Maggie accepts trade
takes place at that price. If Maggie rejects the offer, no trade takes place and
bargaining between the parties ends. Without a new car Davo’s utility is $0. If
Maggie does not sell the Corolla to Davo her payoff is $0.
a. Illustrate this bargaining scenario in a diagram. How do you work out the
outcome of this bargaining situation and why?
b. Assume that bargaining reaches the stage at which Davo is bargaining with
Maggie. What is the price that Davo offers? What is Maggie’s response?
c. Now consider when Shazza is bargaining with Davo. What happens and
why?
d. What is the outcome of this bargaining situation? Compare the surplus that
accrues to the buyer and seller in this outcome with the surplus maximizing
outcome.
2. Ansett, a bankrupt airline, and its administrator have to sell the company assets.
The scrap value for the assets (the planes, its spare part inventory, and so on) is
$100M. Ansett has only one potential buyer for the airline as an ongoing concern
– Singapore Airlines – and negotiations must be completed by close of business
in two days time. This is enough time for two rounds of offers. Negotiations
proceed as follows. Ansett makes an offer that Singapore can accept or reject. If
Singapore accepts, trade takes place at that price. In this case Ansett has a payoff
of the agreed price and Singapore has a payoff of its value ($350 million) minus
the price. If Singapore rejects the offer, the company representatives go back to
their hotels for the night. The following day Singapore Airlines gets to make an
offer that Ansett can accept or reject. If Ansett accepts, trade takes places at that
price. If Ansett rejects, no trade takes place between the companies because they
have reached the bargaining deadline. In this case Singapore Airlines gets a
surplus of $0 and the Ansett company is scrapped, so that Ansett gets a payoff
equal to its scrap value. Singapore values the Ansett company at $350 million
minus the price it has to pay.
a. Draw a game tree to represent these negotiations.
b. Assume that negotiations reach the second day. If this is the case, what would
happen?
c. Now consider what happens on the first day of negotiation. What is the
subgame perfect outcome of this bargain between Ansett and Singapore?
3. Bazza has invented a new drug K and holds the patent for the drug for the next
two years. Bazza, not having any marketing skills, values the patent at $0 in each
of the two years. Raelene, on the other hand, can make $100 in the first year if
she holds the patent and $80 in the second year if she holds the patent. Without
the patent Raelene gets a payoff of $0 in each year. The two parties enter into
negotiations with each other regarding the possibility of Raelene buying the patent
from Bazza. The negotiation process is as follows: At the start of year 1 Bazza
gets to make an offer to Raelene regarding the price of the patent. If Raelene accepts
she buys the patent at the asking price, and gets to use it for the next two years.
If Raelene rejects the offer negotiations are put on hold for a year. At the start of
year 2 Raelene then gets to make an offer to Bazza for the patent. If Bazza
accepts, trade takes place at that price and Raelene gets to use the patent for year
2 only. If Bazza rejects the price offer negotiations cease and both parties receive
a payoff of $0.
What is the outcome in this bargaining game?
4. Consider the following bargaining model. Two firms in different countries – A
and B – are considering trading with one another. Firm A has an input of production
(i.e. a computer chip) that Firm B would like to use in its new plant. A values the
computer at 0; if B gets the computer it is worth $25 for each year that it gets to
use it. There are two years (after which the computer is worth $0 to everyone).
The negotiating situation proceeds as follows. Executives of firms A and B meet
at the start of the first year, and firm A makes an offer to firm B. B can either
accept or reject the offer. If the offer is accepted, A gets the agreed price and B
pays the price and gets to use the computer for two years. If B rejects the offer,
the executives from each firm part company and do not meet until the start of the
second year. At the start of the second year, B gets to make an offer to A. If A
accepts, A gets a payoff of the agreed price and B gets the value of using the chip
for one year, net of the price paid.
Solving backwards, determine the initial price offer by A and argue whether or
not it will be accepted by B.
Chapter 6
1. Why is a typical consumer’s demand curve downward sloping?
2. What causes a movement along a demand curve (or a change in the quantity
demanded)?
a. An unanticipated announcement of the health benefit of the good.
b. A change in the price of the good itself.
c. A change in consumers’ income.
d. An increase in the price of a substitute product.
210 21 Review
3. A consumer’s marginal benefit curve is given by MB = 10 – q, where q is the

quantity consumed. What is the consumer’s demand curve. What is the maximum
price this consumer is willing to pay for the first unit? What is the maximum price
she is willing to pay for the 7th unit?
4. What causes a change in demand (or a shift in the demand curve)?
a. An unanticipated announcement of the health benefit of the good.
b. A decrease in the price of a complementary product.
c. A change in consumers’ incomes.
d. An increase in the price of a substitute product.
e. All of the above
5. What is the market demand curve if there are three consumers in the market each
with individual demand curves of P = 20 – q?
6. What is the market demand curve if there are five consumers in the market each
with individual demand curves of P = 10 – 2q?
Chapter 7
1. Which statement is true?
a. Marginal cost intersects the maximum of average variable cost from above.
b. Marginal cost insects average total cost at the minimum of marginal cost.
c. Marginal cost intersects the minimum of average fixed cost from below.
d. The relationship between marginal cost and average total cost cannot be
determined without specific knowledge of the particular production process.
e. Marginal cost intersects the minimum of average variable cost from below.
2. Arthur is considering becoming chair of a company board that will pay $200,000
a year. Taking this position will require Arthur to give up another directorship
that pays $75,000 a year. The job will involve travel anticipated to have an out-
of-pocket cost of $20,000 a year. It will also require Arthur to buy $150,000 of
shares in the company (which he must sell back to the company for $150,000 when
he ceases to be chair of the board); currently Arthur has the money invested in
the bank earning an interest rate of 10 per cent per year. What is Arthur’s
anticipated economic profit in the business’s first year of operation?
a. $80,000
b. $90,000
c. –$25,000
d. $110,000
e. –$40,000
3. Sascha, a bank teller who works Monday to Friday, is considering buying a new
lounge. Sascha can get the lounge from AllenKey Inc for $150, but this lounge
needs to be delivered (an extra $30) then assembled by Sascha. The only time
Sascha has to assemble the lounge is on a Sunday afternoon. It would take Sascha
three hours to build the lounge, during which time she would otherwise watch
the Aussie rules footy; Sascha loves her footy and would pay up to $100 to see
the game. Sascha’s hourly wage at the bank is $50 an hour. From experience,
Sascha also expects to experience a personal disutility (or cost) from frustration
of $40 and anticipates that she will most likely break her partner Alex’s favourite
vase (replacement cost $75) in the process of putting the lounge together.
As an alternative, Sascha could buy an identical lounge from LazyBones Inc,
which comes full constructed, delivered and installed at one price. What is the
minimum price of the lounge on sale at LazyBones for which Sascha will choose
to buy from AllenKey Inc?
a. $295
b. $445
c. $545
d. $395
e. None of the above
a. Marginal cost intersects average total cost when marginal cost is at its
maximum.
b. Marginal cost intersects average total cost when marginal cost is at its
minimum.
c. Marginal cost intersects average variable cost when average variable cost is
at its maximum.
d. Marginal cost intersects average variable cost when average variable cost is
at its minimum.
e. Marginal cost intersects average fixed cost when average fixed cost is at its
minimum.
5. Explain why in the long run: there are no fixed costs; a firm’s long-run average
cost is the lower envelope of its possible short-run average total cost curves;
and a firm’s long-run marginal cost cannot be higher than its short-run marginal
cost.
6. A firm’s cost curve is TC(q) = 250 + 10q + 2q2, q is the quantity produced. What
is the firm’s marginal cost (MC), average variable cost (AVC) and its average total
cost (ATC)?
7. Draw the typical short-run cost curves for a competitive firm, noting in particular
the relationship between average variable costs, average total costs and marginal
costs. Explain the shape of each curve. What is the difference between the long
run and the short run? Again using a typical firm, show how to derive its long-
run average total cost curve and show examples of economies of scale, dis-
economies of scale and constant returns to scale on your diagram. Finally, for a
given level of output (and assuming constant prices and technology), can a firm’s
costs ever be higher in the long run than in the short run?
212 21 Review
Chapter 8
1. A firm’s supply curve is its marginal cost curve. True or false?
2. Explain why a typical firm’s supply is upwards sloping.
3. Which change will cause a movement along the supply curve (or a change in the
quantity supplied)?
a. An increase in the price of a key input.
b. A change in the production technology.
c. A change in the anticipated price of an input next year.
d. None of the above.
Chapter 9
1. Consider the markets for corn chips and potato chips, two substitutes. We observe
the following facts:
• The price rises and the quantity demanded falls in the corn chip market.
• The price rises and the quantity demanded rises in the market for potato chips.
Which of the following could explain these observations?
a. Dry conditions in potato growing areas hinder the cultivation of potato crops.
b. Frost in corn growing areas hinders the growing of corn.
c. Good rains assist potato cultivation.
d. Damp conditions and warm weather assist corn production.
2. Beer and potato chips are complementary products that are both sold in a
competitive market. Barley is an essential ingredient in the production of beer;
potatoes are an input into making chips. We observe that: price increases and the
quantity traded decreases in the beer market; and both the price and the quantity
traded decrease in the chip market. Which of the following scenarios is consistent
with the observation in the two markets, outline above?
a. An increase in irrigated lands has allowed for an increase the supply of
potatoes.
b. A bumper year with perfect conditions in the grain growing regions increase
yields of barley.
c. Low rainfall reduces the barely yields in grain growing regions.
d. A health campaign by the Federal government has the effect of curtailing beer
consumption.
e. Floods in potato growing regions reduce potato production.
3. Consider the world markets for lamb and beef. Assume that in both markets the
laws of demand and supply hold. What will be the effect of the discovery of mad-
cow disease (which can be transmitted to humans by eating infected products)?
a. The quantity of lamb sold in the market increases; the price for beef decreases.
b. The quantity of lamb sold in the market decreases; the price for beef rises.
c. The quantity of lamb sold in the market decreases; the price for beef decreases.
d. The quantity of lamb sold in the market increases; the price for beef increases.
4. Consider a market in which the laws of demand and supply hold. Which of the
following would result in an increase in equilibrium price and an ambiguous change
in equilibrium quantity?
a. An increase in supply and an increase in demand.
b. An increase in supply and a decrease in demand.
c. A decrease in supply and an increase in demand.
d. A decrease in supply and a decrease demand.
5. Consider a perfectly competitive market with a demand curve given by P = 100
–qd , where P is market price and qd is the quantity demanded. The market supply
curve is given by P = 3qs, where qs is the market quantity supplied.
a. Calculate the market equilibrium price and quantity. Illustrate your answer
using a diagram.
b. What is Consumer Surplus (CS)? What is Producer Surplus (PS)? Calculate
the CS and PS and the total surplus generated in the competitive market, again
illustrating your answer using a diagram. How does this competitive market
equilibrium compare to the outcome that maximizes total surplus? Provide
some intuition for your answer.
6. In a market demand is given by P = 21 – q and supply by P = 2q. In the
competitive-market equilibrium, what are the CS and PS?
Chapter 10
a. The cross-price elasticity for complementary goods is positive.
b. The price elasticity of demand is equal to –1 along any linear demand curve.
c. The price elasticity of demand is constant along a linear demand curve.
d. The price elasticity of demand is equal to 0 if the demand curve is perfectly
elastic.
e. The cross-price elasticity for substitute goods is positive.
a. If a market demand curve is linear (a straight line), revenue in the market is
maximized when the price elasticity of demand is unit elastic.
b. If a market demand curve is linear (a straight line) the price elasticity of
demand is constant along the length of the demand curve.
c. If a market demand curve is linear (a straight line) the price elasticity of
demand can be either elastic or inelastic along the whole curve, depending
slope of the curve.
214 21 Review
d. If a market demand curve is linear (a straight line), revenue is maximized when

