Write an annotated bibliography of two theories and two conceptual framework models. Use your knowledge from Week 4 about writing an annotated bibliography. This week, you will keep the sources to...

Economics for Managers 3/E
G
lo
b
a
l
ed
it
io
n
econom
ics for M
anagers
Farnham
t
h
ir
d
ed
it
io
n
economics for Managers
third edition
Paul G. Farnham
this is a special edition of an established title widely
used by colleges and universities throughout the world.
Pearson published this exclusive edition for the benefit
of students outside the United States and Canada. if you
purchased this book within the United States or Canada
you should be aware that it has been imported without
the approval of the Publisher or author.
Pearson Global Edition
Global
edition
For these Global editions, the editorial team at Pearson has collaborated
with educators across the world to address a wide range of subjects and
requirements, equipping students with the best possible learning tools.
this Global edition preserves the cutting-edge approach and pedagogy of
the original, but also features alterations, customization and adaptation
from the north american version.
Global
edition
FARNHAM_1292060093_mech.indd 1 07/08/14 6:42 pm
Economics
for Managers
Third
Edition
Paul G. Farnham
Georgia State University

Boston Columbus Indianapolis New York San Francisco Upper Saddle River
Amsterdam Cape Town Dubai London Madrid Milan Munich Paris Montréal Toronto
Delhi Mexico City São Paulo Sydney Hong Kong Seoul Singapore Taipei Tokyo
Global Edition
A01_FARN0095_03_GE_FM.INDD 1 21/08/14 1:52 PM
Editor in Chief: Donna Battista
AVP/Executive Editor: David Alexander
Head of Learning Asset Acquisition,
Global Editions: Laura Dent
Senior Editorial Project Manager:
Lindsey Sloan
Director of Marketing: Maggie Moylan
Executive Marketing Manager: Lori Deshazo
Senior Marketing Assistant:
Kimberly Lovato
Managing Editor: Jeffrey Holcomb
Assistant Acquisitions Editor, Global
Editions: Debapriya Mukherjee
Senior Project Editor, Global Editions:
Vaijyanti
Art Director: Jayne Conte
Cover Designer: Lumina Datamatics Ltd.
Cover Art: ©Zoran Orcik/123rf
Media Director: Lisa Rinaldi
Media Production Manager, Global
Editions: M Vikram Kumar
Production Manager: Meghan DeMaio
Senior Manufacturing Production
Controller, Global Editions:
Trudy Kimber
Full-Service Project Management: Integra
Credits and acknowledgments borrowed from other sources and reproduced, with permission,
in this textbook appear on the appropriate page within text.
Microsoft® and Windows® are registered trademarks of the Microsoft Corporation in the U.S.A.
and other countries. Screen shots and icons reprinted with permission from the Microsoft
Corporation. This book is not sponsored or endorsed by or affiliated with the Microsoft
Corporation.
Pearson Education Limited
Edinburgh Gate Harlow
Essex CM20 2JE
England
and Associated Companies throughout the world
Visit us on the World Wide Web at:
www.pearsonglobaleditions.com
© Pearson Education Limited 2014
The rights of Paul G. Farnham to be identified as the author of this work have been asserted by
him in accordance with the Copyright, Designs and Patents Act 1988.
Authorized adaptation from the United States edition, entitled Economics for Managers, 3rd
edition, ISBN 978-0-132-077370-6, by Paul G. Farnham, published by Pearson Education © 2014.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system,
or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or
otherwise, without either the prior written permission of the publisher or a license permitting
restricted copying in the United Kingdom issued by the Copyright Licensing Agency Ltd, Saffron
House, 6–10 Kirby Street, London EC1N 8TS.
All trademarks used herein are the property of their respective owners. The use of any
trademark in this text does not vest in the author or publisher any trademark ownership
rights in such trademarks, nor does the use of such trademarks imply any affiliation with or
endorsement of this book by such owners.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
10 9 8 7 6 5 4 3 2 1
15 14 13 12 11
ISBN 10: 1-292-06009-3
ISBN 13: 978-1-292-06009-5
Typeset in ITC Century Std by Integra
Printed by Courier Kendallville in the United States of America
A01_FARN0095_03_GE_FM.INDD 2 21/08/14 1:52 PM
Dedication
To my friend and colleague, Dr. Jon Mansfield, who
continues to excel at teaching economics for managers.
A01_FARN0095_03_GE_FM.INDD 3 21/08/14 1:52 PM
A01_FARN0095_03_GE_FM.INDD 4 21/08/14 1:52 PM
Part 1 MicroeconoMic analySiS 32
1 Managers and economics 32
2 Demand, Supply, and equilibrium Prices 46
3 Demand elasticities 76
4 techniques for Understanding consumer Demand and Behavior 116
5 Production and cost analysis in the Short run 144
6 Production and cost analysis in the long run 172
7 Market Structure: Perfect competition 200
8 Market Structure: Monopoly and Monopolistic competition 226
9 Market Structure: oligopoly 260
10 Pricing Strategies for the Firm 288
Part 2 MacroeconoMic analySiS 320
11 Measuring Macroeconomic activity 320
12 Spending by individuals, Firms, and Governments on real Goods and Services 350
13 the role of Money in the Macro economy 390
14 the aggregate Model of the Macro economy 416
15 international and Balance of Payments issues in the Macro economy 446
Part 3 inteGration oF the FraMeworkS 482
16 combining Micro and Macro analysis for Managerial Decision Making 482
SolUtionS to even-nUMBereD ProBleMS 501
GloSSary 521
inDex 535
Brief Contents
A01_FARN0095_03_GE_FM.INDD 5 21/08/14 1:52 PM
A01_FARN0095_03_GE_FM.INDD 6 21/08/14 1:52 PM
Preface 17
About the Author 29
Part 1 MicroeconoMic analySiS 32
chapter 1 ManaGerS anD econoMicS 32
caSe For analySiS: Micro- and Macroeconomic influences on the Global
automobile industry 33
Two Perspectives: Microeconomics and Macroeconomics 35
Microeconomic Influences on Managers 36
Markets 36
Managerial rule of thumb: Microeconomic influences on Managers 39
Macroeconomic Influences on Managers 39
Factors Affecting Macro Spending Behavior 41
Managerial rule of thumb: Macroeconomic influences on Managers 43
End of Chapter Resources
Summary 43 • key terms 44 • exercises 44 • application Questions 44
chapter 2 DeManD, SUPPly, anD eQUiliBriUM PriceS 46
caSe For analySiS: Demand and Supply in the copper industry 47
Demand 48
Nonprice Factors Influencing Demand 49
Demand Function 53
Demand Curves 54
Change in Quantity Demanded and Change in Demand 55
Individual Versus Market Demand Curves 56
Linear Demand Functions and Curves 56
Mathematical Example of a Demand Function 57
Managerial rule of thumb: Demand considerations 58
Supply 58
Nonprice Factors Influencing Supply 58
Supply Function 60
Supply Curves 61
Change in Quantity Supplied and Change in Supply 61
Mathematical Example of a Supply Function 62
Summary of Demand and Supply Factors 63
Managerial rule of thumb: Supply considerations 64
Contents
A01_FARN0095_03_GE_FM.INDD 7 21/08/14 1:52 PM
8 contents
Demand, Supply, and Equilibrium 64
Definition of Equilibrium Price and Equilibrium Quantity 64
Lower-Than-Equilibrium Prices 64
Higher-Than-Equilibrium Prices 66
Mathematical Example of Equilibrium 67
Changes in Equilibrium Prices and Quantities 67
Mathematical Example of an Equilibrium Change 70
End of Chapter Resources
Summary 72 • key terms 72 • exercises 72 • application Questions 74
chapter 3 DeManD elaSticitieS 76
caSe For analySiS: Demand elasticity and Procter & Gamble’s Pricing Strategies 77
Demand Elasticity 78
Price Elasticity of Demand 79
The Influence of Price Elasticity on Managerial Decision Making 80
Price Elasticity Values 81
Elasticity and Total Revenue 81
Managerial rule of thumb: estimating Price elasticity 83
Determinants of Price Elasticity of Demand 83
Number of Substitute Goods 84
Percent of Consumer’s Income Spent on the Product 84
Time Period 85
Numerical Example of Elasticity, Prices, and Revenues 85
Calculating Price Elasticities 85
Numerical Example 87
The Demand Function 87
Other Functions Related to Demand 87
Calculation of Arc and Point Price Elasticities 88
Price Elasticity Versus Slope of the Demand Curve 89
Demand Elasticity, Marginal Revenue, and Total Revenue 90
Vertical and Horizontal Demand Curves 92
Vertical Demand Curves 92
Horizontal Demand Curves 93
Income and Cross-Price Elasticities of Demand 94
Income Elasticity of Demand 94
Managerial rule of thumb: calculating income elasticity 95
Cross-Price Elasticity of Demand 95
Elasticity Estimates: Economics Literature 97
Elasticity and Chicken and Agricultural/Food Products 98
Elasticity and Beer 99
Water Demand 100
Elasticity and the Tobacco Industry 100
Elasticity and Health Care 101
Tuition Elasticity in Higher Education 101
Managerial rule of thumb: Price elasticity Decision Making 102
Elasticity Issues: Marketing Literature 102
Marketing Study I: Tellis (1988) 103
Marketing Study II: Sethuraman and Tellis (1991) 104
Marketing Study III: Hoch et al. (1995) 105
Marketing Study Update 105
A01_FARN0095_03_GE_FM.INDD 8 21/08/14 1:52 PM
contents 9
Managerial rule of thumb: elasticities in Marketing and Decision Making 106
End of Chapter Resources
Summary 106 • appendix 3a economic Model of consumer choice 107
• key terms 113 • exercises 113 • application Questions 114
chapter 4 techniQUeS For UnDerStanDinG conSUMer DeManD anD Behavior 116
caSe For analySiS: the Use of new technology to Understand
and impact consumer Behavior 117
Understanding Consumer Demand and Behavior: Marketing Approaches 118
Expert Opinion 118
Consumer Surveys 119
Test Marketing and Price Experiments 120
Analysis of Census and Other Historical Data 121
Unconventional Methods 121
Evaluating the Methods 122
Managerial rule of thumb: Marketing Methods for analyzing consumer Behavior 123
Consumer Demand and Behavior: Economic Approaches 123
Relationship Between One Dependent and One Independent Variable: Simple Regression
Analysis 124
Relationship Between One Dependent and Multiple Independent Variables:
Multiple Regression Analysis 129
Other Functional Forms 131
Demand Estimation Issues 132
Managerial rule of thumb: Using Multiple regression analysis 133
Case Study of Statistical Estimation of Automobile Demand 133
Managerial rule of thumb: Using empirical consumer Demand Studies 137
Relationships Between Consumer Market Data and Econometric Demand Studies 137
Case Study I: Carnation Coffee-mate 137
Case Study II: Carnation Evaporated Milk 138
Case Study III: The Demand for Cheese in the United States 139
Managerial rule of thumb: Using consumer Market Data 141
End of Chapter Resources
Summary 141 • key terms 141 • exercises 142
• application Questions 143
chapter 5 ProDUction anD coSt analySiS in the Short rUn 144
caSe For analySiS: Production and cost analysis in the Fast-Food industry 145
Defining the Production Function 146
The Production Function 146
Fixed Inputs Versus Variable Inputs 146
Short-Run Versus Long-Run Production Functions 147
Managerial rule of thumb: Short-run Production and long-run Planning 147
Productivity and the Fast-Food Industry 147
Model of a Short-Run Production Function 148
Total Product 148
Average Product and Marginal Product 148
Relationships Among Total, Average, and Marginal Product 149
Economic Explanation of the Short-Run Production Function 151
A01_FARN0095_03_GE_FM.INDD 9 21/08/14 1:52 PM
10 contents
Real-World Firm and Industry Productivity Issues 152
Other Examples of Diminishing Returns 152
Productivity and the Agriculture Industry 153
Productivity and the Automobile Industry 154
Productivity Changes Across Industries 155
Model of Short-Run Cost Functions 156
Measuring Opportunity Cost: Explicit Versus Implicit Costs 156
Accounting Profit Measures Versus Economic Profit Measures 157
Managerial rule of thumb: the importance of opportunity costs 158
Definition of Short-Run Cost Functions 159
Fixed Costs Versus Variable Costs 159
Relationships Among Total, Average, and Marginal Costs 160
Relationship Between Short-Run Production and Cost 162
Other Short-Run Production and Cost Functions 163
Managerial rule of thumb: Understanding your costs 164
Empirical Evidence on the Shapes of Short-Run Cost Functions 164
Econometric Estimation of Cost Functions 164
Survey Results on Cost Functions 165
Constant Versus Rising Marginal Cost Curves 166
Implications for Managers 167
End of Chapter Resources
Summary 168 • key terms 168 • exercises 169 • application Questions 170
chapter 6 ProDUction anD coSt analySiS in the lonG rUn 172
caSe For analySiS: the iPhone in china 173
Model of a Long-Run Production Function 174
Input Substitution 174
Model of a Long-Run Cost Function 182
Derivation of the Long-Run Average Cost Curve 182
Economies and Diseconomies of Scale 183
Factors Creating Economies and Diseconomies of Scale 184
Other Factors Influencing the Long-Run Average Cost Curve 185
The Minimum Efficient Scale of Operation 186
Long-Run Average Cost and Managerial Decision Making 189
End of Chapter Resources
Summary 189 • appendix 6a isoquant analysis 190 • key terms 197
• exercises 197 • application Questions 198
chapter 7 Market StrUctUre: PerFect coMPetition 200
caSe For analySiS: competition and cooperative Behavior in the
Potato industry 201
The Model of Perfect Competition 202
Characteristics of the Model of Perfect Competition 202
Model of the Industry or Market and the Firm 203
The Short Run in Perfect Competition 209
Long-Run Adjustment in Perfect Competition: Entry and Exit 209
Adjustment in the Potato Industry 210
Long-Run Adjustment in Perfect Competition: The Optimal Scale of Production 211
A01_FARN0095_03_GE_FM.INDD 10 21/08/14 1:52 PM
contents 11
Managerial rule of thumb: competition Means little control over Price 212
Other Illustrations of Competitive Markets 212
Competition and the Agricultural Industry 213
Competition and the Broiler Chicken Industry 214
Competition and the Red-Meat Industry 215
Competition and the Milk Industry 217
Competition and the Trucking Industry 218
Managerial rule of thumb: adopting Strategies to Gain Market Power