the demand curve intersect the price axis.
e. If a market demand curve is linear (a straight line), revenue is maximized when
the demand curve intersect the quantity axis.
3. For good Y, when the price is P1 = $60 the quantity demanded is q1 = 10 units.
When the price falls to P2 = $40 it is determined formula that the elasticity is –3
according to the initial point elasticity method (use the initial point as the reference
point for calculating the proportional changes in P and q). What is the quantity
demanded, q2, when the price is $40?
a. q2 = 0
b. q2 = 20
c. q2 = 30
d. q2 = 40
4. In a market demand is given by P = 100 – 2qd , where P is price and qd is the
quantity demanded. The supply curve is given by P = 2qs , where qs is the quantity
supplied. At the market equilibrium, what is the elasticity of supply?
a. 0
1
b. ⁄2
c. 1
d. 2
e. 3
5. In a market demand is given by P = 100 – 2qd , where P is price and qd is the
quantity demanded. The supply curve is given by P = 2qs , where qs is the quantity
supplied. At the market equilibrium, what is the price elasticity of demand?
6. A consumer always spends one third of her income on computing services. Using
the point-method, in this case the income-elasticity of demand for computing
services is:
a. 0
b. 0.33
c. 1
d. 3
7. In a market, when the price increased the total expenditure on the good also
increased. Is demand elastic, inelastic or unit elastic in this region? Explain your
answer.
Chapter 12
1. In a perfectly competitive industry, each firm has a total cost function of TC =
400 + 10q + q2 and a marginal cost curve of MC = 10 + 2q if it produces a positive
quantity of output q. If a firm produces zero output it has no costs. The market
price is $50. Which statement is true?
a. Each firm produces 20 units of output; the industry will require entry to reach
its long-run equilibrium.
b. Each firm is producing 25 units; as the firm is making short-run profits, the
industry is not at its long-run equilibrium.
c. Each firm is producing 25 units; the firm is covering its variable costs, but
making a short run loss.
d. Each firm is producing 20 units; the firm will continue producing in the short
run, but will consider exiting in the long run as it is not covering its total costs
of production.
e. Each firm produces 20 units of output; the market is in its long-run equilibrium.
a. In the long run, if the long run supply curve in a competitive industry is
horizontal, the supply curve for each individual firm in the industry is also
horizontal.
b. The long-run supply curve in a constant cost industry is horizontal at the
minimum of average total cost; individual firm supply curves are upward
sloping, being their marginal cost curves when price is above the minimum
of average total cost.
c. The supply curve of a competitive firm in the long-run in a constant cost
industry is horizontal.
d. Both b and c are correct.
3. The short-run marginal and average variable cost curves for a competitive firm
are given by MC = 8 + 8Q and AVC = 8 + 4Q, respectively. The profit-maximizing
level of output (Q) is 2 and the total fixed cost (TFC) is $64. Which of the following
must be true about the firm?
a. The firm is charging a price of $40 and covering its average variable cost,
hence it should continue operating in the short run.
b. The firm is charging a price of $40 and making a short-run loss, and hence
the firm must shut down immediately.
c. The firm is charging a price of $24 and making a zero profit, and hence the
firm should shut down eventually.
d. The firm is charging a price of $24, covering its average variable costs, but
in the long run at this price the firm should exit in the long-run.
a. Marginal cost intersects average fixed cost when average fixed cost is at its
minimum.
b. In the short run, a perfectly competitive profit maximizing firm that has not
shut down is not operating on the upward-sloping portion of its AVC.
c. A perfectly competitive profit maximizing firm whose long run economic profit
is exactly zero should expand production to try to earn a positive profit.
216 21 Review
d. It is possible for average variable cost to be greater than average total cost at
high levels of output.
5. Consider the Sydney taxi market. Assume that the market is competitive with free
entry. Further, assume that each firm has an increasing marginal cost curve and
u-shaped average variable cost and average total cost curves. The market is
initially in long-run equilibrium.
a. Using a diagram, show and explain a firm’s long-run supply curve. What will
the price be in the long run and what profits will the firm make? Explain your
answer.
b. Due to an increase in population, there is an increase in the market demand
for taxis. What will happen to price, the quantity the firm sells and its profit
in the short run? Use a diagram to help explain your answer.
c. Following this demand increase, the NSW government does two things: it bans
any new firms (taxis) from entering the market – that is, there are no more
taxi allowed; and, to raise revenue, it starts charging taxi owners for the right
to have a taxi (often called a licence fee). With no new entry, what will be
the market price and the quantity sold by each firm? What would be the
maximum licence fee a firm would be willing to pay for the rights to drive a
taxi in this market with no new entry.
a. In the long run in a constant-cost industry, a firm has a perfectly elastic supply
curve.
b. In the long run in a constant-cost industry, firms have an upward-sloping
supply curve and face a downward-sloping demand curve.
c. In the long run in a constant cost industry, firms have an upward-sloping supply
curve and the industry supply curve is the horizontal summation of the supply
curves of firms currently in the industry.
d. In the long run in a constant cost industry, firms have a perfectly elastic supply
curve and the long-run industry supply curve is horizontal.
e. In the long run in a constant-cost industry, firms have an upward sloping supply
curve and the industry supply curve is perfectly elastic.
7. Assume a constant-cost industry with free entry is initially at its long-run
equilibrium. Following an unanticipated increase in demand:
a. In the short run, firms currently in the market face a higher price, make short-
run profits; in the long run, entry reduces price somewhat, but positive
economic profits for the firms initially in the market remain.
b. In the short run, firms incur higher marginal costs for the last good produced
but make positive economic profits; in the long run the firms maximize profit
by selling a quantity so that MC for the last unit sold is the same as it was in
the original long-run equilibrium (before the increase in demand).
c. In the short run firms do not change the quantity they sell, but they earn positive
economic profits; in the long run economic profits are zero.
d. In the short run each firm increases its output it sells and makes positive profits;
in the long run entry causes profits to return to zero but each firm in the industry
sells a higher quantity than in the initial long-run equilibrium (before the
increase in demand).
e. All firms continue to make zero profits in both the short and long run.
Chapter 13
1. A monopolist has a constant MC = 20 per unit produced, has no fixed costs and
faces a demand curve of 2q = 100 – P. If the monopolist sells at a single price,
which statement is true about the monopolists profit-maximizing quantity and price
and its profit?
2. Consider a monopolist with MC = 2q and a fixed cost of FC = $50, facing a market-
demand curve of P = 40 – q. What is the profit of a single-price monopolist?
3. Which statement is FALSE?
a. A monopolist changing a linear (or single) price raises price above the MB of
the final good sold.
b. A monopolist regulated with marginal-cost pricing regulation produces no
deadweight loss (assuming it produces, and does not exit the industry).
c. A monopolist charging a linear (single) price sets its profit-maximizing
quantity where MR = MC.
d. A monopolist regulated by average total cost pricing regulation, ensuring that
it earns zero profits, produces no deadweight loss.
e. A monopolist with positive marginal costs and facing a linear demand curve
always sets a quantity (or price) such that it sells on the elastic section of the
demand curve.
4. Consider a monopoly market with a demand curve of P = 60 – q. The monopolist
has a marginal cost of production of MC = q. If the monopolist charges a linear
price (a single-price monopolist), what is the DWL of monopoly?
a. 0
b. 80
c. 100
d. 120
e. 160
5. A scientific research lab has invented a new cancer drug. The lab is willing to sell
the patent for the drug to the highest bidder, and the holder of the patent will have
a monopoly for the drug. The drug costs $40 to produce per unit (that is, MC =
$40 per unit) and the demand curve for the medication is 1/2q = 200 – P. What
would be the maximum amount that a firm thinking would be willing to pay for
the patent?
a. 8,800
b. 12,800
218 21 Review
c. 16,000
d. 6,400
e. 4,800
6. A scientific research lab has invented a new cancer drug. The government is willing
to buy the patent and make the patent publicly available, so that any firm can
produce the drug (that is, there will be a competitive market in production of the
new drug). The drug costs $40 to produce per unit (MC = $40 per unit) and the
demand curve for the medication is 1/2q = 200 – P. What would be the maximum
that the government would be willing to pay for the patent? (Hint: the government
would like to maximize the total surplus of producers and consumers generated
by trade of this product in the market.)
a. 6,400
b. 12,800
c. 19,200
d. 12,400
e. 25,600
7. Consider a market with a demand curve given by P = 100 – qd , where P is market
price and qd is the quantity demanded. Assume that there is only one firm selling
in this market, and that this monopolist has a marginal cost curve of MC = 3q and
has no fixed costs. The monopolist charges a linear price (or a single price) to all
customers.
a. What is the monopolist’s marginal revenue (MR) curve, and its profit-
maximizing price and quantity? Explain the monopolist’s MR, and illustrate
your answer on a diagram. What is the monopolist’s profit?
b. Compare total surplus in the competitive market with the monopoly outcome.
Using economic intuition (and a diagram where appropriate) explain the
differences between the two market outcomes.
8. Consider Donna, a monopolist, selling to a market with a demand curve P =
20 – q, where P is the market price and q is the quantity demanded. Donna has a
constant marginal cost of production of $2 per unit and a fixed cost of $20.
a. If Donna charges a linear price (or a single price), what is the profit-
maximizing price and quantity, Donna’s profit and any deadweight loss that
arises. Explain your answer with the help of a diagram.
b. What is first-degree price discrimination? Assuming the demand curve given
above is the demand curve for one consumer, if Donna can engage in first-
degree price discrimination using a two-part tariff, what price, quantity, profit
and deadweight loss will result? Again, explain your answer with the help of
a diagram.
9. Which of the following is an example of second degree price discrimination?
a. A theatre that offers lower prices for children.
b. A computer firm that offer a computer and a printer in a package at a lower
price than it would cost to purchase both separately.
c. A rail service offering lower priced tickets to students than non-students.

d. All of the above.
10. Consider a monopolist producing a good with zero marginal costs and $20 fixed
costs. There are two consumers in the market. Consumer 1 has a demand curve
q1 = 10 – P, where q1 is the quantity demanded and P is the per-unit price.
Consumer 2 has a demand curve q2 = 15 – P, where q2 is the quantity demanded
by consumer 2. If the monopolist perfectly price discriminates between two
consumers using a two-part tariff.
a. The monopolist will charge a fixed fee of $100 to consumer 1 and a per unit
fee of $0; and the fixed fee of $225 to consumer 2 and a per-unit fee of $0.
b. The monopolist will charge a fixed fee of $12.5 to consumer 1 and a per unit
fee of $5; and a fixed fee of $25 to consumer 2 and a per-unit fee of $5.
c. The monopolist will not produce as it makes negative profits if it does.
d. The monopolist will charge a fixed fee of $50 to consumer 1 and a per unit
fee of $0; and the fixed fee of $112.5 to consumer 2 and a per-unit fee of $0.
Chapter 14
1. Consider the following statement. Which statements are characteristics of a
monopolistic competitive industry?
a. In the long run all firms in the industry make zero economic profits.
b. Firms sell differentiated products.
c. There is free entry in the long run.
d. All of the above
2. The entry and exit of firms in a monopolistically competitive market guarantees
that:
a. In the long run, economic profits and economic losses are driven back to zero.
b. Economic profits can survive in the long run, but no firm can make an
economic loss.
c. A firm can earn an economic losses in the long run, but not a positive
economic profit.
d. Both economic profits and economic losses will exist in the long run.
3. Jon runs one of many restaurants in the industry in a city. This industry is
monopolistically competitive and it is initially in its long-run equilibrium.
a. Assume that Jon has typical u-shaped average cost curves. Draw a diagram
illustrating Jon’s profit-maximizing output, price and profit. Explain your
answer.
b. On your diagram, show the consumer surplus generated, and any DWL.
Explain your answer.
c. The government is contemplating regulating the industry so that prices must
equal marginal cost. What would Jon’s response be to this regulation in the
long run?
220 21 Review
4. Because the average total cost of a firm is not minimized in the long run if the
market is monopolistically competitive, the level of entry is too high and it is
inefficient. Discuss.
Chapter 15
1. Consider the following game. Amazon can choose to either Enter of Not Enter
into the market for e-books. If Amazon chooses Not Enter then the payoffs are
$1 to Amazon and $6 to Apple. If Amazon chooses to Enter then Apple observes
this and chooses to either Punish or Accommodate. If Apple chooses to Punish
the payoffs are $1 to Apple and –$1 to Amazon. If Apple Accommodates the
payoffs are $2 to each party. What are all of the Nash equilibria in this game?
a. (Enter, Accommodate).
b. (Not Enter, Punish).
c. (Enter, Punish).
d. (Not Enter, Punish) and (Enter, Accommodate).
e. (Not Enter, Punish), (Not Enter, Punish) and (Enter, Accommodate).
2. Consider the following game. Amazon can choose to either Enter of Not Enter
into the market for e-books. If Amazon chooses Not Enter then the payoffs are
$1 to Amazon and $6 to Apple. If Amazon chooses to Enter then Apple observes
this and chooses to either Punish or Accommodate. If Apple chooses to Punish
the payoffs are $1 to Apple and –$1 to Amazon. If Apple Accommodates the
payoffs are $2 to each party. What are all of the subgame perfect equilibria in this
game?
a. (Enter, Accommodate).
b. (Not Enter, Punish).
c. (Enter, Punish).
d. (Not Enter, Punish) and (Enter, Accommodate).
e. (Not Enter, Punish), (Not Enter, Punish) and (Enter, Accommodate).
3. Considered the following market. Chrome can choose when launching its new
product either to do it LARGE or as NICHE. After Chrome has chosen its action,
Firefox observes Chrome’s choice and then can choose to ADAPT to RETAIN
its own product. After Firefox has chosen its action, the game ends and the
payoffs are made. The payoffs are as follows. If Chrome chooses LARGE and
Firefox ADAPT, the payoffs are 25 and 40 to Chrome and Firefox, respectively.
If Chrome goes LARGE and Firefox RETAINS the payoffs are 30 and 50 to
Chrome and Firefox. If Chrome plays NICHE and Firefox ADAPTS, the payoffs
are (40 Chrome, 30 Firefox) and if Chrome plays NICHE and Firefox RETAINS
the payoffs are (20, 20) for Chrome and Firefox, respectively.
a. What is a Nash equilibrium? Outline the Nash equilibrium or equilibria in the
game, and explain your answer.
b. What is a subgame perfect (or credible) equilibrium and how would you find
such an equilibrium? What is the outcome in the subgame perfect equilibrium
in this game? Explain your answer.
c. Does the subgame perfect equilibrium (or credible) equilibrium result in the
outcome that maximizes total surplus? If it is, explain why. If not, is there
some transfer between the two parties that can help achieve the surplus-
maximizing result? Explain your answer.
4. Sonia the Terrible lands her ship and crew of scurvy-pirate raiders onto a foreign
beach. Up on the headland is a village, led by Jen the Brave. At this point, Sonia
has a choice: she can SLASH her water barrels, eliminating her water supplies;
or she can NOT SLASH, which keeps her water supplies completely intact.
Having seen the action taken by Sonia, Jen can choose to CHARGE or to WAIT.
The payoffs are as follows: if Sonia the Terrible chooses to SLASH and Jen
CHARGEs, the payoffs are (100, 20) to Sonia and Jen, respectively. If Sonia
SLASHes and Jen WAITs, the payoffs are (10, 30). On the other hand, if Sonia
chooses to NOT SLASH and Jen CHARGEs, the payoff is (20, 15). If Sonia opts
to NOT SLASH and Jen chooses to WAIT, the payoffs are (40, 10), respectively.
What is the subgame perfect (or credible) equilibrium outcome?
a. Sonia the Terrible chooses to SLASH; Jen the Brave chooses to CHARGE.
b. Sonia the Terrible chooses to NOT SLASH; Jen the Brave chooses to WAIT.
c. Sonia the Terrible chooses to NOT SLASH; Jen the Brave chooses to
CHARGE.
d. Sonia the Terrible chooses to SLASH; Jen the Brave chooses to WAIT.
5. Consider the following game involving Sony and Nintendo. Each has to
simultaneously decide on a new platform for its new generation of games. The
two choices are Platform 1 and Platform 2. If both firms choose Platform 1 the
payoff is 20 to each firm. If both firms choose Platform 2, the payoff is 30 to each
firm. If Sony chooses Platform 2 and Nintendo Platform 1 the payoffs are 8 to
Sony and 2 to Nintendo. Finally, if Sony opts for Platform 1 and Nintendo for
Platform 2, the payoffs are 5 to Sony and 7 to Nintendo. What is Sonys preferred
outcome? What is Nintendos prefer? What are the Nash equilibrium or equilibria
of the game? Explain your answer.
6. Define Nash equilibrium. Give an example of a prisoner’s dilemma in a business
context. Explain why your game is a prisoner’s dilemma.
7. Consider the following game between firm A and firm B. In each period firm A
and B each decide at the same time whether to advertise (Adv) or not advertise
(NA). The payoffs in each period are: if both firms advertise they each get a payoff
of 2; if one firm advertises and the other chooses NA the payoffs are 6 to the
advertiser and 1 to the firm that did not advertise; and, finally, if they both choose
to not advertise each gets 5.
a. If the game only goes for one period, what is the Nash equilibrium?
b. Now assume that there are three periods; that is, the two firms play the
Adv/not advertise game three times in a row. What is the Nash equilibrium?
222 21 Review
Chapter 16
1. Consider two markets for haircuts in both Sydney and Melbourne. Supply in both
cities is perfectly elastic and at the same price. Demand in Sydney is relative
inelastic compared with demand for haircuts in Melbourne, in the relevant range.
The Federal government introduces a tax of t in both markets. Which statement
is true?
a. The deadweight loss created by the tax is larger in Sydney than in Melbourne;
the reduction in quantity is larger in Sydney than in Melbourne.
b. The deadweight loss created by the tax is larger in Sydney than in Melbourne;
the reduction in quantity is larger in Melbourne than in Sydney.
c. The deadweight loss created by the tax is larger in Melbourne than in Sydney;
the reduction in quantity is larger in Sydney than in Melbourne.
d. The deadweight loss created by the tax is larger in Melbourne than in Sydney;
the reduction in quantity is larger in Melbourne than in Sydney.
a. A positive tax always creates a deadweight loss.
b. A tax that does not raise any revenue has a zero deadweight loss.
c. If a tax is imposed on a market, the economic incidence of the tax depends
on who legally has to pay for the tax.
d. If demand is downward sloping and supply if perfectly inelastic, consumers
pay for all of a tax.
3. In a market the demand curve is given my P = 30 – 2qd , where P is price and qd
is the quantity demanded. Supply is given by P = qs . The government imposes a
per-unit tax to be legally paid by suppliers of $6 per unit. Following the imposition
of the tax, which statement is true?
a. Consumers pay $4 more than without the tax; producers receive $2 less with
the tax than before it was imposed.
b. Consumers pay $16 with the tax; producers receive a price of $10 with the
tax.
c. Producers receive $4 with the tax; consumers pay $10.
d. Consumers pay $12 with the tax; producers receive a price of $6 after the tax
is imposed.
e. Producers pay for all of the economic incidence of the tax.
4. In a market the demand curve is given my P = 30 – 2qd , where P is price and qd
is the quantity demanded. Supply is given by P = qs . The government imposes a
per-unit tax to be legally paid by suppliers of $6 per unit. Following the imposition
of the tax, what is the DWL?
5. In a market the demand function is given by qd = 100 – P, where qd is the quantity
demanded and P is the price. Supply is perfectly elastic at P = $40. The government
implements a subsidy to producers of $10 a unit. With the subsidy, what is the
DWL?
6. The deadweight loss from a per-unit tax in a product market is:
a. Positively related to the (absolute) elasticity of demand, but is not affected by
the elasticity of supply.
b. Negatively related to the (absolute) elasticity of demand and to the (absolute)
elasticity of supply.
c. Positively related to the (absolute) elasticity of demand and to the (absolute)
elasticity of supply.
d. Positively related to the (absolute) elasticity of supply and not affected by the
elasticity of demand.
e. It depends on which side of the market the tax is levied on.
7. Consider the following market. Demand is given by qd = 150 – 2P, where qd is
the quantity demanded and P is the price. Supply is given by qs = P, where qs
is the quantity supplied.
a. What is the market equilibrium?
b. The government implements a tax of $30 per unit to be paid by consumers.
What is the new market equilibrium? What is the economic incidence of the
tax (that is, who pays for the tax)? How would your answer change if the
government implemented a production tax of $30 per unit instead?
c. With the aid of a diagram, calculate the deadweight loss that arises as a result
of the tax. Provide intuition for the result.
Chapter 17
1. The government issues tradeable pollution permits to deal with an externality. With
tradeable pollution permits:
a. The cost of reducing pollution is minimized, regardless of their initial
distribution among polluting firms.
b. The firms that are allocated permits will continue to emit more pollution than
firms with fewer, or zero, permits.
c. Although the price of permits reflects the opportunity cost of polluting, this
only applies to firms without permits.
d. The firms that are allocated permits will trade all their permits to other firms,
and keep the money.
e. The cost of reducing pollution is minimized, although firms receiving permits
will emit more pollution than other firms.
2. Consider a market for sport in which the private marginal benefit (PMB) is given
by PMB = 120 – 2q, where q is the quantity consumed. The marginal cost of
providing the service is given by MC = q. In this market there is also a positive
consumption externality of $30 per unit consumed.
a. Define externality.
224 21 Review
b. For this market, compare the market outcome with the socially efficient
outcome. What is the deadweight loss, if any? With the aid of a diagram,
provide some intuition for your answer.
c. In this market, the government is restricted in that it cannot by law subsidize
or tax consumers, but it can legally tax or subsidize producers. Can the
government intervene in the market to achieve the surplus-maximizing
outcome? Again, explain your answer.
3. What is the Coase theorem? Why might the Coase theorem fail?
4. Consider a market in which there is a demand curve of qd = 10 – P, where qd
is the quantity demanded and P is the price. The market supply curve is given by
qs = P, where qs is the quantity supplied. There is, however, a negative production
externality of $2 per unit produced. What is the deadweight loss (the loss of surplus)
in the market outcome?
a. $1.
b. $2.
c. $3.
d. $4.
e. It is not possible to ascertain with the information provided.
5. The private demand for education is given by qd = 500 – 5P. The supply of
education is given by qs = 5P.
a. What is the market equilibrium?
b. Education generates a positive externality of $5 a unit. What is the social
optimum? What is the deadweight loss?
c. Outline a possible intervention in the market by the government that could
improve the market outcome.
6. There has been some debate about the use of single-use plastic bags.
a. If there is a negative consumption externality compare the welfare effects of
a ban on plastic bags compared with the market outcome. Is there a level of
externality for which a zero level of output is efficient?
b. What is an alternative policy to a policy of banning plastic bags?
7. The cost of reducing each unit of pollution costs firm A are $10 for the first unit
of pollution abated (or reduced), $20 for the second unit abated, $30 for the third,
$40 for the fourth, $50 for the fifth, $60 for the sixth and $70 for the seventh unit
of pollution that is not emitted. For firm B the costs of reducing pollution are $5
for the first unit not emitted, $15 for the second unit abated, $25 for the third unit,
$35 for the fourth unit, $45 for the fifth unit, $55 for the sixth unit not emitted
and $65 for the seventh unit not emitted. The government imposes a tax of $37
for each unit of pollution emitted. What is the total level of abatement (or the total
reduction) in pollution after the tax has been implemented?
8. Consider an externality between a beekeeper (and her bees) and almond production
from an almond orchard. It costs the beekeeper $25 to maintain each beehive. If
the beekeeper places her hives near the almond orchard almond production
increases by $10 per hive. The benefit to the beekeeper from placing her hives
near the almond orchard is $50 for the first hive, $40 for the second hive, $30 for
the third hive, $20 for the fourth hive, $10 for the fifth hive and $0 for any
subsequent hives. If the almond orchard buys the beekeeping operation (that is,
the two activities are performed by the same firm) what is the number of hives
that will be placed near the almond orchard? Explain your answer in context of
the literature on market mechanisms to deal with externalities.
Chapter 18
1. Consider a public good that has a marginal benefit for consumer 1 of MB =
20 – q and a marginal benefit for consumer 2 of MB = 30 – 1.5q. If the marginal
cost of provision is $10 per unit, what is the socially optimal level of output?
2. A public goods is
a. Non-rival and excludable.
b. A private good that benefits many people.
c. Rival in consumption, but not excludable.
d. Non-rival and non-excludable.
3. The local council is considering providing access to a public good. Beth, Cathy
and David are the only potential users of the good. Their willingness to pay for
the good are as follows: PBeth = 10 – Q; PCathy = 30 – 3Q; and PDavid = 20 – 2Q,
where a total of Q units are provided. If there are no fixed costs, and the marginal
cost of providing the good is given by MC = 20 + 2Q, what is the optimal quantity?
Chapter 19
1. There is a negative externality associated with consuming an imported product of
$6 per unit. This imported good costs $12 on the world market. There is no negative
externality from consuming the locally produced good, which costs $16 to make.
What is the appropriate policy response from government?
a. Not to intervene in the market.
b. Subsidise local producers $4 a unit
c. Implement a tariff (a tax on imports) of $6 per unit.
d. Implement a consumption tax of $6 per unit.
2. Consider Donna, a monopolist, selling to a market with a demand curve P =
20 – q, where P is the market price and q is the quantity demanded. Donna has a
constant marginal cost of production of $2 per unit and a fixed cost of $20.
a. If Donna charges a linear price (or a single price), what is the profit-
maximizing price and quantity, Donna’s profit and any deadweight loss that
arises. Explain your answer with the help of a diagram.
226 21 Review
b. Now consider that there is a negative production externality of $10 per unit?
What is the deadweight loss generated by the market outcome? What is an
optimal policy response by the government? Explain your answer.
Chapter 20
1. Suppose that the federal government imposes a tariff on imported cars to protect
the Australian car industry from foreign competition.
a. Assuming that Australia is a price taker in the world car market, use a diagram
to help analyse the welfare effects of the policy.
b. Who is going to be opposed to the removal of the tariff? Who is going to be
in favour for the removal of the tariff? Why might reducing tariff be a
potentially difficult political issue?
2. What is an import quota. Using a diagram, analyse the impact of a binding import
quota. Show that a quota can produce the same effect on local production and
on the number of imports as a tariff set a given rate. What is the value of the import
licences?
3. Guyana, a small country, is considering opening up its aluminium market to
international trade – that is international trade is currently not permitted. The
government knows that the world price for aluminium is above the price that is
prevailing in the domestic market. Knowing little else, the government hires you
as a consultant. The government asks you the following questions.
a. If international trade is allowed, what will be the effect on domestic price,
domestic production and domestic consumption? Use a diagram to help
explain your answer.
b. Again using the assistance of a diagram, explain what happens to PS and CS
if international trade is allowed. What happens to total surplus?
c. Is the policy of not allowing international trade in aluminium pareto efficient?
Explain your answer.
4. Critically assess the arguments for protection of a domestic industry from
international trade.
5. Consider the market for bottled water in Australia (a small country). The quantity
demanded in Australia qd is given by qd = 90 – P, where P is the market price.
The domestic quantity supplied qs is given by *qs = 2P. The world price for bottled
water is $10.
a. With the aid of a diagram, calculate the change in consumer surplus, producer
surplus and overall welfare if Australia removes a ban on international trade
of bottled water. Is the original ban of international trade in bottled water
efficient? Give reasons for your answer.
b. After the Australian government has removed the ban on international trade,
it then implements an import tax of $10. Again, with the help of a diagram,
calculate the change in quantity demanded, quantity domestic quantity
produced, imports and exports and the change consumer surplus, producer
surplus and government revenue. Make sure you explain the economic
intuition behind your working and your answer.
c. The government, instead of the import tax, considers implementing a $10 per
unit production subsidy for domestic producers. What are the implications for
the market, in particular, the change in consumer, producer surplus and tax
revenue and the impact on overall welfare? Compare the outcome to that in
part b. Make sure you explain the economic reasoning behind your answer.
d. You are hired as an economic consultant by the Australian government. What
is your ranking of the three alternative policies outline in parts a, b, and c (that
is free trade, a tariff and a production subsidy)? Which policy would you
recommend and why?
Selected answers
Chapter 1
1. Opportunity cost of an activity is the next best forgone opportunity.
2. Opportunity cost is the next best forgone opportunity – this includes both explicit
costs and implicit costs (for example, what a person might alternatively do with
their time).
3. e. All of the activities listed involve tradeoffs.
4. d.
5. a.
Chapter 2
1. P = 100 – 0.5q
2. C′(q) = dC/dq = 10 + q
3. εd = –4
4. q = 30, P = 60
5. q* = 20 units
Chapter 3
1. d.
2. a.
3. c.
4. c.
228 21 Review
5. (PP, B) and (B, PP), where the first term in each brackets is Vlad’s strategy and
the second term is Guillaume’s strategy.
6. Japan chooses OUT, and Australia AGREE.
7. Japan chooses IN and Australia AGREE.
8. (On, Off) and (Off, On) where the first strategy in each parentheses is Martin’s
and the second is Jim’s. The first Nash equilibrium produces higher total surplus.
The two guitarist might know to play this equilibrium from experience practicing
and playing together in the past.
9. (Lead, Rhythm) and (Rhythm, Lead).
Chapter 4
2. b is not true.
3. a. The party that is more productive has the absolute advantage. Suzie has the
absolute advantage in x, no one has the absolute advantage in y.
b. Comparative advantage means having the lower opportunity cost. Bob has the
comparative advantage in y – his opportunity cost is two units of x compared
with Suzie’s opportunity cost of 3 units of x. Suzie has the opportunity cost
in producing x – her opportunity cost is 1⁄3 unit of y compared with Bob’s
opportunity cost of 1⁄2 unit of y.
c. The slope of the PPF represents the opportunity cost for each individual
changing production from one good to the other.
d. Each individual will only be willing to trade if they can either buy a good
for a lower price than their opportunity cost, or if they are selling a good, they
must receive a price that is at least as large as their opportunitycost. The
maximum price Bob is willing to pay for x is 1⁄2 a unit of y. The minimum
price Suzie will accept for selling a unit of x is 1⁄3 a unit of y. The minimum
price that Bob would sell y for is 2 units of x. The maximum price that Suzie
would pay for a unit of good y is 3 units of x.
Together, these minima and maxima give the range of price in between
which trade is mutually beneficial.
4. In this case, Bob still has a comparative advantage in task y. Suzie has a
comparative advantage in task x. To maximize output of the firm Bob should
specialize in y and Suzie in task x. (Examples of this abound from real firms. For
instance, a lawyer might be both better at providing legal advice and doing the
filing than her assistant, but the total output of the firm might be greater if she lets
the assistant do the filing while she concentrates on providing legal advice to
clients.)
In this way, the idea that it is someone’s relative opportunity cost (their
comparative advantage) that determines what they should specialize in, and not
their absolute advantage, also applies to the allocation of tasks within a firm, as
well as to trading patterns we observe in the market.
There are several things different when we are considering internal organization
of production, as opposed to trade in markets. Here are a few of the important
differences. First, task allocation in a firm is chosen by the manager – they are not
determined by voluntary trading choices of individuals in the market. That is, there
is nothing automatic in regards to the decision as to task allocation in a firm – it
is up to a clever manager to see her workers opportunity costs and allocate jobs
accordingly. Second, we are abstracting in this argument about the balance between
tasks required and the output mix that a firm would want to produce (that is, which
goods they want to sell on the market). But to the extent that profits will be
increased by a firm that is able to produce more output with the same (or less)
inputs, comparative advantage is a key driver in how a successful firm will allocate
tasks among its employees.
5. Both increase in population and technological progress shift the PPF to the right.
The advantage of technological progress is that it leads to an increase in the per
worker (or per person) consumption.
Imagine an economy that produced only coffee. Increasing the number of
workers who can spend time producing coffee would increase coffee production.
This would not necessarily increase the consumption of coffee per person. (In
principle, it could if the new workers were relatively more productive than the old
workers but let’s assume that this is not the case.) Alternatively, if the technological
progress occurred and machines became more efficient, then the economy could
produce more coffee with the same number of workers. This would lead to an
increase in per person consumption of coffee.
This example illustrates the idea that living standards are determined by
productivity.
Chapter 5
1. a. To work out the outcome of these negotiations we work backwards. By
solving backwards we capture the fact that these bargaining agents are forward
looking; they will all take into account possible future actions or events when
thinking about what to do (in the present).
b. Maggie’s payoff if she rejects the offer is 0, so she will accept any offer greater
or equal to 0. Knowing this (working backwards) Davo offers Maggie a price
for the Corolla of 0 (or an offer very close to zero). So at this stage of the
negotiations (if reached), Davo offers a price $0 and Maggie accepts that offer.
c. Davo in a sense has an outside option – he can always reject Shazza’s offer,
start bargaining with Maggie and end up with a payoff of $4,000. As a result,
Davo will not accept any offer that gives him less surplus than $4000. In other
words 20,000 minus the price must be ⭓ 4,000. Hence, the maximum price
Davo is willing to accept will be $16,000; that is Davo will accept any offer
from Shazza of 16,000 or less. Shazza knows this, so she offers the highest
price that Davo will accept; Shazza offers $16,000. In summary, Shazza
offers a price $16,000 and Davo accepts this offer.
230 21 Review
d. In this bargaining outcome Shazza gets $16,000 surplus, Davo gets $4,000
surplus and Maggie gets $0. Thus total surplus is $20,000. The maximum
possible surplus in this situation is $20,000 (when Davo ends up with the
Falcon). As Davo does buy the Falcon here, total surplus is maximize.
2. b. Singapore offers a price of $100 and Ansett accepts.
b. Ansett offers a price of $100, and Singapore accepts; Singapore gets a net
payoff of 350 – 100 = $250 million, and Ansett $100 million.
3. Bazza offers a price of $100 and Raelene accepts. Trade takes place immediately
in the first year. Note here that there is a cost of delay. Bazza knows that if Raelene
rejects the offer the potential gains from the patent in year 1 will be forgone. This
gives Bazza some bargaining power – in fact, he captures the potential gains from
using the patent in year 1.
4. Solving backward. At the very beginning of the game, A offers a price of $25 and
B accepts this offer.
Chapter 6
1. Due to the consumer’s diminishing marginal benefit for the good or service.
2. b.
3. MB = 10 – q is the consumer’s demand curve, although it is usually written as
P = 10 – q. The maximum price for the first unit (for a very small unit) is 10 –
the intercept of the P axis. The maximum price a consumer is willing to pay for
the 7th unit is 3.
4. e.
5. P = 20 – 1⁄3Q.
6. P = 10 – 2⁄5Q.
Chapter 7
1. e.
2. b.
3. d.
4. d.
6. MC = 10 + 4q, AVC = 10 + 2q and ATC = 250/q + 10 + 2q.
Chapter 8
1. True. Notwithstanding a few nuances regarding the short and long run, if a firm
is going to produce a positive level of output its marginal cost curve is its supply
curve.
2. A firm’s supply curve is its MC curve. It is upwards sloping in the short run due
to its diminishing marginal product.
3. d.
Chapter 9
1. b.
2. c.
3. a.
4. c.
5. a. q* = 25, P* = 100 – 25 = 75.
b. Consumer surplus: MB minus price paid for each unit consumed. It
measures the net benefit a consumer receives from consuming a good, as they
perceive it.
Producer surplus: is the difference between the price received by a firm
for selling a good and the minimum price they would have accepted to
sell the good, for every good sold. It is the difference between the price
received and the marginal cost of producing a good for every good sold:
CS = (100 – 75)(25)/2 = $312.5; and PS = (25)(75)/2 = $937.5.
The competitive market outcome maximizes surplus as all trades for which
MB ⭓ MC take place. In a competitive market, consumers keep buying
provided MB ⭓ P, and firms keep selling as long as P ⭓ MC; thus all mutually
beneficial trades take place (when MB ⭓ MC) and all the gains from trade are
realized.
6. CS = $24.5; PS = $49.
Chapter 10
1. e.
2. a.
3. b.
4. c.
5. –1.
6. c.
Chapter 12
1. e.
2. b.
3. d.
232 21 Review
4. e.
6. e.
7. b.
Chapter 13
1. qm = 20 units, PM = 60, ␲ = 800.
2. ␲ = $150.
3. d.
4. c.
5. b.
6. e.
7. a. MR = 100 – 2q; Profit maximization when MR = MC. qM = 20
and PM = 100 – 20 = 80. ␲ = $1000.
For a monopolist charging a single price must drop its price to all infra-
marginal consumers in order to sell an extra unit. Hence, MR < P (for all units
but the very first unit). Like all firms will maximize profit continuing to sell
when MR ⭓ MC, so it maximized profit here when MR = MC.
b. Surplus is maximized when all trades for which MB ⭓ MC. Here this means
that 25 units are traded. This is the same as the competitive market outcome
(as all trades for which MB ⭓ MC take place with that market structure).
For a monopolist, profit is maximized when MR = MC – here at 20 units.
As the monopolists MR < P, this quantity will be less than the quantity required
so that MB = MC for the last unit traded. That is, in this case between an output
of 20 and 25 units, the MB of a good is greater than its MC of production –
there are potential gains from trade (surplus could increase), but these potential
gains are forgone due to the monopolist restricting output. A DWL (or loss of
potential surplus) of $50 results in this case. The DWL = $40.5.
8. a. MR = 20 – 2q; PM = 11, qm = 9, ␲ = 61, DWL = 40.5.
b. Fixed fee F = 162, per unit price p = 2; ␲ = 142.
9. b.
10. d.
Chapter 14
1. d.
2. a.
4. ATC is not minimized in the long run in a monopolistically competitive firm.
In the LR the firm equates MR and MC, and makes zero profit. This requires
P = ATC. As P > MC, ATC is not minimized.
However, as the firms in a monopolistically competitive industry are not selling