in competitive industries 219
End of Chapter Resources
Summary 220 • appendix 7a industry Supply 220 • key terms 222
• exercises 222 • application Questions 223
chapter 8 Market StrUctUre: MonoPoly anD MonoPoliStic coMPetition 226
caSe For analySiS: changing Market Power for eastman kodak co. 227
Firms with Market Power 228
The Monopoly Model 228
Comparing Monopoly and Perfect Competition 230
Sources of Market Power: Barriers to Entry 231
Managerial rule of thumb: Using lock-in as a competitive Strategy 241
Changes in Market Power 241
Measures of Market Power 243
Antitrust Issues 246
Managerial rule of thumb: Understanding antitrust laws 251
Monopolistic Competition 251
Characteristics of Monopolistic Competition 252
Short-Run and Long-Run Models of Monopolistic Competition 252
Examples of Monopolistically Competitive Behavior 253
Managerial rule of thumb: Maintaining Market Power in Monopolistic
competition 256
End of Chapter Resources
Summary 256 • key terms 256 • exercises 257 • application Questions 257
chapter 9 Market StrUctUre: oliGoPoly 260
caSe For analySiS: oligopoly Behavior in the airline industry 261
Case Studies of Oligopoly Behavior 262
The Airline Industry 262
The Soft Drink Industry 264
The Doughnut Industry 265
The Parcel and Express Delivery Industry 266
Oligopoly Models 267
Noncooperative Oligopoly Models 268
The Kinked Demand Curve Model 268
Game Theory Models 269
Strategic Entry Deterrence 272
Predatory Pricing 273
A01_FARN0095_03_GE_FM.INDD 11 21/08/14 1:52 PM
12 contents
Cooperative Oligopoly Models 275
Cartels 275
Tacit Collusion 281
Managerial rule of thumb: coordinated actions 283
End of Chapter Resources
Summary 283 • key terms 283 • exercises 283 • application Questions 285
chapter 10 PricinG StrateGieS For the FirM 288
caSe For analySiS: airline Pricing Strategies: will they Start charging for the
Use of the lavatories? 289
The Role of Markup Pricing 290
Marginal Revenue and the Price Elasticity of Demand 291
The Profit-Maximizing Rule 292
Profit Maximization and Markup Pricing 292
Business Pricing Strategies and Profit Maximization 294
Markup Pricing Examples 295
Managerial rule of thumb: Markup Pricing 296
Price Discrimination 297
Definition of Price Discrimination 297
Theoretical Models of Price Discrimination 298
Price Discrimination and Managerial Decision Making 305
Marketing and Price Discrimination 312
Macroeconomics and Pricing Policies 313
End of Chapter Resources
Summary 315 • key terms 315 • exercises 316 • application Questions 317
Part 2 MacroeconoMic analySiS 320
chapter 11 MeaSUrinG MacroeconoMic activity 320
caSe For analySiS: Measuring changes in Macroeconomic activity:
implications for Managers 321
Measuring Gross Domestic Product (GDP) 322
The Circular Flow in a Mixed, Open Economy 322
Managerial rule of thumb: Spending Patterns 324
National Income Accounting Systems 324
Characteristics of GDP 325
Real Versus Nominal GDP 326
Alternative Measures of GDP 329
Other Important Macroeconomic Variables 337
Price Level Measures 337
Measures of Employment and Unemployment 341
Managerial rule of thumb: Price level and Unemployment 343
Major Macroeconomic Policy Issues 343
What Factors Influence the Spending Behavior of the Different Sectors of the
Economy? 344
How Do Behavior Changes in These Sectors Influence the Level of Output and
Income in the Economy? 344
Can Policy Makers Maintain Stable Prices, Full Employment, and Adequate Economic
Growth over Time? 344
A01_FARN0095_03_GE_FM.INDD 12 21/08/14 1:52 PM
contents 13
How Do Fiscal, Monetary, and Balance of Payments Policies Influence
the Economy? 346
What Impact Do These Macro Changes Have on Different Firms
and Industries? 346
Managerial rule of thumb: competitive Strategies and the Macro environment 346
End of Chapter Resources
Summary 347 • key terms 347 • exercises 348 • application Questions 349
chapter 12 SPenDinG By inDiviDUalS, FirMS, anD GovernMentS on real GooDS
anD ServiceS 350
caSe For analySiS: Mixed Signals on the U.S. economy in Summer 2012 351
Framework for Macroeconomic Analysis 352
Focus on the Short Run 352
Analysis in Real Versus Nominal Terms 353
Treatment of the Foreign Sector 353
Outline for Macroeconomic Analysis 353
The Components of Aggregate Expenditure 354
Personal Consumption Expenditure 354
Gross Private Domestic Investment Expenditure 362
Government Expenditure 371
Net Export Expenditure 372
Aggregate Expenditure and Equilibrium Income and Output 375
Aggregate Expenditure 375
Equilibrium Level of Income and Output 377
Effect of the Interest Rate on Aggregate Expenditures 382
End of Chapter Resources
Summary 383 • appendix 12a numerical example of equilibrium and the
Multiplier 383 • appendix 12B algebraic Derivation of the aggregate expenditure
Function 385 • key terms 388 • exercises 388 • application Questions 389
chapter 13 the role oF Money in the Macro econoMy 390
caSe For analySiS: the chairman’s Quandary 391
Money and the U.S. Financial System 392
Definition of Money 392
Measures of the Money Supply 392
Depository Institutions and the Fractional Reserve Banking System 393
The Central Bank (Federal Reserve) 396
Tools of Monetary Policy 398
Managerial rule of thumb: Federal reserve Policy 407
Equilibrium in the Money Market 407
The Supply of Money 407
The Demand for Money 409
Equilibrium in the Money Market 411
Change in the Supply of Money 411
Change in the Demand for Money 412
Overall Money Market Changes 413
End of Chapter Resources
Summary 413 • appendix 13a Monetary tools and the Market for Bank
reserves 413 • key terms 414 • exercises 415 • application Questions 415
A01_FARN0095_03_GE_FM.INDD 13 21/08/14 1:52 PM
14 contents
chapter 14 the aGGreGate MoDel oF the Macro econoMy 416
caSe For analySiS: what role for inflation? 417
The Model of Aggregate Demand and Supply 418
The Aggregate Demand Curve 418
Fiscal and Monetary Policy Implementation 422
The Aggregate Supply Curve 427
Using the Aggregate Model to Explain Changes in the Economy from 2007 to 2008
and from 2011 to 2012 434
Impact of Macro Changes on Managerial Decisions 438
Measuring Changes in Aggregate Demand and Supply 440
Managerial rule of thumb: Judging trends in economic indicators 442
End of Chapter Resources
Summary 442 • appendix 14a Specific and General equations for
the aggregate Macro Model 442 • key terms 444 • exercises 444
• application Questions 445
chapter 15 international anD Balance oF PayMentS iSSUeS in the Macro econoMy 446
caSe For analySiS: Uncertainty in the world economy in 2012 447
Exchange Rates 448
Managerial rule of thumb: currency exchange rates 451
Equilibrium in the Open Economy 452
U.S. International Transactions in 2011 (Balance of Payments) 453
The Current Account 453
The Financial Account 454
Revenue or T-Account 455
Deriving the Foreign Exchange Market 457
The Demand for and Supply of Dollars in the Foreign Exchange Market 457
Equilibrium in the Foreign Exchange Market 459
Managerial rule of thumb: the Foreign exchange Market 460
Exchange Rate Systems 460
Flexible Exchange Rate System 462
Fixed Exchange Rate System 463
The Effect on the Money Supply 465
Sterilization 465
Policy Examples of International Economic Issues 466
The U.S. Economy, 1995–2000 466
The U.S. Economy, 2007–2008 and 2010–2012 468
Effects of the Euro in the Macroeconomic Environment 470
Euro Macro Environment Effects on Managerial Decisions 473
Southeast Asia: An Attempt to Maintain Fixed Exchange Rates 474
Macro and Managerial Impact of the Chinese Yuan Since 2003 476
Policy Effectiveness with Different Exchange Rate Regimes 478
End of Chapter Resources
Summary 479 • appendix 15a Specific and General equations for
the Balance of Payments 479 • key terms 480 • exercises 480
• application Questions 481
A01_FARN0095_03_GE_FM.INDD 14 21/08/14 1:52 PM
contents 15
Part 3 inteGration oF the FraMeworkS 482
chapter 16 coMBininG Micro anD Macro analySiS For ManaGerial DeciSion MakinG 482
caSe For analySiS: Strong headwinds for McDonald’s 483
Microeconomic and Macroeconomic Influences on McDonald’s and the
Fast-Food Industry 484
Shifting Product Demand 484
Oligopolistic Behavior 485
Strategies to Offset Shifting Demand 487
Cost-Cutting Strategies 488
Innovations for Different Tastes 488
Drawing on Previous Experience 489
2012 and Beyond: A Focus on China and Other Emerging Markets 490
Economic and Political Issues 492
Responses of Other Fast-Food Competitors 494
Calorie Counts on Menus 495
Macroeconomic Influences on the Fast-Food Industry in 2011 and 2012 497
End of Chapter Resources
Summary: Macro and Micro influences on the Fast-Food industry 497
• appendix 16a Statistical estimation of Demand curves 498
• exercises 500 • application Questions 500
SolUtionS to even-nUMBereD ProBleMS 501
GloSSary 521
inDex 535
A01_FARN0095_03_GE_FM.INDD 15 21/08/14 1:52 PM
A01_FARN0095_03_GE_FM.INDD 16 21/08/14 1:52 PM
The third edition of Economics for Managers builds on the strengths of the first
two editions, while updating the case studies and examples, the data, and the
references supporting the discussion. Economics for Managers, Third Edition,
does not attempt to cover all the topics in traditional principles of economics
texts or in intermediate microeconomic and macroeconomic theory texts. As in
the previous editions, the goal of this text is to present the fundamental ideas of
microeconomics and macroeconomics and then integrate them from a manage-
rial decision-making perspective into a framework that can be used in a single-
semester course for Master of Business Administration (MBA), Executive MBA
(EMBA), and other business students.
What’s New in this Edition?
This edition has been completely revised to update the industry cases and examples
in the microeconomics section and the data and analysis in the macroeconomics
section.
• Twelve of the sixteen chapters have entirely new cases, while the cases in the
remaining four chapters have been updated extensively.
• New cases include: Micro- and Macroeconomic Influences on the Global Automo-
bile Industry; Demand Elasticity and Procter & Gamble’s Pricing Strategies; The
iPhone in China; Changing Market Power for Eastman Kodak Co.; Airline Pricing
Strategies: Will They Start Charging for the Use of the Lavatories?; Mixed Signals
on the U.S. Economy in Summer 2012; The Chairman’s Quandary; and Strong
Headwinds for McDonald’s.
• Linkage to the marketing literature, particularly in Chapter 4, Techniques for
Understanding Consumer Demand and Behavior, and in Chapter 10, Pricing Strat-
egies for the Firm, has been increased.
• The macroeconomics section of the text has been completely rewritten, given the
changes in the macroeconomy since 2008, when the second edition was drafted.
• The macroeconomic data in the tables have been updated to 2011, and the data in
the figures show trends from 2000 to first quarter 2012.
• The macroeconomics discussion, which makes extensive use of Federal Reserve
Monetary Policy Reports to Congress and reports and analyses by the Congressio-
nal Budget Office, includes recent policy issues such as the impact of the Ameri-
can Recovery and Reinvestment Act of 2009, the fiscal cliff debates in 2012, and
the Federal Reserve’s use of nontraditional policy tools to stimulate the economy.
• An extensive discussion of the situation in the European Union from 2010 to 2012,
which includes the banking, sovereign debt, and growth crises and the impact of
these events on managerial decision making, is presented.
Preface
A01_FARN0095_03_GE_FM.INDD 17 21/08/14 1:52 PM
18 Preface
Motivation for the text
Most micro/managerial economics and intermediate macroeconomics texts are
written for economics students who will spend an entire semester using each
text. The level of detail and style of writing in these texts are not appropriate for
business students or for the time frame of a single-semester course. However,
business students need more than a principles of economics treatment of these
topics because they have often been exposed to that level of material already.
The third edition of Economics for Managers will continue to present economic
theory that goes beyond principles of economics, but the text is not as detailed
or theoretical as a standard intermediate economics text given the coverage of
both micro- and macroeconomics and the additional applications and examples
included in this text. The compactness of the text and the style of writing are
more appropriate for MBA students than what is typically found in large, compre-
hensive principles texts.
As in the previous editions, each chapter of Economics for Managers, Third
Edition, begins with a “Case for Analysis” section, which examines events drawn
from the current news media that illustrate the issues in the chapter. Thus, stu-
dents begin the study of each chapter with a concrete, real-world example that
highlights relevant economic concepts, which are then explained with the appro-
priate economic theory. Numerous real-world examples are used to illustrate the
theoretical discussion. This approach appeals to MBA students who typically
want to know the relevance and applicability of basic economic concepts and
how these concepts can be used to analyze and explain events in the business
environment.
Intended Audience
This text is designed to teach economics for business decision making to students
in MBA and EMBA programs. It includes fundamental microeconomic and macro-
economic topics that can be covered in a single quarter or semester or that can be
combined with other readers and case studies for an academic year course. The
book is purposely titled Economics for Managers and not Managerial Economics
to emphasize that this is not another applied microeconomics text with heavy
emphasis on linear programming, multiple regression analysis, and other quantita-
tive tools. This text is written for business students, most of whom will not take
another course in economics, but who will work in firms and industries that are
influenced by the economic forces discussed in the text.