identical products (as they are in perfect comp), each new firm offers a new product
to consumers. As consumers appreciate this variety, entry by new firms can
increase surplus.
As a consequence, it is not immediately obvious that just because ATC is not
minimized that entry is excessive.
Note a firm will enter a market if its (variable) profit covers its fixed costs. In
terms of society’s welfare, the increase in fixed costs from entry must be
outweighed by the increase in surplus (both to the entering firm and to consumers).
These two objectives do not necessarily coincide.
In comparison to total welfare, there can be excessive or too little entry into a
market that is monopolistically competitive. Moreover, there are two effects that
influence total welfare and the number of firms in the market. First, the product
variety effect suggests that there will be too little entry. Because a firm increases
product variety that increases consumer surplus; to the extent that this increase in
surplus is not captured by the entry firm, there is too little incentive to enter.
Second, an entering firm steals business off existing firms, providing it an incentive
to enter but not actually increasing overall surplus; this effect suggests excessive
entry.
Hence, one effect suggests too little entry (or number of firms in the market);
the other suggests excessive entry. Depending on which of these effects dominate
will determine whether there are too many or too few firms in the market with
respect to maximizing total welfare.
Chapter 15
1. d.
2. a.
4. c.
5. The NE are Platform 1, Platform 1 and Platform 2, Platform 2.
This is a coordination game – the parties like to coordinate their actions. More
generally, you can think of firms wanting to do the same thing (as in this example,
where they want compatible platforms) or wanting to do different (or opposite)
things, for example you might want to locate your restaurant where your rival is
not (or Sony might want to make their system incompatible with Nintendos). Note
that even though both firms prefer the equilibrium Platform 2, Platform 2, there
is no unilateral incentive to deviate out of the Platform 1, Platform 1 if that’s were
the firms find themselves.
7. a. (Adv, Adv).
b. The NE is – (Adv, Adv), (Adv, Adv) and (Adv, Adv) – play Adv in every period.
234 21 Review
Chapter 16
1. d.
2. e.
3. a.
4. $6.
5. 50.
6. c.
7. a. P* = $50, q* = 50.
b. Producers receive $30 per unit; Consumers pay $60 per unit; quantity with
the tax is 30 units.
Incidence of the tax: Consumers pay for $10; producers pay $20 per unit
for the tax (note, in total they pay for the entire tax $30 per unit)
If the tax was implemented on the production side the economic incidence
of the tax would be identical (as would the quantity bought and sold)
The intuition for this is that the burden of the tax depends on the relative
elasticities of supply and demand (the side of the market that is relatively more
inelastic paying more of the tax), not who legally is required to pay the tax
to the government. Thus, the economic incidence of the tax is the same as
with a consumption tax. Further, the gap between the consumer and producer
price is required to be equal to the size of the tax – this means that the quantity
sold in the market needs to be reduced just enough in order that the gap
between S and D is equal to the tax. It does not matter, therefore, who legally
has to pay for the tax as the market output ends up being the same level
regardless as to whether it is a consumption or a production tax.
c. The tax effectively shifts the demand curve vertically down by the size of the
tax. This is because consumers must not reduce the tax they must pay to the
government from their MB they received from each unit they consume – it is
as if their MB was reduced by the size of the tax for each unit they consume.
The new equilibrium quantity is where the new demand curve with the tax
intersects the supply curve – in this case qt = 30 units. The DWL = $300.
A DWL arises because the tax places a wedge between the MB of the last
unit consumed and the MC of the last unit produced. Because the consumer
has to cover the tax, they need to receive at least enough MB from any good
they buy to cover the price plus the tax. That means that the last good bought
has the characteristic that MB – T = MC. Beyond this level of consumption,
the consumer does not receive enough benefit from purchasing a unit to cover
the price plus the tax (the full cost of buying a unit).
This wedge between MB and MC means that not all the gains from trade
are realized. Between qt and q*, each extra unit would provide a greater benefit
than it costs to produce – there are unrealized potentially mutually beneficial
trades that are not made. The DWL is the total loss in surplus from these
unrealized gains from trade as a result of the tax.
Chapter 17
1. a.
4. a.
5. a. P = 50, q = 250 units.
b. q* = 262.5 units; DWL = 31.25.
c. The government could implement a subsidy for education of $5 a unit. This
means the private benefit from education equals the benefit from the education
itself, plus the $5 subsidy from the government. In this case, the private
incentives to get education are MB = 105 – qd /5, which aligns society’s and
the individual’s incentives to get education.
6. a. With a negative externality e per unit, the MSC curve is shifted down from
the MPB curve by e per unit. The efficient quantity is q* below qm
b. In order for zero level of output to be the efficient quantity, the externality
needs to be sufficiently large so that the MSB curve and the MSC curve never
intersect. That is, the negative externality e must be at least as large as the
gap between the MPB and MSC at zero level of output.
7. Firm A reduces emission when the cost of abatement is less than the tax, so Firm
A reduces emissions by 3 units.
Firm B reduces emissions by 4 units.
Total reduction is 7 units.
The general principal is that a firm will try to minimize the negative impact –
they will reduce emissions when that is less costly than the tax. When the loss to
the firm from reducing emissions is greater than the tax, a firm will choose to pay
the tax (and continue to emit that unit of pollution).
8. 4; participants to trade have an incentive to maximize surplus (if they can) because
with larger surplus at least one person can be made better off without making
anyone worse off (or everyone can be made better off) – refer to Coase theorem.
Chapter 18
1. 16 units.
2. d.
3. Q* = 5.
Chapter 19
1. c.
2. a. PM = 11, qM = 9 units; DWL = $40.5.
b. The efficient quantity is where MB = MSC = 12; the efficient quantity is
q* = 8. Hence the monopoly output is still too high, so a tax of $2 per unit
will induce the efficient level of output, because MR = MC + t, or 20 – 2(8)
236 21 Review
= 2 + t, hence t = 2. Note, that the monopolist reduces the level of output

compared with a competitive industry, so a smaller tax is required. If this were
a competitive industry (with price-taking behaviour) the tax would need to be
$10 per unit.
Chapter 20
3. a. Prior to international trade, equilibrium is determined by the intersection of
domestic supply and demand. After international trade is allowed, domestic
price immediately rises to be equal to the world price Pw . At this higher price,
domestic quantity supplied increases, and domestic quantity demanded
decreases. The difference between domestic quantity supplied and domestic
quantity demand is exported.
b. Trade has made producers better off and consumers worse off. Part of this is
a transfer from consumers to producers, but trade has increased total surplus
(the gains to the winners exceed the loses to the losers).
c. Pareto Efficiency – an outcome is PE if it is not possible to make one party
better off without making any other party worse off.
The outcome without trade is not PE. Trade increases total surplus, so it
makes it possible for the winners from allowing the change to adequately
compensate the losers for any losses they may incur. As international trade
increases total surplus, it is possible to compensate for producers to fully
compensate producers, plus have some extra surplus remaining.
5. a. No international trade: Price $30, q = 60.
At world price (with trade): P = 10, qd = 80, qs = 20.
CS increases by $1400; PS falls by 800; overall surplus increases by $600
with international trade. The original ban is not efficient.
b. With tariff domestic price is Pt = 20; qdt = 70, qst = 40. Imports are 30 units
CS falls by 750; PS increases by 300, Government revenue is 300; total welfare
falls with the tariff by 150.
c. With a production subsidy, the domestic price remains at the world price.
But now effectively domestic producers receive 20 for each unit sold.
Hence qds = 80, qss = 40.
CS is unchanged from free trade. PS increases by 300. Government
expenditure (a minus) is 400. So total welfare falls by 100.
With a production subsidy there is no consumption DWL, there is just the
production DWL. So a production subsidy results in higher welfare than the
tariff (but less than free trade).
d. If maximizing total surplus is the objective, free trade is the best policy,
followed by the subsidy, the tariff, then the ban on international trade.
Remember, with more total surplus it is possible to fully compensate the losers
from a policy change and still make others better off.
Index
absolute advantage 32–3 consumption 31; externalities 164, 166–9;