A course using this text would ideally require principles of microeconomics and
macroeconomics as prerequisites. However, the text is structured so that it can be
used without these prerequisites. Coverage of the material in this text in one semes-
ter does require a substantial degree of motivation and maturity on the part of the
students. However, the style of writing and coverage of topics in Economics for
Managers will facilitate this process and are intended to generate student interest
in these issues that lasts well beyond the end of the course.
Economics for Managers can be used with other industry case study books, such
as The Structure of American Industry by James Brock. These books present
extensive discussions of industry details from an economic perspective. Although
they focus primarily on microeconomic and managerial topics, these texts can be
used with Economics for Managers to integrate influences from the larger mac-
roeconomic environment with the microeconomic analysis of different firms and
industries.
A01_FARN0095_03_GE_FM.INDD 18 21/08/14 1:52 PM
Preface 19
Organization of the text
The text is divided into three parts. Part 1, Microeconomic Analysis, focuses on how
individual consumers and businesses interact with each other in a market economy.
Part 2, Macroeconomic Analysis, looks at the aggregate behavior of different sec-
tors of the economy to determine how changes in behavior in each of these sectors
influence the overall level of economic activity. And finally, Part 3, Integration of the
Frameworks, draws linkages between Parts 1 and 2.
Although many of the micro- and macroeconomic topics are treated similarly in
other textbooks, this text emphasizes the connections between the frameworks,
particularly in the first and last chapters. Changes in macroeconomic variables,
such as interest rates, exchange rates, and the overall level of income, usually affect
a firm through microeconomic variables such as consumer income, the price of
the inputs of production, and the sales revenue the firm receives. Managers must
be able to analyze factors relating to both market competition and changes in the
overall economic environment so they can develop the best competitive strategies
for their firms.
To cover all this material in one text, much of the detail and some topics found
in other micro and macro texts have been omitted, most of which are not directly
relevant for MBA students. There is no calculus in this text, only basic algebra and
graphs. Algebraic examples are kept to a minimum and used only after the basic
concepts are presented intuitively with examples. Statistical and econometric tech-
niques are covered, particularly for demand estimation, at a very basic level, while
references are provided to the standard sources on these topics. The text places
greater emphasis than other texts on how managers use nonstatistical and market-
ing strategies to make decisions about the demand for their products, and it draws
linkages between the statistical and nonstatistical approaches.
Economics for Managers, Third Edition, includes little formal analysis of input
or resource markets, either from the viewpoint of standard marginal productivity
theory or from the literature on the economics of organization, ownership and
control, and human resource management. The latter are interesting topics that
are covered in other texts with a focus quite different from this one. The mac-
roeconomics portion of this text omits many of the details of alternative macro
theories discussed elsewhere. Students are given the basic tools that will help
them understand macroeconomics as presented in business sources, such as the
Wall Street Journal, that emphasize how the national government and the Federal
Reserve manage the economy to promote full employment, a stable price level, and
economic growth.
Chapter-by-Chapter Breakdown: What’s New
in this Edition?
Part 1: Microeconomic Analysis
The third edition of Economics for Managers includes new and updated cases from
2010 to 2012 that introduce each chapter. In some chapters, the cases are on the
same topic as in previous editions (e.g., the copper industry in Chapter 2) to facili-
tate the transition for current users of the text.
Chapter 1 introduces an entirely new case on the global automobile industry,
which includes a discussion of the microeconomic factors influencing competi-
tion among the major players in the industry, and the impact of macroeconomic
A01_FARN0095_03_GE_FM.INDD 19 21/08/14 1:52 PM
20 Preface
changes on the entire industry. The chapter focuses on the competition between
Japanese and American auto makers, how the American industry has been making
a comeback in recent years, and how that change intensified competition among
the American producers. I also discuss the impact of the 2011 earthquake and
tsunami on the Japanese auto industry and the effect of the 2010 recall and quality
issues on Toyota. Automobile production and demand changes in China are major
issues in this chapter. Moreover, the role of China regarding both individual firms’
strategies and in the larger macroeconomic environment will be a significant factor
throughout this text.
Another theme introduced in this chapter is the impact of the global financial cri-
sis and recession on managerial strategies. General Motors and Chrysler received
a bailout from the U.S. government to help them survive. The ongoing economic
crisis in Europe, which I discuss throughout the text, created major challenges for
all players in the global automobile industry. I also discuss the role of currency
exchange rates, particularly the impact of the strong yen on the Japanese auto
industry.
As in previous editions, this chapter presents the frameworks for the microeco-
nomic and macroeconomic analyses used throughout the text. I introduce the role
of relative prices and discuss the different models of market competition. I also
present the circular flow macroeconomic model that focuses on consumption (C),
investment (I), and government spending (G), and spending on exports (X) and
imports (M). I introduce macro policy issues, including the U.S. Federal Reserve
policy since 2008 of targeting historically low interest rates and fiscal policy issues
such as the American Recovery and Reinvestment Act of 2009.
These microeconomic and macroeconomic issues will be discussed again in the
context of the fast-food industry in Chapter 16. The use of two well-known indus-
tries to frame both the microeconomic and macroeconomic discussion is a unique
feature of Economics for Managers, Third Edition.
Chapter 2 updates the case on the copper industry that introduces the concepts
of demand and supply and shows the extreme volatility of prices in a competitive
industry. The current discussion highlights the issues of the global demand for cop-
per, the particular influence of China, and the problem of copper thefts due to its
high price. I have retained much of the discussion of the copper industry from pre-
vious editions to illustrate the impact of these changes over time. Even though this
chapter focuses on the microeconomic concepts of demand and supply, the copper
industry has been given the name “Dr. Copper,” because strong demand and high
prices can indicate the overall health of the economy.
New examples of the non-price factors influencing demand include (1) the impact
on the Zippo Manufacturing Co. of changing attitudes on cigarette smoking; (2) the
Chinese demand for pecans; (3) the effect of the Japanese earthquake on the
demand for luxury goods in that country; (4) the increased marketing of beer and
other products to the Hispanic community; (5) the return of the practice of layaway
in department stores; and (6) the effect of substitutes on Nestle bottled water.
New examples of the non-price factors influencing supply include (1) the effect
of new technology on pecan growers; (2) the impact of high pecan prices as inputs
for bakers; (3) the impact of high oil prices on the supply of natural gas; and (4) the
effect of Chinese demand on the number of lumber producers. The extensive
numerical example on the copper industry that is used throughout Chapter 2 has
been updated to reflect recent events in the copper industry.
Chapter 3 begins with a new case on the relationship between Procter & Gamble’s
pricing strategy and the price elasticity of demand. I have updated information on
price elasticity for airline prices, gasoline, and illegal substances such as cocaine
and heroin. The discussion of income elasticity now includes the demand for wines,
while the cross-elasticity discussion includes the relationship between airline and
automobile travel, which influenced the regulation of child airline safety seats, and
A01_FARN0095_03_GE_FM.INDD 20 21/08/14 1:52 PM
Preface 21
consumer demand for wireline and wireless phones. I have updated Table 3.7 with
recent estimates of demand elasticity for food, water, and higher education. I have
also increased the discussion of the relationship between the economic and mar-
keting approaches to consumer demand, and I have updated an earlier marketing
demand elasticity study included in previous editions.
Chapter 4 presents a new case on how firms use cable television and Visa/Mas-
terCard information to better understand and impact consumer behavior. Although
successful from the firm’s viewpoint, these strategies have raised concerns over the
invasion of consumer privacy. In the discussion of marketing techniques to estimate
consumer demand, I have drawn extensively on two major marketing references:
Vithala R. Rao, Handbook of Pricing Research in Marketing, 2009, and Thomas T.
Nagle et al., The Strategy and Tactics of Pricing, 5th ed., 2011. I have also updated
the econometric references on estimating consumer demand, and I added a new
case, “Case Study III: The Demand for Cheese in the United States.” I retained the
case study of automobile demand and the illustrations of the use of consumer mar-
ket data in econometric demand studies from the previous editions.
In Chapter 5, I updated the opening case, “Production and Cost Analysis in the
Fast-Food Industry,” by adding a discussion of fast-food delivery in various parts
of the world. New productivity examples in the chapter include (1) the use of addi-
tional workers versus robots by Amazon and Crate & Barrel; (2) eliminating dimin-
ishing returns in hospital emergency rooms; (3) a discussion of Toyota’s quality
problems; and (4) an updated discussion of overall industry productivity increases.
Chapter 6 begins with a new case, “The iPhone in China,” that focuses on the
long-run decisions of Apple Inc. to produce iPhones in China and the controversy
that followed over working conditions in those Chinese factories. I also discuss the
location and production decisions of smaller manufacturers such as Standard Motor
Products of North Carolina. New examples of long-run production and cost decisions
include (1) the use of robots in mining operations and hospitals; (2) crowdsourcing
or farming out production tasks to the general public; (3) law firms’ increased use of
software for the discovery process; (4) the trade-off between airline use of smaller
jets to cut costs and increased time for refueling; and (5) the decreased use of bags
in grocery stores. I also describe the limits of lean production that arose during the
Japanese earthquake and tsunami, and I update the discussion of the use of nurse-
to-patient ratios to regulate hospital staffing decisions.
Chapter 7 begins with an update of the case of the potato industry from the previ-
ous editions of the book. Earlier editions focused on how potato farmers attempted
to move away from the competitive market, where they had very little control
over price. They formed farmers’ cooperatives to help control production and
keep prices high. These moves recently faced consumer challenges as price-fixing
arrangements. I discuss other recent influences on the potato industry, including
obesity concerns resulting from potato consumption and a move to eliminate white
potatoes from federally subsidized school breakfasts and lunches. I also update
the analysis of competitive strategies in the broiler chicken, red meat, milk, and
trucking industries. The red meat industry discussion includes both the issues of
“pink slime” and the National Cattlemen’s Beef Association’s MBA or Master of
Beef Advocacy, a training program to promote and defend red meat. The theme in
all of these cases is how firms deal with the volatility of the competitive market.
Chapter 8 begins with a new case, “Changing Market Power for Eastman Kodak
Co.,” which illustrates how the changing markets for cameras and film eroded the
market power of this well-known company. In the section on the sources of market
power, I have updated the discussion of mergers in the banking, beer, and airlines
industries and examined mergers among pharmacy-benefit managers and law firms.
New licensing examples include the case of interior decorators and the contro-
versy over who can perform teeth whitening. I have updated the patent discus-
sion with the case of Pfizer’s blockbuster drug Lipitor and the patent infringement
A01_FARN0095_03_GE_FM.INDD 21 21/08/14 1:52 PM
22 Preface
case between Apple Inc. and Samsung Electronics Co. The changing market power
section now includes a discussion of how bricks-and-mortar retailers are fighting
showrooming and the consumer use of phone apps to compare prices and then
purchase online.
I have updated the antitrust section with a discussion of the August 2010 revisions
in the antitrust guidelines and the use of the Herfindahl-Hirschman Index (HHI).
I also provided new references on the guidelines. Although the Microsoft antitrust
case was an important illustrative example in the two previous editions of the text,
it has become dated. I replaced this case with a discussion of the failed merger of
AT&T and T-Mobile in 2011. This case clearly illustrates for students the contro-
versies over the existence and use of market power. This discussion is extended in
the monopolistic competition section with an update on the cases of independent
drugstores and booksellers.
The opening case in Chapter 9 builds on the previous case of interdependent air-
line pricing behavior by adding current examples of oligopolistic strategies. I have
also updated the examples of oligopolistic behavior between Coke and Pepsi; in
the doughnut industry; and among DHL, Federal Express, and UPS. I added a dis-
cussion of the predatory pricing case of Spirit Airlines versus Northwest Airlines,
included new references on cartel behavior, and updated the discussion of OPEC
and the diamond cartel.
Chapter 10 begins with the case, “Airline Pricing Strategies: Will They Start Charg-
ing for the Use of the Lavatories?” This case illustrates revenue or yield manage-
ment strategies where the airlines have unbundled their services and are charging
separately for different services based on demand elasticity and consumer willing-
ness to pay. I have extended this discussion throughout the chapter and have drawn
extensively on articles in the Journal of Revenue and Pricing Management, a
source that would be very useful for MBA students. As in Chapter 4, I have included
more linkages with the marketing literature by including examples and citations
from Vithala R. Rao, Handbook of Pricing Research in Marketing, and Thomas
T. Nagle et al., The Strategy and Tactics of Pricing, 5th ed. I updated the discussion
of major league baseball ticket pricing and peak load pricing with smart electric
meters, and I added an example of revenue management by the Atlanta Symphony
Orchestra. At the end of the chapter, I added material to the discussion of the macro
impacts on pricing in 2011–2012.
Part 2: Macroeconomic Analysis
Part 2, Macroeconomic Analysis, continues with the framework in the second edi-
tion. After introducing the macroeconomic variables in Chapter 11, the text dis-
cusses real spending by individuals, firms, and governments (C + I + G + X − M) in
Chapter 12. This material draws on the analyses students see daily in the Wall Street
Journal and other business publications. A discussion of money, money markets,
and Federal Reserve policy is presented in Chapter 13. These elements are com-
bined using the aggregate demand–aggregate supply (AD–AS) model in Chapter 14.
Monetary and fiscal policy implementation issues are also presented in this chapter.
Chapter 15 continues to focus on exchange rate and balance of payments issues and
presents an updated discussion of controversies over the role of the euro and the
Chinese yuan. The text continues to describe the impacts of policy changes in these
areas on the U.S. and foreign economies. However, as in the previous editions,
and unlike the presentation in other texts, Economics for Managers, Third Edi-
tion, has an extensive discussion in both Chapters 14 and 15 of the impact of macro
policy changes on the competitive strategies of both domestic and international
firms. This is a unique feature of this textbook, which makes it most appropriate
for MBA students who will probably never make macroeconomic policy, but who
will work in firms and industries influenced by these policy changes.