ad valorum tax 151 subsidies 159–60; tax 152; see also
advertising 131 demand
allocation 27, 149–50 coordination failure 135–6
anti-dumping measures 198 coordination game 19–20, 134–6
arbitrage 112–13, 115–16, 118 correlation 6
arc method 80–1 cost function 57
autarky 189–92 costs: fixed 57–8, 118–19; long run 59–61;
average costs 58–9, 98–101; long-run 60; opportunity 4–5, 32–3; short run 57–9;
price regulation 121 total 62; transaction 171–2
credible threats 136–8
backwards induction 22–3 cross-price elasticity 86–7
bargaining 37–42, 170–2 curves 9; elasticity 83; individual demand
barriers 194–8 48–50; Laffer 159; marginal benefit 48;
barter 28 market demand 50–1; supply 65–8
battle of the sexes 19–20
benefit 47–8, 164 deadweight loss see DWL
best response 16 decreasing-cost industry 102–3
binding 148–9; commitment 132 demand 47–51, 71–3; elasticity of 81–5;
breakdown 42 excess 70; international trade 190–3
business stealing 127 differentiation 9–10, 55
diminishing marginal benefit 48
capital 53–4 discrimination 150
causation 6 diseconomies of scale 60
ceteris paribus 5–6 dominant strategy 15–16, 18
change see elasticity DWL 111–12, 119–21, 155, 157–9, 162, 167,
change in demand 50 195–6
change in quantity demanded 49–50
change in supply 66–7 economic incidence 155–6, 160–1
Coase Theorum 170–2 economic profit 62
commitment 138–9 economics 3–6
common resources 182 economies of scale 60
comparative advantage 32–3 elasticity 10, 79–81, 88; cross-price 86–7;
comparative static analysis 71–3 of demand 81–5; income 87; of supply
competition, monopolistic 91–2, 123–8 85–6
competitive markets 65, 78, 93, 104; emissions 174–6
long run 96–103; short run 94–6; employment protection 199
see also equilibrium; welfare entry condition 97–8, 101–2, 125–6
constant average costs 60–1 environmental standards 198–9
constant-cost industry 100–2 equations 7–9
consumer surplus 73, 76, 110–11, 150–1, equilibrium 69–73, 77; international trade
191–3 190–3; Nash 16–18, 21–3, 134–7, 141–2;
238 Index
perfect competition 96–104; see also long run 53–4, 56; costs 59–61; monopolistic
externalities competition 125–7; perfect competition
excess demand 70, 149 96–103; price controls 151
excess supply 69, 148 losses: monopolistic competition 126–8;
exchange 27–8; see also international trade; perfect competition 95–6, 98–100;
trade see also DWL
exit condition 97–8, 101, 125–6
exporting 190–2 machines see capital
extensive form 20–1, 23 marginal: analysis 5; benefit 5, 48–9, 180; cost
externalities 163–5, 176–7; government 5, 58–60, 65–7, 120–1; product 55–6;
solutions 172–6; problem of 165–9; revenue 106–7; social benefit 164, 166,
solutions 169–72 168–9; social cost 165–6, 168–9
market 3–4, 91–2; clearing price 69, 71;
firm 53–4, 63; costs 57–61; perfect competition comparative static analysis 71–3; demand
94–8, 100–4; producer surplus 74–6; 50–1; equilibrium 69–73, 77; power 112;
production 54–7; profit 62; supply 65–7; supply 67–8, 95, 100–3; see also monopoly;
see also monopoly; oligopoly; perfect oligopoly; perfect competition
competition market failures 163; common resources 182;
first-degree price discrimination 113–15 price regulation 147–51; public goods
first-mover advantage 139–51 180–2; Theory of Second Best 185–6; see
fixed costs 57–8, 118–19 also externalities
free market see competitive markets mathematics 7–11
free ride 142 maxima 10
free trade 198–9 midpoint method 80–1
minima 10
gains 77; bargaining 37–42; trade 27–35 minimal product differentiation 135
game theory 13–14, 129–30, 134–6; money 28
bargaining 37–42; sequential 20–4, 136–43; monopoly 91–2, 105–6, 122; competition 91–2,
simultaneous-move 14–20, 130–4 123–8; natural 118–22; price discrimination
government 181–2, 186; externalities 172–6; 112–18; single-price 106–12
international trade 194–8; ownership 120; MP 55–6
price regulation 147–51; subsidies 159–62, multiple-offer 40–1
172, 174–5; taxes 151–9, 173–5
Nash equilibrium 16–18, 21–3, 134–7, 141–2
hidden information 121–2 natural monopoly 118–22
negative externality 163–5, 167–9, 173–4
importing 192–8 non-binding 148–9
incidence: of subsidy 160–1; of tax 155–7 non-excludable 179
income elasticity 87 non-rivalrous 179
increasing-cost industry 102 normal form 14–15
individual demand 48–50
infant industries 198 offers 38–41
information 112–13, 115, 118, 121 oligopoly 91–2, 129–30, 143; product choice
input 53–7, 59–63 134–6; sequential games 136–43;
international trade 189, 199; barriers to simultaneous move games 130–4
194–8; welfare 189–93 opportunity cost 4–5, 32–3
output 30–1, 34–5, 53–63; see also supply
labour 31, 53–6, 199
Laffer curve 159 Pareto efficiency 77, 191, 193
law of demand 49 per-unit tax 151
law of supply 66–7 perfect competition 91–3, 104; long run
legal incidence of tax 155–6 96–103; short-run 94–6
Index 239
perfect price discrimination 113–15 simultaneous equations 10–11