A01_FARN0095_03_GE_FM.INDD 22 21/08/14 1:52 PM
Preface 23
The macro section of the second edition of Economics for Managers was
revised just as the U.S. economy was slipping into recession in 2008. Given that
the third edition was revised in 2012, I rewrote most of the macro discussion to
reflect the substantial economic and policy changes during that period. I updated
the references for the national income and product accounts and for the under-
ground economy (Chapter 11). I used 2011 data for all the tables, while the figures
show trends from 2000 to the first quarter of 2012. I updated the discussion of
each component of GDP with recent data and events (Chapter 12), and I made
extensive use of Federal Reserve Monetary Policy Reports to Congress and the
reports and analyses by the Congressional Budget Office. All of the cases reflect
the uncertainty about the U.S. and global economies in 2011–2012 and the slow
recovery from the recession of 2007–2009. In the case in Chapter 13, “The Chair-
man’s Quandary,” I discuss the dilemma facing Federal Reserve officials in sum-
mer 2012 as they made decisions on future monetary policy. This discussion
includes issues related to the continuation of historically low targeted interest
rates, Operation Twist, future bond buying, and the use of public statements to
achieve monetary policy goals. I also refer students to Federal Reserve Chair-
man Ben Bernanke’s 2012 College Lecture Series, The Federal Reserve and the
Financial Crisis.
In Chapter 14, I discuss recent fiscal policy issues, including the American Recovery
and Reinvestment Act of 2009, the fiscal cliff debate in 2012, the impacts of fiscal
multipliers and how they are estimated, the role of automatic stabilizers, and the
interactions between fiscal and monetary policies. On the supply side, I updated the
discussion of productivity growth and the natural rate of unemployment. I kept the
section on the use of the aggregate macro model to explain changes in the economy
from 2007 to 2008, but I added a similar discussion for the period 2011–2012. I also
added a summary of the U.S. auto bailout in 2008–2009, and I discussed the impact
of the uncertain economic recovery on managerial decisions in 2011–2012.
Chapter 15, which is built around the opening case, “Uncertainty in the World
Economy in 2012,” focuses on the U.S. economy and international issues from 2010
to 2012. The chapter includes an analysis of the weakness in the Chinese economy,
the worsening situation in Europe, and capital flows among industrialized and emerg-
ing economies. I have updated all tables and figures, included balance of payments
data for 2011, and I updated the discussion and references on balance of payment
issues and the role of fixed versus flexible exchange rates. I include a discussion
and extensive references on the euro zone situation that involves the banking crisis,
the sovereign debt crisis, the growth crisis, and the issue of the sustainability of the
euro, and I show the impact of the euro crisis on managerial decision making. I also
updated the discussion of the Southeast Asia crisis from the second edition, and
I have included recent policy issues related to the Chinese yuan.
Part 3: Integration of the Frameworks
As noted earlier in this section, in Part 3 we return to the issues first discussed
in Chapter 1, the relationship between microeconomic and macroeconomic influ-
ences on managerial decision making. Chapter 16 presents the case, “Strong
Headwinds for McDonald’s,” which examines the effects of changes in the micro-
economic and macroeconomic environment on McDonald’s competitive strategies.
I discuss current challenges facing the company and how these challenges were
met in the past. I then broaden the discussion to include McDonald’s major rivals in
the fast-food industry, Burger King, Subway, Wendy’s, and Starbucks, and I discuss
the opportunities and challenges facing all of these companies as they enter emerg-
ing markets. I have added a discussion of how these companies are facing public
health concerns over obesity, and I present a detailed discussion of the impact of
regulations requiring calorie counts on menus. I have also kept a statistical study
A01_FARN0095_03_GE_FM.INDD 23 21/08/14 1:52 PM
24 Preface
of fast-food industry demand that was included in previous editions of the text.
I discuss macroeconomic influences on the fast-food industry with details from the
International Monetary Fund Global Economic Report in October 2012.
The text ends by emphasizing its major theme: Changes in the macro environ-
ment affect individual firms and industries through the microeconomic factors
of demand, production, cost, and profitability. Firms can either try to adapt to
these changes or undertake policies to try to modify the environment itself. This
theme is particularly important in this third edition of Economics for Managers,
given the impact of the slow recovery from the 2007–2009 recession on the overall
economy and on the strategies of different firms operating in this environment.
Unique Features of the text
Chapter Opening Cases for Analysis
Each chapter begins with a “Case for Analysis” section, which examines a case
drawn from the current news media that illustrates the issues in the chapter. Thus,
students begin the study of each chapter with a concrete, real-world example that
highlights relevant economic issues, which are then explained with the appro-
priate economic theory. For example, Chapter 2 begins with a case on the cop-
per industry that illustrates forces on both the demand and supply sides of the
market that influence the price of copper and have caused that price to change
over time. This example leads directly to a discussion of demand and supply func-
tions and curves, the concept of equilibrium price and quantity, and changes in
those equilibria. Within this discussion, I include numerous real-world examples
to illustrate demand and supply shifters. The chapter concludes by reviewing how
formal demand and supply analysis relates to the introductory case. Students thus
go from concrete examples to the relevant economic theory and then back to real-
world examples.
Interdisciplinary Focus
Economics for Managers, Third Edition, continues to have an interdisciplinary
focus. For example, Chapter 3 presents demand price elasticity estimates drawn
from both the economics and marketing literature. Empirical marketing and eco-
nomic approaches to understanding consumer demand are both discussed in Chap-
ter 4. The production and cost analysis in Chapters 5 and 6 relates to topics covered
in management courses, while the pricing discussion in Chapter 10 draws exten-
sively on the marketing literature. Thus, the third edition of Economics for Manag-
ers is uniquely positioned to serve the needs of instructors who are trying to inte-
grate both micro- and macroeconomic topics and who want to relate this material
to other parts of the business curriculum.
Focus on Global Issues
Global and international examples are included in both the microeconomic and
macroeconomic sections of the text. For example, Chapter 2 discusses how demand
from China, an earthquake in Chile, and the financial crisis in Europe affected the
copper industry. I revisit these international issues again in Chapters 15 and 16.
Analyses of the impact of changing consumer demand, new production technolo-
gies, and rising input costs on both U.S. and international firms are included in
many of the microeconomic chapters. Chapters 14 and 15 include discussions
of the effects of U.S. and international macroeconomic policy changes on firms
located around the world.
A01_FARN0095_03_GE_FM.INDD 24 21/08/14 1:52 PM
Preface 25
As noted previously, Economics for Managers, Third Edition, takes the unique
approach in Chapters 1 and 16 to discuss the impact of both microeconomic and
macroeconomic factors on firms’ competitive strategies in international markets.
The analysis of the global automobile industry in Chapter 1 and the fast-food indus-
try in Chapter 16 helps students see how economic and political issues around the
world impact managerial decision making. This integration of micro and macro
tools in the global setting has been a key feature of all editions of Economics for
Managers.
Managerial Decision-Making Perspective
Economics for Managers is developed from a firm and industry decision-making
perspective. Thus, the demand and elasticity chapters focus on the implications
of elasticity for pricing policies, not on abstract models of consumer behavior. To
illustrate the basic models of production and cost, the text presents examples of
cost-cutting and productivity-improving strategies that firms actually use. It dis-
cusses the concept of input substitution intuitively with examples, but places the
formal isoquant model in an appendix to Chapter 6. The text then compares and
contrasts the various models of market behavior, incorporating discussions and
examples of the measurement and use of market power, most of which are drawn
from the current news media and the industrial organization literature.
Throughout the chapters you will find “Managerial Rule of Thumb” features,
which are shortcuts for using specific concepts and brief descriptions of important
issues for managers. For example, Chapter 3 contains several quick approaches for
determining price and income elasticities of demand. Chapter 4 includes some key
points for managers to consider when using different approaches to understanding
consumer behavior.
Macroeconomics presents a particular challenge for managers because the sub-
ject matter is traditionally presented from the viewpoint of the decision makers,
either the Federal Reserve or the U.S. Congress and presidential administration.
Although Economics for Managers, Third Edition, covers the models that include
this policy-making perspective, the text also illustrates how the actions of these
policy makers influence the decisions managers make in various firms and indus-
tries. This emphasis is important because most students taking an MBA economics
course will never work or make policy decisions for the Federal Reserve or the U.S.
government, but they are or will be employed by firms that are affected by these
decisions and policies.
End-of-Chapter Exercises
As you will see, some of the end-of-chapter exercises are straightforward calcula-
tion problems that ask students to compute demand-supply equilibria, price elastic-
ities, and profit-maximizing levels of output, for example. However, many exercises
are broader analyses of cases and examples drawn from the news media. These
exercises have a managerial perspective similar to the examples in the text. The
goal is to make students realize that managerial decisions usually involve far more
analysis than the calculation of a specific number or an “optimal” mathematical
result. One of the exercises at the end of each chapter is related to the “Case for
Analysis” discussed at the beginning of that chapter.
Instructor Resource Center
Economics for Managers is connected to the Instructor Resource Center available
at www.pearsonglobaleditions.com/farnham. Instructors can access a variety of
print, digital, and presentation resources available with this text in downloadable
A01_FARN0095_03_GE_FM.INDD 25 21/08/14 1:52 PM
26 Preface
format. Registration is simple and gives you immediate access to new titles and
new editions. As a registered faculty member, you can download resource files and
receive immediate access and instructions for installing course management con-
tent on your campus server. If you ever need assistance, our dedicated technical
support team is ready to help with the media supplements that accompany this text.
Visit http://247pearsoned.custhelp.com/ for answers to frequently asked ques-
tions and toll-free user support phone numbers. The following supplements are
available to adopting instructors:
• Instructor’s Manual
• Test Item File also available in TestGen software for both Windows and Mac
computers
• PowerPoint Presentations containing all figures and tables from the text
Acknowledgments
As with any major project, I owe a debt of gratitude to the many individuals who
assisted with this book.
I first want to thank my friend and colleague, Jon Mansfield, who worked with
me in developing materials for the book. Jon and I have discussed the integration
of microeconomics and macroeconomics for business students for many years as
we both experimented with new ideas for teaching a combined course. We even
team-taught one section of the course for EMBA students so that we could directly
learn from each other. Jon is a great teacher, and his assistance in developing this
approach has been invaluable.
I next want to thank the generations of students I have taught, not only in the
MBA and EMBA programs, but also in the Master of Public Administration, Master
of Health Administration, and Master of Public Health programs at Georgia State.
They made it quite clear that students in professional master’s degree programs
are different from those in academic degree programs. Although these students
are willing to learn theory, they have insisted, sometimes quite forcefully, that the
theory must always be applicable to real-world managerial situations.
I also want to thank my colleagues Professors Harvey Brightman and Yezdi Bha-
da, now retired from Georgia State’s Robinson College of Business, for their teach-
ing seminars and for backing the approach I have taken in this book. I always knew
that business and other professional students learned differently from economics
students. Harvey and Yezdi provided the justification for these observations.
I want to acknowledge the following graduate research assistants supported by
the Department of Economics, Georgia State University, for their contributions to
various editions of the text: Mercy Mvundura, Djesika Amendah, William Holmes,
and Sarah Beth Link. They provided substantial assistance in finding the sources
used in the text and in developing tables and figures for the book.
The Prentice Hall staff has, of course, been of immense help in developing
the third edition of the text. I would especially like to thank David Alexander,
Executive Editor, Pearson Economics, for his support and Lindsey Sloan, Senior
Editorial Project Manager, Economics, Pearson Higher Education, who has been
available to answer all my questions at every step of the project. I would also
like to thank Fran Russello, Pearson Production Manager, and Anand Natarajan,
Project Manager at Integra Software Services, for their assistance in producing
the text.
A01_FARN0095_03_GE_FM.INDD 26 21/08/14 1:52 PM
Preface 27
I would like to thank all those who assisted with supporting materials. Professor Leonie
Stone of SUNY Geneseo contributed to the end-of-chapter questions in the micro section of
the text. I also want to acknowledge the assistance of all the reviewers of the various drafts
of the text. These include:
Gerald Bialka, University of North Florida; John Boschen, College of William and
Mary; Vera Brusentsev, University of Delaware; Chun Lee, Loyola Marymount Univer-
sity; Mikhail  Melnik, Niagara University; Franklin E. Robeson, College of William and
Mary; Dorothy R. Siden, Salem State College; Ira A. Silver, Texas Christian University;
Donald L. Sparks, The Citadel; Kasaundra Tomlin, Oakland University; Doina Vlad, Seton
Hill University; John E. Wagner, SUNY-ESF; E. Anne York, Meredith College.
Finally, I want to thank my wife, Lynn, and daughters, Ali and Jen, for bearing with me dur-
ing the writing of all editions of this text.
—Paul G. Farnham
Pearson would like to thank and acknowledge the following people for their work on the
Global Edition. For her contribution: CHAN Ka Yu Yuka, The Open University of Hong
Kong. And for their reviews: Erkan Ilgün, International Burch University; Rajkishan Nair,
IILM Graduate School of Management; Ozlem Olgu Akdeniz, College of Administrative
Sciences and Economics.
A01_FARN0095_03_GE_FM.INDD 27 21/08/14 1:52 PM
A01_FARN0095_03_GE_FM.INDD 28 21/08/14 1:52 PM
Paul G. Farnham is Associate Professor Emeritus of Economics at Georgia State
University. He received his B.A. in economics from Union College, Schenectady,
New York, and his M.A. and Ph.D. in economics from the University of California,
Berkeley. For over 30 years, he specialized in teaching economics to students in
professional master’s degree programs including the Master of Business Admin-
istration and Executive MBA, Master of Public Administration, Master of Health
Administration, and Master of Public Health. He has received both teaching awards
and outstanding student evaluations at Georgia State. Dr. Farnham’s research
focused first on issues related to the economics of state and local governments and
then on public health economic evaluation issues where he has published articles in
a variety of journals. He co-authored three editions of Cases in Public Policy Analy-
sis (1989, 2000, and 2011), contributed to both editions of Prevention Effectiveness:
A Guide to Decision Analysis and Economic Evaluation (1996, 2003), and wrote
a chapter for the Handbook of Economic Evaluation of HIV Prevention Programs
(1998). He is currently a Senior Service Fellow in the Division of HIV/AIDS Preven-
tion at the Centers for Disease Control and Prevention in Atlanta. Dr. Farnham can
be reached at [email protected].
About the Author
A01_FARN0095_03_GE_FM.INDD 29 21/08/14 1:52 PM
A01_FARN0095_03_GE_FM.INDD 30 21/08/14 1:52 PM
Economics
for Managers
A01_FARN0095_03_GE_FM.INDD 31 21/08/14 1:52 PM
Managers and Economics
Pa
rt
1

M
ic
ro
ec
on
om
ic
A
na
ly
si
s
Why should managers study economics? Many of you are probably asking yourself this question as you open this text. Students in Master of Business Administration (MBA) and Executive MBA programs usually have some knowledge of
the topics that will be covered in their accounting, marketing, finance, and
management courses. You may have already used many of those skills on
the job or have decided that you want to concentrate in one of those areas in
your program of study.
But economics is different. Although you may have taken one or two in-
troductory economics courses at some point in the past, most of you are not
going to become economists. From these economics classes, you probably
have vague memories of different graphs, algebraic equations, and terms such
as elasticity of demand and marginal propensity to consume. However, you
may have never really understood how economics is relevant to managerial
decision making. As you’ll learn in this chapter, managers need to under-
stand the insights of both microeconomics, which focuses on the behavior of
individual consumers, firms, and industries, and macroeconomics, which ana-
lyzes issues in the overall economic environment. Although these subjects are
typically taught separately, this text presents the ideas from both approaches
and then integrates them from a managerial decision-making perspective.
As in all chapters in this text, we begin our analysis with a case study. The
case in this chapter, which focuses on the global automobile industry, pro-
vides an overview of the issues we’ll discuss throughout this text. In particu-
lar, the case illustrates how the automobile industry is influenced by both the
microeconomic issues related to production, cost, and consumer demand and
the larger macroeconomic issues including the uncertainty in global economic
activity, particularly in Europe, and the value of various countries’ currencies
relative to the U.S. dollar.
1
32
M01_FARN0095_03_GE_C01.INDD 32 11/08/14 5:17 PM
Case for analysis
Micro- and Macroeconomic Influences on the Global automobile Industry
1Jeff Bennett, “Corporate News: Passenger Cars Lift U.S. Sales—
Big Gains for Toyota, Honda, Chrysler: Pickup Weakness Weighs
on GM, Ford,” Wall Street Journal (Online), October 3, 2012.
2Christina Rogers, “August U.S. Car Sales Surge,” Wall Street
Journal (Online), September 4, 2012.
In September 2012, U.S. automobile sales increased to 1.19
million cars and light trucks per month, a 12.8 percent increase
from a year earlier. This increase represented an annualized rate
of 14.94 million vehicles, the highest sales rate since March
2008 before the recession began in the United States. Much of
the increase was driven by passenger car sales at Toyota Motor
Corp., Honda Motor Co., and Chrysler Group LLC. There was
a significant increase in sales for Toyota and Honda from the
previous year, as both companies were recovering from the
earthquake that hit Japan in March 2011.1 Analysts noted simi-
lar increases in August 2012 that were attributed to pent-up
consumer demand for replacing aging vehicles and the low-
interest financing and other incentives Japanese auto makers
offered to regain market share lost in 2011 due to the lack of
availability of their cars.2
Automobile production in the United States had expanded
in 2012, given favorable foreign exchange rates and a plenti-
ful supply of affordable labor. Toyota, Honda, and Nissan
Motor Co. all increased their production capacity in the
United States with the goal of shipping automobiles to
Europe, Korea, the Middle East, and other countries. The
strong value of the yen, and conversely the weak U.S. dollar,
gave Japanese producers the incentive to produce cars in the
United States for export around the world. This investment
by foreign automobile producers helped the U.S. economy
that was still struggling to recover from the recession of
2007–2009. Automobile industry employment in the United
States was estimated to increase from 566,400 in 2010 to
756,800 in 2015. Although these estimates were well below
the 1.1 million automobile workers employed in 1999, they
indicated that the economic recovery was moving forward.
General Motors Co., which had once encouraged auto parts
suppliers to relocate in low-wage countries, now encouraged
them to locate near U.S. auto plants.3
U.S. auto producers, who had once essentially lost the com-
petition to their Japanese rivals in the 1980s and 1990s and
who went through government-backed (GM and Chrysler)
or private (Ford) restructurings during the U.S. recession,
regained profitability and invested in the engineering and rede-
sign of their cars. Several Fords were designed with a voice-
operated Sync entertainment system, and the Chevrolet Cruze
that was launched in 2010 came with 10 air bags compared
with 6 for the Toyota Corolla. As the U.S. economy recovered,
Americans also began purchasing more trucks and sport-utility
vehicles (SUVs), which helped to restore profits and market
share for the Detroit auto makers. Trucks and SUVs made up
47.3 percent of the U.S. market in 2009, 50.2 percent in 2010,
and 50.8 percent in 2011. This segment of the market had been
hit particularly hard during the U.S. recession.4
As the U.S. automobile industry revived, the competition
between Ford and GM again became more intense. In 2008,
Ford supported the government bailout for GM and Chrysler
because Ford was worried that a collapse of these companies
would also impact the auto parts industry. As the domestic
auto industry recovered, Ford, which had often focused just
on Toyota as its key competitor, began developing strategies to
counter GM. Ford realized that customers who had long been
loyal to Asian brands were again looking at U.S. cars, given the
generally perceived quality increases in the U.S. auto industry.5
3Joseph B. White, Jeff Bennett, and Lauren Weber, “Car Makers’
U-Turn Steers Job Gains,” Wall Street Journal (Online),
January 23, 2012; Neal Boudette, “New U.S. Car Plants Signal
Renewal for Manufacturing,” Wall Street Journal (Online),
January 26, 2012.
4Mike Ramsey and Sharon Terlep, “Americans Embrace SUVs
Again,” Wall Street Journal (Online), December 2, 2011; Jeff
Bennett and Neal E. Boudette, “Revitalized Detroit Makes Bold
Bets on New Models,” Wall Street Journal (Online), January 9,
2012.
5Sharon Terlep and Mike Ramsey, “Ford and GM Renew a Bitter
Rivalry,” Wall Street Journal (Online), November 23, 2011.
33
M01_FARN0095_03_GE_C01.INDD 33 11/08/14 5:17 PM
34 Part 1 Microeconomic Analysis
Japanese auto makers in 2011 and 2012 faced managerial
decisions that were influenced both by the nature of the com-
petition from their rivals and by macroeconomic conditions,
most importantly the value of the exchange rate between the
yen and the U.S. dollar.6 Production by both Toyota and Honda
was hit by the earthquake and tsunami in Japan in March 2011
and by subsequent flooding in Thailand that disrupted the sup-
ply of electronics and other auto parts made there. Toyota sales
were also influenced by the recall and quality issues in 2010
related to the gas pedal and floor mat design. Honda’s rede-
signed 2012 Civic was criticized for its technology and less-
than-luxurious interior. The car was dropped from Consumer
Reports’ recommended list in August 2011. Honda officials
acknowledged that they had underestimated the competition
from U.S. producers.
The strong yen, which made exports from Japan less price
competitive, also gave the Japanese producers the incentive
to produce their cars in the United States. Honda, which had
produced 1.29 million vehicles in North America in 2010,
planned to open a new plant in Mexico and expand produc-
tion in all seven of its existing assembly plants to 2 million
cars and trucks per year. Production abroad was a particu-
lar issue for Toyota, which made half of its automobiles
in Japan, compared to Honda and Nissan, which produced
about one-third of their output in Japan. The president of
Toyota, Akio Toyoda, grandson of the company founder, had
made a public commitment to build at least 3  million cars
in Japan annually, half of which would be for export. Some
company officials argued for streamlining production in
Japan by decreasing production without raising costs, essen-
tially redefining the economies of scale in the company’s
production process. These officials believed the company
could meet domestic goals with high-precision production,
cost- cutting, and collaboration on new technology with parts
suppliers.
Auto producers also focused on China during this period,
although there was concern about the slowing Chinese econ-
omy.7 Auto sales in China increased only 2.5 percent in 2011
compared with increases of 46 percent in 2009 and 32 percent
in 2010. However, the size of the Chinese economy contin-
ued to be the major incentive for expansion in that country. In
April 2012, Ford announced that it would build its fifth fac-
tory in eastern China as part of its plan to double its produc-
tion capacity and sales outlets in the country by 2015. This
production increase would make the company capable of
producing 1.2 million passenger cars in China, approximately
half of the number of cars it built in North America in 2011.
Ford lagged behind other major auto producers in entering
the world’s largest car market. Ford’s strategy was to build
cars from platforms developed elsewhere to minimize costs.
However, these platforms might not provide enough space
in the back seats to appeal to affluent Chinese, who often
employed drivers. General Motors developed a partnership
with Chinese SAIC Motor Corp. to become the dominant for-
eign competitor in China. This partnership resulted in produc-
tion changes such as designing Cadillacs with softer corners,
dashboards with more gadgets, and increasing the comfort of
the rear seats to appeal to Chinese consumers. The challenge
for GM was that SAIC could also use GM’s expertise and
technology to make itself a major competitor with the U.S.
company. In 2012, the Chinese automobile industry began
increasing exports, although these were not thought to be a
threat in developed markets in the United States and Europe,
given perceived quality issues including lack of air-condition-
ing and power windows. However, Chinese producers were
making inroads into emerging markets in Africa, Asia, and
Latin America.
The other major influence on the global auto industry in
2011 and 2012 was the recession and economic crisis in
Europe.8 In October 2012, Ford announced a plan to cut its
operating losses in Europe by closing three auto-assembly and
parts factories in the region, reduce its workforce by 13 per-
cent, and decrease automobile production by 18 percent. Ford
predicted a loss of $1.5 billion in Europe in 2012 and a similar
loss in 2013. The cost-cutting in Europe was combined with
the introduction of several new commercial vans and SUVs and
the introduction of the Mustang sports car for the first time. All
European auto makers faced decreased car sales and chronic
overcapacity at this time. Daimler AG, maker of Mercedes-
Benz automobiles, announced that it would not achieve its
profit targets, while PSA Peugeot Citroen SA announced a
government bailout of its financing arm and a cost-sharing
pact with General Motors. There had been a smaller decrease
in auto-producing capacity in Europe since the 2008 financial
crisis compared with that during the restructuring of the U.S.
auto industry that was influenced by the federal government
bailout.
8This discussion is based on Sharon Terlep and Sam
Schechner, “GM, Peugeot Take Aim at Europe Woes,” Wall
Street Journal (Online), July 12, 2012; Mike Ramsey, David
Pearson, and Matthew Curtin, “Daimler Warns as Europe Car
Makers Cut Back,” Wall Street Journal (Online), October 24,
2012; and Marietta Cauchi and Mike Ramsey, “Ford to Shut
3 Europe Plants,” Wall Street Journal (Online), October 25,
2012.
7This discussion is based on Andrew Galbraith, “Car Makers
Still Look to China,” Wall Street Journal (Online), April 19,
2012; Sharon Terlep and Mike Ramsey, “Ford Bets $5 Billion on
Made in China,” Wall Street Journal (Online), April 20, 2012;
Chester Dawson and Sharon Terlep, “China Ramps Up Auto
Exports,” Wall Street Journal (Online), April 24, 2012; and
Sharon Terlep, “Balancing the Give and Take in GM’s Chinese
Partnership,” Wall Street Journal (Online), August 19, 2012.
6The following discussion is based on Jeff Bennett and Neal
E. Boudette, “Revitalized Detroit Makes Bold Bets on New
Models”; Mike Ramsey and Yoshio Takahashi, “Car Wreck:
Honda and Toyota,” Wall Street Journal (Online), November 1,
2011; Chester Dawson, “For Toyota, Patriotism and Profits May
Not Mix,” Wall Street Journal (Online), November 29, 2011;
Mike Ramsey and Neal E. Boudette, “Honda Revs Up Outside
Japan,” Wall Street Journal (Online), December 21, 2011; and
Yoshio Takahashi and Chester Dawson, “Japan Auto Makers on
a Roll,” Wall Street Journal (Online), April 22, 2012.
M01_FARN0095_03_GE_C01.INDD 34 11/08/14 5:17 PM
ChaPter 1 Managers and Economics 35
Two Perspectives: Microeconomics
and Macroeconomics
As noted above, microeconomics is the branch of economics that analyzes the
decisions that individual consumers and producers make as they operate in a mar-
ket economy. When microeconomics is applied to business decision making, it is
called managerial economics. The key element in any market system is pricing,
because this type of system is based on the buying and selling of goods and ser-
vices. As we’ll discuss later in the chapter, prices—the amounts of money that
are charged for different goods and services in a market economy—act as signals
that influence the behavior of both consumers and producers of these goods and
services. Managers must understand how prices are determined—for both the
outputs, or products sold by a firm, and the inputs, or resources (such as land,
labor, capital, raw materials, and entrepreneurship) that the firm must purchase
in order to produce its output. Output prices influence the revenue a firm derives
from the sale of its products, while input prices influence a firm’s costs of produc-
tion. As you’ll learn throughout this text, many managerial actions and decisions
are based on expected responses to changes in these prices and on the ability of a
manager to influence these prices.