permits 175–6 simultaneous-move games 14–20, 130–4
point method 80 single-price monopolist 106–12
pollution 174–6 small country 189–93
population 30–1 SPE 22–4, 38–9
positive externality 163–4, 166–9, 172 specialization 28–34
PPF 28–32 stag hunt 20
pre-game communication 132 standard of living see consumption
prevent arbitrage 112–13, 115–16, 118 straight lines 7–9
price 27–8, 47–8, 73–8; ceiling 149–51; strategic interaction see oligopoly
discrimination 112–18; floor 147–9; strategic trade policy 198
maker 124; regulation 120–1, 147–51; strategy see game theory
takers 65, 68; wars 130–1; see also subgame perfect equilibrium 22–4, 38–9
elasticity; equilibrium subsidies 159–62, 172, 174–5
prisoner’s dilemma 18–19, 130–4 sunk costs 4–5
private good 179–80 supply 65–8; elasticity of 85–6; excess 69;
producer surplus 74–6, 110–11, 151 long run 96–103; short run 94–6
product: choice 134–6; differentiation 124–5, surplus 37–42, 73–6, 191–2
134–5; variety 127–8
production 54–7, 100; externalities 164–5, take-it-or-leave-it 38–40
167–9; function 54–5; possibility frontier tariffs 194–6
28–32; subsidies 160; tax 152–3 taxes 151–9, 173–5
profit 17–18, 62; maximization 107–9, 124–6; technology 29–31
monopolistic competition 126–8; perfect Theory of Second Best 185–6
competition 95–6, 98–100 third-degree price discrimination 115–16
property rights 170–1, 182 tit-for-tat 133–4
public goods 179–81 tools see game theory
total benefit 47–8
quantity regulation 174 total costs 62
queuing 149 total revenue 62, 84–5, 111–12
quotas 196–8 total surplus 76
trade 27–35; bargaining 37–42; see also
rationality 5 international trade
Reagan, R. 159 tradeable permits 175–6
regulation see government Tragedy of the Commons 182
repeated game 132 transaction costs 171–2
resources 3, 28–32, 53–4, 77 trigger strategy 133
returns to scale 56 two-part tariff 113–14
revenue: marginal 106–7; total 62, 84–5,
111–12 value 27–8, 37–9
variable costs 58
scarcity 4–5 Viking game 138–9
second-degree price discrimination 117–18
second-mover advantage 141–3 welfare 73–7; international trade 189–93,
self-interest 47 195–7; monopolies 110–12, 120, 127–8;
selling see supply price regulation 148–51; subsidies 161–2;
sequential games 20–4, 136–43 tax 154–5
short run 53–4, 56; costs 57–9; monopolistic willingness to pay 47–8, 117
competition 124–5; perfect competition world market see international trade
94–6 WTP 47–8, 117
shut-down 94
side payments 150 zero profits 99

Essentials of Microeconomics

Uploaded by

Copyright:

Available Formats

Essentials of Microeconomics

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Essentials of Microeconomics

Uploaded by

Copyright:

Available Formats

Essentials of Microeconomics

‘Essentials of Microeconomics, by Bonnie Nguyen and Andrew Wait, prepares students

Essentials of Microeconomics is an excellent introduction to microeconomics. It presents

Bonnie Nguyen and Andrew Wait

Economics Cover image: © Thinkstock

Essentials of Microeconomics is an excellent introduction to microeconomics. It

Andrew Wait is an Associate Professor in the School of Economics at the University

‘Essentials of Microeconomics, by Bonnie Nguyen and Andrew Wait, prepares students

‘Nguyen and Wait’s Essentials of Microeconomics is really covering the essentials of

‘The authors have done a great job explaining complex materials.’

Bonnie Nguyen and

First published 2016

ISBN: 978-1-138-89135-7 (hbk)

Typeset in Times and Helvetica

List of illustrations vii

Part I Key concepts and tools 1

1 Key economic concepts 3

2 Key mathematical tools 7

3 Key strategic tools 13

Part II Gains from trade 25

4 Trade and the PPF 27

Part III Market fundamentals 45

7 Production and costs 53

9 Equilibrium and welfare 69

Part IV Types of markets 89

14 Monopolistic competition 123

Part V Market failures 145

16 Price regulation, taxes and subsidies 147

18 Public goods and common resources 179

19 The Theory of Second Best 185

Part VI International trade 187

20 International trade 189

Part VII Review 201

21 Questions and answers 203

3.1 The normal form of a game 15

13.12 Under average-cost price regulation, the government sets the

16.17 Government revenue and deadweight loss as a result of a subsidy 162

11.1 Types of market structure 92

Page Intentionally Left Blank

Page Intentionally Left Blank

Economics is the study of choice under scarcity. Typically consumers want

4 1 Key concepts and tools

answer to these questions by regulating or intervening in the market. Consequently,

1.2 Scarcity and opportunity cost

Key economic concepts 5

1.3 Marginal analysis

1.4 Ceteris paribus

The notion of ceteris paribus is also an important foundation of economic analysis.

6 1 Key concepts and tools

1.5 Correlation and causation

1.6 Concluding comments

Even at an introductory level, the analytical nature of economics requires us

Often we are interested in the relationship between two variables, x and y.

2.2.1 Straight lines

8 2 Key concepts and tools

Example. Suppose we are interested in the relationship between the price of a

2.2.2 The slope of a straight line

2.2.3 Determining the equation of a straight line

Key mathematical tools 9

of a straight line is y = mx + c. Substituting in m = –2 and (4, 3) yields 3 = –2 ×