Managerial decisions are also influenced by events that occur in the larger eco-
nomic environment in which businesses operate. Changes in the overall level of
economic activity, interest rates, unemployment rates, and exchange rates both
at home and abroad create new opportunities and challenges for a firm’s compet-
itive strategy. This is the subject matter of macroeconomics, which we’ll cover
in the second half of this text. Managers need to be familiar with the underly-
ing macroeconomic models that economic forecasters use to predict changes in
the macroeconomy and with how different firms and industries respond to these
changes. Most of these changes affect individual firms via the pricing mechanism,
so there is a strong connection between microeconomic and macroeconomic
analysis.9
In essence, macroeconomic analysis can be thought of as viewing the econ-
omy from an airplane 30,000 feet in the air, whereas with microeconomics the
observer is on the ground walking among the firms and consumers. While on the
ground, an observer can see the interaction between individual firms and con-
sumers and the competitive strategies that various firms develop. At 30,000 feet,
however, an observer doesn’t see the same level of detail. In macroeconomics,
we analyze the behavior of individuals aggregated into different sectors in the
economy to determine the impact of changes in this behavior on the overall level
of economic activity. In turn, this overall level of activity combines with changes
in various macro variables, such as interest rates and exchange rates, to affect
the competitive strategies of individual firms and industries, the subject matter of
microeconomics. Let’s now look at these microeconomic influences on managers
in more detail.
Microeconomics
The branch of economics that
analyzes the decisions that
individual consumers, firms, and
industries make as they produce,
buy, and sell goods and services.
Managerial economics
Microeconomics applied to
business decision making.
Prices
The amounts of money that are
charged for goods and services in
a market economy. Prices act as
signals that influence the behavior
of both consumers and producers
of these goods and services.
Outputs
The final goods and services
produced and sold by firms in
a market economy.
Inputs
The factors of production, such as
land, labor, capital, raw materials,
and entrepreneurship, that are used
to produce the outputs, or final
goods and services, that are bought
and sold in a market economy.
Macroeconomics
The branch of economics that
focuses on the overall level of
economic activity, changes in the
price level, and the amount of
unemployment by analyzing group
or aggregate behavior in different
sectors of the economy.
9Note that the terms micro and macro are used differently in various business disciplines. For example, in
Marketing Management, The Millennium Edition (Prentice Hall, 2000), Philip Kotler describes the “macro
environment” as dealing with all forces external to the firm. His examples include both (1) the gradual open-
ing of new markets in many countries and the growth in global brands of various products (microeconomic
factors for the economist) and (2) the debt problems of many countries and the fragility of the international
financial system (macroeconomic problems from the economic perspective). In each business discipline,
you need to learn how these terms and concepts are defined.
M01_FARN0095_03_GE_C01.INDD 35 11/08/14 5:17 PM
36 Part 1 Microeconomic Analysis
Microeconomic Influences on Managers
The discussion of the global automobile industry in the opening case illustrates sev-
eral microeconomic factors influencing managerial decisions. In 2012, Japanese auto
makers used low-interest financing and other incentives to regain market share lost
in previous years. Toyota had to recover from the impact of its recall and negative
quality issues in 2010, while Honda stumbled on the redesign of its 2012 Civic by not
incorporating features offered by its competitors. U.S. auto makers reengineered and
redesigned their production processes to add features with greater customer appeal.
They also responded to the increased demand for trucks and SUVs, a market seg-
ment that had been negatively impacted by the recession. Ford and GM began reen-
gaging in their traditional market rivalry. All producers who planned to sell in China,
the world’s largest automobile market, had to recognize the difference in tastes and
preferences of Chinese consumers, such as the desire for larger back seats.
Decisions about demand, supply, production, and market structure are all micro-
economic choices that managers must make. Some decisions focus on the factors
that affect consumer behavior and the willingness of consumers to buy one firm’s
product as opposed to that of a competitor. Thus, managers need to understand
the variables influencing consumer demand for their products. Because consumers
typically have a choice among competing products, these choices and the demand
for each product are influenced by relative prices, the price of one good in rela-
tion to that of another, similar good. Relative prices are the focus of microeco-
nomic analysis. The Japanese auto makers’ use of low-interest financing and other
pricing incentives noted above is an example of a strategy based on influencing
relative prices. All auto makers discussed in the case had to respond to changing
consumer demand over time and to variations in consumer tastes and preferences
that influenced demand in different countries.
Production technology and the prices paid for the resources used in production
influence a company’s final costs of production. The relative prices of these resources
or factors of production will influence the choices that managers make among dif-
ferent production methods. Whether a production process uses large amounts of
plant and equipment relative to the amount of workers and whether a business oper-
ates out of a small office or a giant factory are microeconomic production and cost
decisions managers must make. As noted in the case, Ford Motor Co. used produc-
tion platforms developed elsewhere to minimize its production costs as it entered
the Chinese market. However, this cost-minimizing strategy was not appropriate for
producing cars with larger back seats that appealed to affluent Chinese customers.
General Motors also had to redesign its Cadillac to meet Chinese demand.
Markets
All of the auto makers in the opening case made strategic decisions in light of their
knowledge of the market environment or structure. Markets, the institutions and
mechanisms used for the buying and selling of goods and services, vary in structure
from those with hundreds or thousands of buyers and sellers to those with very few
participants. These different types of markets influence the strategic decisions that
managers make because markets affect both the ability of a given firm to influence
the price of its product and the amount of independent control the firm has over
its actions.
There are four major types of markets in microeconomic analysis:
1. Perfect competition
2. Monopolistic competition
3. Oligopoly
4. Monopoly
relative prices
The price of one good in relation to
the price of another, similar good,
which is the way prices are defined
in microeconomics.
Markets
The institutions and mechanisms
used for the buying and selling of
goods and services. The four major
types of markets in microeconomic
analysis are perfect competition,
monopolistic competition,
oligopoly, and monopoly.
M01_FARN0095_03_GE_C01.INDD 36 11/08/14 5:17 PM
ChaPter 1 Managers and Economics 37
These market structures can be located along a continuum, as shown in Figure 1.1.
At the left end of the continuum, there are a large number of firms in the market,
whereas at the right end of the continuum there is only one firm. (We’ll discuss other
characteristics that distinguish the markets later in the chapter.)
The two market structures at the ends of the continuum, perfect competition
and monopoly, are essentially hypothetical models. No real-world firms meet all
the assumptions of perfect competition, and few could be classified as monopo-
lies. However, these models serve as benchmarks for analysis. All real-world firms
contain combinations of the characteristics of these two models. Managers need
to know where their firm lies along this continuum because market structure will
influence the strategic variables that a firm can use to face its competition.
The major characteristics that distinguish these market structures are
1. The number of firms competing with one another that influences the firm’s
control over its price
2. Whether the products sold in the markets are differentiated or undifferentiated
3. Whether entry into and exit from the market by other firms is easy or difficult
4. The amount of information available to market participants
The Perfect Competition Model The model of perfect competition, which
is on the left end of the continuum in Figure 1.1, is a market structure character-
ized by
1. A large number of firms in the market
2. An undifferentiated product
3. Ease of entry into the market
4. Complete information available to all market participants
In perfect competition, we distinguish between the behavior of an individual
firm and the outcomes for the entire market or industry, which represents all firms
producing the product. Economists make the assumption that there are so many
firms in a perfectly competitive industry that no single firm has any influence on
the price of the product. For example, in many agricultural industries, whether an
individual farmer produces more or less product in a given season has no influence
on the price of these products. The individual farmer’s output is small relative to
the entire market, so the market price is determined by the actions of all farm-
ers supplying the product and all consumers who purchase the goods. Because
individual producers can sell any amount of output they bring to market at that
price, we characterize the perfectly competitive firm as a price-taker. This firm
does not have to lower its price to sell more output. In fact, it cannot influence the
price of its product. However, if the price for the entire amount of output in the
market increases, consumers will buy less, and if the market price of the product
decreases, they will buy more.
In the model of perfect competition, economists also assume that all firms in an
industry produce the same homogeneous product, so there is no product differen-
tiation. For example, within a given grade of an agricultural product, potatoes or
peaches are undifferentiated. This market characteristic means that consumers do
not care about the identity of the specific supplier of the product they purchase.
They may not even know who supplies the product, and that knowledge would be
irrelevant to their purchase decision, which will be based largely on the price of
the product.
Perfect competition
A market structure characterized
by a large number of firms in
an industry, an undifferentiated
product, ease of entry into the
market, and complete information
available to participants.
Price-taker
A characteristic of a perfectly
competitive firm in which the firm
cannot influence the price of its
product, but can sell any amount of
its output at the price established
by the market.
Perfect Competition Monopolistic Competition Oligopoly Monopoly
Large Number of Firms Single Firm FIGure 1.1
Market Structure
M01_FARN0095_03_GE_C01.INDD 37 11/08/14 5:17 PM
38 Part 1 Microeconomic Analysis
The third assumption of the perfectly competitive model is that entry into the
industry by other firms is costless. This means that if a perfectly competitive firm
is making a profit (earning revenues in excess of its costs), other firms will also
enter the industry in an attempt to earn profits. However, these actions will com-
pete away excess profits for all firms in a perfectly competitive industry.
The final assumption of the perfectly competitive model is that complete infor-
mation is available to all market participants. This means that all participants
know which firms are earning the greatest profits and how they are doing so.
Thus, other firms can easily emulate the strategies and techniques of the profit-
able firms, which will result in greater competition and further pressure on any
excess profits.
While the details of this process will be described in later chapters, these four
assumptions mean that perfectly competitive firms have no market power—the
ability to influence their prices and develop other competitive strategies that allow
them to earn large profits over longer periods of time. All of the other market struc-
tures in Figure 1.1 represent imperfect competition, in which firms have some
degree of market power. How much market power these firms have and how they
are able to maintain it differ among the market structures.
The Monopoly Model At the right end of the market structure continuum in
Figure 1.1 is the monopoly model, in which a single firm produces a product for
which there are no close substitutes. Thus, as we move rightward along the contin-
uum, the number of firms producing the product keeps decreasing until we reach
the monopoly model of one firm. A monopoly firm typically produces a product
that has characteristics and qualities different from the products of its competi-
tors. This product differentiation often means that consumers are willing to pay
more for this product because similar products are not considered to be close
substitutes.
In the monopoly model, there are also barriers to entry, which are structural,
legal, or regulatory characteristics of the market that keep other firms from easily
producing the same or similar products at the same cost and that give a firm mar-
ket power. However, while market power allows a firm to influence the prices of its
products and develop competitive strategies that enable it to earn larger profits, a
firm with market power cannot sell any amount of output at a given market price,
as in perfect competition. If a monopoly firm raises its price, it will sell less output,
whereas if it lowers its price, it will sell more output.
The Monopolistic Competition and Oligopoly Models The intermediate
models of monopolistic competition and oligopoly in Figure 1.1 better character-
ize the behavior of real-world firms and industries because they represent a blend
of competitive and monopolistic behavior. In monopolistic competition, firms
produce differentiated products, so they have some degree of market power.
However, because these firms are closer to the left end of the continuum in
Figure 1.1, there are many firms competing with one another. Each firm has only
limited ability to earn above-average profits before they are competed away over
time. In oligopoly markets, a small number of large firms dominate the market,
even if other producers are present. Mutual interdependence is the key character-
istic of this market structure because firms need to take the actions of their rivals
into account when developing their own competitive strategies. Oligopoly firms
typically have market power, but how they use that power may be limited by the
actions and reactions of their competitors.
The opening case of this chapter did not explicitly discuss the market structure
of the major auto producers. However, because all of these firms are large multina-
tional companies that sell globally, they obviously have substantial market power
Profit
The difference between the total
revenue that a firm receives for
selling its product and the total cost
of producing that product.
Market power
The ability of a firm to influence the
prices of its products and develop
other competitive strategies that
enable it to earn large profits over
longer periods of time.
Imperfect competition
Market structures of monopolistic
competition, oligopoly, and
monopoly, in which firms have
some degree of market power.
Monopoly
A market structure characterized by
a single firm producing a product
with no close substitutes.
Barriers to entry
Structural, legal, or regulatory
characteristics of a firm and its
market that keep other firms from
easily producing the same or similar
products at the same cost.
Monopolistic competition
A market structure characterized
by a large number of small firms
that have some market power as a
result of producing differentiated
products. This market power can be
competed away over time.
Oligopoly
A market structure characterized by
competition among a small number
of large firms that have market
power, but that must take their
rivals’ actions into account when
developing their own competitive
strategies.
M01_FARN0095_03_GE_C01.INDD 38 11/08/14 5:17 PM
ChaPter 1 Managers and Economics 39
and are located far from the model of perfect competition on the continuum in
Figure 1.1. The case noted that U.S. automobile sales were at an annualized rate of
14.94 million vehicles in 2012.
Large national or multinational companies typically find themselves operating in
multiple markets, making the analysis of market structure more complicated as the
market environment may differ substantially among these markets. Each of these
markets has its own characteristics in terms of the number and size of the competi-
tors and product characteristics. Differences between the Chinese and U.S. mar-
kets were discussed throughout the case.
The Goal of Profit Maximization In all of the market models we have just
presented, we assume that the goal of firms is profit maximization, or earning
the largest amount of profit possible. Because profit, as defined above, represents
the difference between the revenues a firm receives for selling its output and its
costs of production, firms may develop strategies to either increase revenues or
reduce costs in an effort to increase profits. Profits act as a signal in a market
economy. If firms in one sector of the economy earn above-average profits, other
firms will attempt to produce the same or similar products to increase their prof-
itability. Thus, resources will flow from areas of low to high profitability. As we
will see, however, the increased competition that results from this process will
eventually lead to lower prices and revenues, thus eliminating most or all of these
excess profits.
Profitability is the standard by which firms are judged in a market economy.
Profitability affects stock prices and investor decisions. If firms are unprofit-
able, they will go out of business, be taken over by other more profitable com-
panies, or have their management replaced. Subsequently, we model a firm’s
profit- maximization decision largely in terms of static, single-period models where
information on consumer behavior, revenues, and costs is known with certainty.
Real-world managers must deal with uncertainty in all of these areas, which
may lead to less-than-optimal decisions, and managers must be concerned with
maximizing the firm’s value over time. The models we present illustrate the basic
forces influencing managerial decisions and the key role of profits as a motivating
incentive.
Profit maximization
The assumed goal of firms, which
is to develop strategies to earn the
largest amount of profit possible.
This can be accomplished by focus-
ing on revenues, costs, or both.
Managerial rule of thumb
Microeconomic Influences on Managers
To develop a competitive advantage and increase their firm’s profitability, managers need to
understand:
How consumer behavior affects their revenues
How production technology and input prices affect their costs
How the market and regulatory environment in which managers operate influences their ability to
set prices and to respond to the strategies of their competitors ■
Macroeconomic Influences on Managers
The discussion of the impact of the global recession, the continued problems
in Europe’s financial recovery, and the role of currency exchange rates in the
case that opened this chapter can be placed within the circular flow model of
Circular flow model
The macroeconomic model that
portrays the level of economic
activity as a flow of expenditures
from consumers to firms, or
producers, as consumers purchase
goods and services produced by
these firms. This flow then returns
to consumers as income received
from the production process.
M01_FARN0095_03_GE_C01.INDD 39 11/08/14 5:17 PM
40 Part 1 Microeconomic Analysis
macroeconomics, shown in Figure 1.2. This model portrays the level of economic
activity in a country as a flow of expenditures from the household sector to busi-
ness firms as consumers purchase goods and services currently produced by these
firms and sold in the country’s output markets. This flow then returns to consum-
ers as income received for supplying firms with the inputs or factors of produc-
tion, including land, labor, capital, raw materials, and entrepreneurship, which are
bought and sold in the resource markets. These payments, which include wages,
rents, interest, and profits, become consumer income, which is again used to pur-
chase goods and services—hence, the name circular flow. Figure 1.2 also shows
spending by firms, by governments, and by the foreign sector of the economy.
Corresponding to these total levels of expenditures and income are the amounts of
output produced and resources employed.
The levels of expenditures, income, output, and employment in relation to the
total capacity of the economy to produce goods and services will determine whether
resources are fully employed in the economy or whether there is unemployed labor
and excess plant capacity. This relationship will also determine whether and how
much the absolute price level in the economy is increasing. The absolute price level
is a measure of the overall price level in the economy as compared with the micro-
economic concept of relative prices, which refers to the price of one particular good
compared to that of another, as we discussed earlier.
Economists use the circular flow model in Figure 1.2 to define and analyze the
spending behavior of different sectors of the economy, including
Personal consumption expenditures (C) by all households on durable goods,
nondurable goods, and services
Gross private domestic investment spending (I) by households and firms
on nonresidential structures, equipment, software, residential structures, and
inventories
absolute price level
A measure of the overall level of
prices in the economy.
Personal consumption
expenditures (C)
The total amount of spending by
households on durable goods,
nondurable goods, and services in a
given period of time.
Gross private domestic
investment spending (I)
The total amount of spending
on nonresidential structures,
equipment, software, residential
structures, and business inventories
in a given period of time.
Foreign Sector
M X
Borrowing
Revenue
Expenses
BorrowingBorrowing
Domestic Markets for
Currently Produced
Goods and Services
Government Sector
Financial Markets
Income:
Wages,
Rent,
Interest,
Profit
Resource
Markets
TP
S
G
Y
C
I
TB
Household Sector Firm Sector
FIGure 1.2
GDP and the Circular Flow
C = consumption spending
I = investment spending
G = government spending
X = export spending
M = import spending
Y = household income
S = household saving
TP = personal taxes
TB = business taxes
M01_FARN0095_03_GE_C01.INDD 40 11/08/14 5:17 PM
ChaPter 1 Managers and Economics 41
Federal, state, and local government consumption expenditures and gross
investment (G)
Net export spending (F) or total export spending (X) minus total import
spending (M)
Consumption spending (C) is largely determined by consumer income (Y), but it
is also influenced by other factors such as consumer confidence, as noted below.
Much business investment spending (I) is derived from borrowing in the financial
markets and is, therefore, affected by prevailing interest rates. The availability of
funds for borrowing is influenced by the amount of income that consumers save (S)
or do not spend on goods and services.10 Some consumer income (Y) is also used
to pay personal taxes (TP) to the government sector to finance the purchase of its
goods and services. The government also imposes taxes on business (TB). If govern-
ment spending (G) exceeds the total amount of taxes collected (T = TP + TB), the
resulting deficit must be financed by borrowing in the financial markets. This gov-
ernment borrowing may affect the amount of funds available for business invest-
ment, which in turn may cause interest rates to change, influencing firms’ costs of
production.
The foreign sector also plays a role in a country’s circular flow of expenditures
because some currently produced goods and services are purchased by residents
of other countries, exports (X), while a given country’s residents use some of
their income to purchase goods and services produced in other countries, imports
(M). Net export spending (F), or export spending (X) minus import spending (M),
measures the net effect of the foreign sector on the domestic economy. Import
spending is subtracted from export spending because it represents a flow of expen-
ditures out of the domestic economy to the rest of the world.11
Spending by all these sectors equals gross domestic product (GDP), the com-
prehensive measure of overall economic activity that is used to judge how an
economy is performing. Gross domestic product measures the market value of all
currently produced final goods and services within a country in a given period of
time by domestic and foreign-supplied resources. GDP equals the sum of consump-
tion spending (C), investment spending (I), government spending (G), and export
spending (X) minus import spending (M).
Factors Affecting Macro Spending Behavior
In macroeconomics, we develop models that explain the behavior of these differ-
ent sectors of the economy and how changes in this behavior influence the overall
level of economic activity, or GDP. These behavior changes arise from
1. Changes in the consumption and investment behavior of individuals and firms
in the private sector of the economy
2. New directions taken by a country’s monetary or fiscal policy-making institu-
tions (its central bank and national government)
3. Developments that occur in the rest of the world that influence the domestic
economy
Government consumption
expenditures and gross
investment (G)
The total amount of spending
by federal, state, and local
governments on consumption
outlays for goods and services,
depreciation charges for existing
structures and equipment, and
investment capital outlays for newly
acquired structures and equipment
in a given period of time.
Net export spending (F)
The total amount of spending on
exports (X) minus the total amount
of spending on imports (M) or
(F = X − M) in a given period of time.
export spending (X)
The total amount of spending
on goods and services currently
produced in one country and
sold abroad to residents of other
countries in a given period of time.
Import spending (M)
The total amount of spending
on goods and services currently
produced in other countries and
sold to residents of a given country
in a given period of time.
Gross domestic product
(GDP)
The comprehensive measure of the
total market value of all currently
produced final goods and services
within a country in a given period
of time by domestic and foreign-
supplied resources.
10Households also borrow from the financial markets, but they are net savers on balance.
11If a country’s export spending and import spending do not balance, there will be a flow of financial capital
among different countries. This flow will affect a country’s currency exchange rate, the rate at which one
country’s currency can be exchanged for another (Chapter 15).
M01_FARN0095_03_GE_C01.INDD 41 11/08/14 5:17 PM
42 Part 1 Microeconomic Analysis
Changes in Private-Sector Behavior Although there are many factors
that influence consumption spending (C) and investment spending (I), credit
availability, consumer wealth in the housing and stock markets, and confidence
on the part of both consumers and businesspeople were extremely important fac-
tors influencing the U.S. economy in the recession of 2007–2009 and the slow
economic recovery since that time.
Monetary Policies In response to the slowing U.S. economy in 2007, the
Federal Reserve, the central bank in the United States, began lowering its tar-
geted interest rate, which had been 5.25 percent since June 2006. In December
2008, the Federal Reserve cut the targeted rate to historic lows of between 0 and
0.25 percent. This policy has been maintained up through the writing of this chap-
ter. These rate changes were reactions to sluggish growth in consumer spending,
employment and manufacturing activity, continued turmoil in the housing and
financial markets, and sharp drops in the stock market. Managers in any econ-
omy must be aware of the monetary policies of their country’s central bank
that influence interest rates and the amount of funds available for consumer and
business loans.
Fiscal Policies To respond to the recession and financial crisis in the United
States, Congress passed the American Recovery and Reinvestment Act (ARRA)
in February 2009. This legislation represented changes in fiscal policy—taxing
and spending policies by a country’s national government that can be used to
either stimulate or restrain the economy (T = TP + TB and G in the circular flow
model in Figure 1.2). Fiscal policy decisions are made by a country’s executive
and legislative institutions, such as the president, his or her administration, and
the Congress in the United States. As a result, fiscal policy actions may be under-
taken to promote political as well as economic goals.
The ARRA had numerous spending and revenue provisions that can be grouped
as follows: (1) providing funds to states and localities, including aid for education
and support for transportation projects; (2) supporting people in need through
measures such as extending unemployment benefits; (3) purchasing goods and
services including construction and other investment activities; and (4) providing
temporary tax relief for individuals and businesses.12
Changes in the Foreign Sector The opening case for this chapter noted
that the strong yen, which made exports from Japan less price competitive, gave
Japanese producers the incentive to produce cars in the United States. This was
a strategic problem for Toyota, whose president had made a public commitment
to build at least 3 million cars in Japan annually. The value at which a country’s
currency can be exchanged for another currency affects the flow of imports and
exports to and from the country and the level of economic activity in the country.
Policies to keep that exchange rate at a certain level can have negative effects on
other economic goals and can be offset by the actions of currency traders in finan-
cial markets. Exchange rate policies need to be coordinated with monetary and
fiscal policies to maintain the proper rate of economic growth.
Monetary policies
Policies adopted by a country’s
central bank that influence the
money supply, interest rates, and
the amount of funds available for
loans, which, in turn, influence
consumer and business spending.
Fiscal policy
Changes in taxing and spending
by the executive and legislative
branches of a country’s national
government that can be used to
either stimulate or restrain the
economy.
12Congressional Budget Office. Estimated Impact of the American Recovery and Reinvestment Act on
Employment and Economic Output from April 2012 Through June 2012. August 2012. Available at
www.cbo.gov.
M01_FARN0095_03_GE_C01.INDD 42 11/08/14 5:17 PM
ChaPter 1 Managers and Economics 43
Managerial rule of thumb
Macroeconomic Influences on Managers
Changes in the macro environment affect individual firms and industries through the microeco-
nomic factors of demand, production, cost, and profitability. Managers don’t have control over
these changes in the larger macroeconomic environment. However, managers must be aware of the
developments that will have a direct impact on their businesses. Managers sometimes hire outside
consultants for reports on the macroeconomic environment, or they ask in-house staff to prepare
forecasts. In any case, they need to be able to interpret these forecasts and then project the impact
of these macroeconomic changes on the competitive strategies of their firms. Although overall
macroeconomic changes may be the same, their impact on various firms and industries is likely to
be quite varied. ■
Summary
In this chapter, we discussed the reasons why both microeconomic and
macroeconomic analyses are important for managerial decision making.
Microeconomics focuses on the decisions that individual consumers and firms
make as they produce, buy, and sell goods and services in a market economy,
while macroeconomics analyzes the overall level of economic activity, changes
in the price level and unemployment, and the rate of economic growth for the
economy. All of these factors affect the decisions managers make in developing
competitive strategies for their firms.
We illustrated these issues by discussing the challenges and problems facing the
global automobile industry in 2011 and 2012. Some of these challenges arose from
changes in consumer preferences and demand over time, while others resulted
from differences in preferences in various markets. However, all automobile pro-
ducers were affected by the slow global economic recovery at this time and by
fluctuating values of currency exchange rates.
We then briefly introduced the concept of market structure and presented the
four basic market models: perfect competition, monopolistic competition, oligop-
oly, and monopoly. We also showed how the economic activity between consumers
and producers fits into the aggregate circular flow model of macroeconomics, and
we defined the basic spending components of that model: consumption, invest-
ment, government spending, and spending on exports and imports. We illustrated
the effects of changes in monetary policy by a country’s central bank and changes
in fiscal policy by the national administrative and legislative institutions on the
overall level of economic activity.
We will next analyze these issues in more detail. We first focus on the microeco-
nomic concepts of demand and supply, pricing, production and cost, and market
structures (Chapters 2 through 10). We’ll then turn our attention to macroeco-
nomic models and data (Chapters 11 through 15). We return to integrate these
issues further where we’ll look at more examples of the combined impact of both
microeconomic and macroeconomic variables on managerial decision making
(Chapter 16).
M01_FARN0095_03_GE_C01.INDD 43 11/08/14 5:17 PM
44 Part 1 Microeconomic Analysis
Application Questions
1. Give illustrations from the opening case in this
chapter of how both microeconomic and macro-
economic factors influence the global automobile
industry.
2. In each of the following examples, discuss which
market model appears to best explain the behav-
ior described:
a. Corn prices reached record highs in the United
States in August 2012, given the worst drought
in decades. However, by October these prices
started to drop again as countries including
China, Japan, and South Korea began to pur-
chase from producers in other countries such
as Argentina and Brazil.13
b. In 2012, Staples Inc., OfficeMax Inc., and
Office Depot Inc. were all closing many stores,
decreasing the size of their stores, and focus-
ing more on online operations. All three chains
struggled to deal with changing consumer
shopping habits as consumers tested equip-
ment in the stores and then made purchases
online.14
13Andrew Johnon Jr., “Weak Exports Hurt Corn,” Wall Street Journal (Online), November 2, 2012.
14Ann Zimmerman and Shelly Banjo, “New Web Victim: Office-Supply Store,” Wall Street Journal (Online),
September 25, 2012.
Key Terms
absolute price level, p. 40
barriers to entry, p. 38
circular flow model, p. 39
export spending (X), p. 41
fiscal policy, p. 42
government consumption
expenditures and gross
investment (G), p. 41
gross domestic product (GDP), p. 41
gross private domestic investment
spending (I), p. 40
imperfect competition, p. 38
import spending (M), p. 41
inputs, p. 35
macroeconomics, p. 35
managerial economics, p. 35
market power, p. 38
markets, p. 36
microeconomics, p. 35
monetary policies, p. 42
monopolistic competition, p. 38
monopoly, p. 38
net export spending (F), p. 41
oligopoly, p. 38
outputs, p. 35
perfect competition, p. 37
personal consumption
expenditures (C), p. 40
prices, p. 35
price-taker, p. 37
profit, p. 38
profit maximization, p. 39
relative prices, p. 36
Exercises
Technical Questions
1. What are the differences between the microeco-
nomic and macroeconomic perspectives on the
economy?
2. Why are both input and output prices important to
managers?
3. What are the four major types of markets in micro-
economic analysis? What are the key characteris-
tics that distinguish these markets?
4. Since a monopolist has some degree of market
power, and can also take measures to keep
competitors away from the market, a monopo-
list can set the price of their product as high
as they want. The higher the price charged, the
higher the revenue. Do you agree? Explain your
answer.
5. In macroeconomics, what are the five major cat-
egories of spending that make up GDP? Are all five
categories added together to determine GDP?
6. Discuss the differences between fiscal and mon-
etary policies.
M01_FARN0095_03_GE_C01.INDD 44 11/08/14 5:17 PM
ChaPter 1 Managers and Economics 45
15Anton Troianovski, “T-Mobile Finds a New Lifeline,” Wall Street Journal (Online), October 2, 2012.
16Shirley Leung, “Big Chains Talk the Talk, But Can’t Walk the Wok,” Wall Street Journal, January 23, 2003.
c. In fall 2012, T-Mobile announced it was close
to a merger with its smaller rival MetroPCS.
This merger would strengthen T-Mobile’s posi-
tion as the fourth-largest wireless operator in
the United States. The merger would allow the
combined company to cut costs and operate on
a larger scale.15
d. Chinese cooking is the most popular food in
America that isn’t dominated by big national
chains. Chinese food is typically cooked in a
wok that requires high heat and a special stove.
Specialized chefs are also required. Small mom-
and-pop restaurants comprise nearly all of the
nation’s 36,000 Chinese restaurants, which have
more locations than McDonald’s, Burger King,
and Wendy’s combined.16
3. HSBC’s revenue after insurance claims fell from
$68.3 billion in 2012 to $64.6 billion in 2013.17 Does
it necessarily mean that HSBC made less profit in
2013 than in 2012? Explain your answer.
4. The slow recovery from the recession of 2007–
2009 forced many firms to develop new competi-
tive strategies to survive. Find examples of these
strategies in various business publications.
17Howard Mustoe and Gavin Finch, “HSBC’s 2013 Profit Misses Estimates on Cost Reductions,” Bloomberg
(Online), February 25, 2014.
M01_FARN0095_03_GE_C01.INDD 45 11/08/14 5:17 PM
In this chapter, we analyze demand and supply—probably the two most famous words in all of economics. Demand—the functional relationship between the price of a good or service and the quantity demanded by con-sumers in a given period of time, all else held constant—and supply—the
functional relationship between the price of a good or service and the quan-
tity supplied by producers in a given period of time, all else held constant—
provide a framework for analyzing the behavior of consumers and producers
in a market economy. Managers need to understand these terms to develop
their own competitive strategies and to respond to the actions of their com-
petitors. They also need to understand that the role of demand and supply
depends on the environment or market structure in which a firm operates.
We begin our discussion of demand and supply by focusing on an analysis
of the copper industry from 1997/98 to 2011. In our case analysis, we’ll discuss
how factors related to consumer behavior (demand) and producer behav-
ior (supply) determine the price of copper and cause changes in that price.
In the remainder of the chapter, we’ll look at how the factors from the cop-
per industry fit into the general demand and supply framework of economic
theory. We’ll develop a conceptual analysis of demand functions and demand
curves; discuss the range of factors that influence consumer demand; analyze
how demand can be described verbally, graphically, and symbolically using
equations; and look at a specific mathematical example of demand. We’ll then
describe the supply side of the market and the factors influencing supply in
the same manner. Finally, we’ll discuss how demand- and supply-side factors
determine prices and cause them to change.
2 Demand, Supply, and Equilibrium Prices
46
M02_FARN0095_03_GE_C02.INDD 46 13/08/14 1:42 PM
The copper industry illustrates all the factors on the demand
and supply side of a competitive market that we discuss in this
chapter. Shifts in these factors can cause current and expected
future prices of copper to change rapidly. In addition, the cop-
per industry serves as a signal for the status of the global econ-
omy. Because copper is used in so many industries around the
world, the metal has been given the name “Dr. Copper,” since a
strong demand and high prices for it can indicate that the over-
all economy is healthy.1
In February 2011, copper prices reached an all-time high of
$4.62 per pound, having almost quadrupled after a two-year series
of increases. At that time there was a fear that this rally in prices
had stopped, given speculation about events in China. Previously,
traders and industry observers had thought that China had an
insatiable demand for the metal. However, rising interest rates in
China could have forced speculators to sell copper to reduce their
financing costs, while consumers kept their inventories low to
save capital. At this time previously unreported stockpiles of cop-
per were also discovered in China, many of which were in bonded
warehouses where traders stored goods before moving them in or
out of the country. Analysts observed that these supplies could
easily have been moved into the market.2
In April 2011, copper analysts worried about further
decreases in prices. The worldwide economic downturn had
caused demand to decrease in key markets, such as housing and
construction. Copper consumers had reacted to previous high
prices by seeking cheaper alternative substitute materials, such
as aluminum and plastic.3 In June 2011, analysts reported that
copper prices surged to the highest level in two weeks due to
the reporting of better-than-expected Chinese industrial produc-
tion data. Unfavorable U.S. economic data and concern over
Chinese inflation had caused prices to decrease, but the indus-
trial output report indicated that demand could increase again.4
However, later that month concerns over economic conditions
in Europe, an important consumer of copper for plumbing and
electrical wiring, put further downward pressure on prices.5
During the summer of 2011, copper prices increased in
response to a U.S. Department of Labor report that new claims
for unemployment benefits fell for the first time in three weeks.
The widespread use of the metal in construction and manufac-
turing meant that any changes in unemployment could impact
copper prices. There was also concern on the supply side,
given that recent severe winter weather in Chile and a potential
strike at a large copper plant could disrupt production.6
The extreme volatility of the copper market was illustrated
in September 2011. On September 27 the Wall Street Journal
reported that copper prices rose sharply, given a report that the
European Union might expand its support of the Euro zone’s
troubled banks and a Federal Reserve Bank of Chicago report
showing increased manufacturing output in the Midwest region
of the United States.7 However, one day later it was reported
that price declines had erased the previous market increase of
more than 5 percent as investors continued to worry about the
European financial crisis and whether previously anticipated
strong imports into China might not occur.8
Unforeseen events have also influenced the copper market.
Copper prices reached a seven-week high after a massive earth-
quake hit Chile in March 2010. There were concerns that supply
from the world’s largest copper producer would be impacted by
the quake. Analysts attempted to determine as quickly as possi-
ble how much the country’s infrastructure had been damaged.9
Similar factors affected the copper market in 2006 and
2007.10 Analysts predicted a decrease in the supply of copper
in 2007 after many strikes limited production in 2006. This
decreased production along with strong worldwide demand
caused the price of copper to remain at historic highs during
that year. Much of this demand was stimulated by the eco-
nomic growth in China. A lack of new mining projects also
limited supply, given that many large, known copper deposits
were in areas with unstable governments or were difficult to
reach.11 Another impact of the high prices was the increased
theft of copper coils in air-conditioning units, copper wires,
and copper pipes used for plumbing in homes and businesses
Case for Analysis
Demand and Supply in the Copper Industry
1Carolyn Cui and Tatyana Shumsky, “Dr. Copper Offers a Mixed
Prognosis,” Wall Street Journal (Online), April 11, 2011.
2Cui and Shumsky, “Dr. Copper Offers a Mixed Prognosis.”
3Andrea Hotter, “Lofty Copper Prices Remain at Risk,” Wall
Street Journal (Online), April 28, 2011.
4Matt Day, “Copper Rises on China Industrial-Production Data,”
Wall Street Journal (Online), June 14, 2011.
5Amy D’Onofrio, “Copper Falls on Uncertainty Over Global
Economy,” Wall Street Journal (Online), June 20, 2011.
6Matt Day, “Copper Surges on Improved U.S. Labor Market
View,” Wall Street Journal (Online), July 7, 2011.
7Matt Day, “Copper Continues Gains as Global Markets Rally,”
Wall Street Journal (Online), September 27, 2011.
8Matt Day, “Copper Slides to a 13-Month Low on Worries About
Demand,” Wall Street Journal (Online), September 28, 2011.
9Allen Sykora, “Copper Prices Rise Following Quake,” Wall
Street Journal (Online), March 1, 2010.
10Allen Sykora, “Copper Surplus is Foreseen in ’07,” Wall Street
Journal, February 28, 2007.
11Patrick Barta, “A Red-Hot Desire for Copper,” Wall Street
Journal, March 16, 2006.
47
M02_FARN0095_03_GE_C02.INDD 47 13/08/14 1:42 PM
48 PArt 1 Microeconomic Analysis
16Aaron Lucchetti, “Copper Limbo: Just How Low Can It Go?”
Wall Street Journal, February 23, 1998.
in many parts of the United States.12 Thefts, even of cem-
etery bronze vases containing large amounts of copper, con-
tinued with the relatively high prices of copper in subsequent
years.13
Analysts predicted that increased quantities of copper
would be available in 2007 due to several factors. (1) The
strikes that occurred in 2006 were not expected to continue the
next year. (2) The higher copper prices encouraged companies
to mine lower-grade copper that would not have been econom-
ically feasible with lower prices. However, the high copper
prices also gave many copper users the incentive to find sub-
stitutes for the metal. Aluminum producers benefited from the
high copper prices, and these prices stimulated the increased
use of plastic piping in home construction.
Forecasts of future prices and production can be very uncer-
tain, given the variety of factors operating on both the demand
and supply side of the market. One report estimated a surplus
of 108,000 metric tons for the first 11 months of 2006, while
another estimated a surplus of 40,000 metric tons for the entire
year. The extent of substitution with other products was also
difficult to estimate, as was the substitution with scrap metal.14
Moreover, in February 2007, the first impacts of the slowing
housing market on the U.S. economy were just beginning to
appear. This is another example of changes in the macroecon-
omy impacting this industry, leading to the name “Dr. Copper.”
Copper prices continued to be influenced by the demand
from China. This demand slowed in 2008 as the Chinese drew
down their inventories when global prices were high and shut
down some industrial activity preceding the Olympics in August
2008. The slowing Chinese economy in fall 2008 also impacted
the world copper market where prices continued to fall.15
An analysis of a substantial decline in copper prices 10
years earlier from November 1997 to February 1998 illus-
trated many of these same factors.16 The 1997 financial cri-
sis and recession in Southeast Asia had a significant impact
on the copper industry, as did uncertain demand from China
and the increased use of copper in communications technol-
ogy in North America. Expectations also played a role as many
copper users were hesitant to buy because they thought prices
might continue their downward trend.
On the supply side, the low price of copper forced mining
companies to decide whether certain high-cost mines should
be kept in operation. However, a new mining process called
“solvent extraction” also allowed some companies to mine
copper at a lower cost, which permitted more copper mines to
stay in business.
We can see from this discussion that a variety of factors
influence the price of copper and that these factors can be cat-
egorized as operating either on the demand (consumer) side or
the supply (producer) side of the market. Sometimes the influ-
ence of one factor in lowering prices is partially or completely
offset by the impacts of other factors that tend to increase
prices. Thus, the resulting copper prices will be determined by
the magnitude of the changes in all of these variables.
Note also that the case discusses general influences on the
copper industry. There is no discussion of the strategic behav-
ior of individual firms. This focus on the entire industry is a
characteristic of a perfectly or highly competitive market,
where there are many buyers and sellers and the product is rela-
tively homogeneous or undifferentiated. Prices are determined
through the overall forces of demand and supply in these mar-
kets. All firms, no matter where they are located on the market
structure continuum, face a demand from consumers for their
products. The factors influencing demand, which are discussed
in this chapter, thus pertain to firms operating in every type
of market. However, the demand/supply framework and the
resulting determination of equilibrium prices apply only to
perfectly or highly competitive markets. We’ll now examine
the concepts of demand and supply in more detail to see how
managers can use this framework to analyze changes in prices
and quantities of different products in various markets.
12Sara Schaefer Munoz and Paul Glader, “Copper and Robbers:
Homeowners’ Latest Worry,” Wall Street Journal, September 6,
2006.
13Joe Barrett, “Sky-High Metal Prices Lead to a Grave Situation,”
Wall Street Journal (Online), November 22, 2011.
14Sykora, “Copper Surplus is Foreseen in ’07.”
15Allen Sykora, “China Copper Need Set to Rise,” Wall Street
Journal, August 25, 2008; James Campbell and Matthew Walls,
“China Drags Down Metals: Slump in Real Estate: Export
Industries May Keep Lid on Oil Prices, Wall Street Journal,
October 29, 2008; Allen Sykora, “Copper Is Vulnerable to Falling
Further,” Wall Street Journal, November 24, 2008.
Demand
Although demand and supply are used in everyday language, these concepts have
very precise meanings in economics.17 It is important that you understand the dif-
ference between the economic terms and ordinary usage. We’ll look at demand
first and turn our attention to supply later in the chapter.
17Even basic terms such as demand may be defined differently in various business disciplines. For example,
in Marketing Management, The Millennium Edition (Prentice Hall, 2000), Philip Kotler defines market
demand as “the total volume that would be bought by a defined customer group in a defined geographical
area in a defined time period in a defined marketing environment under a defined marketing program”
(p. 120). Since advertising and marketing expenditures are the focus of this discipline, demand is defined
to emphasize these issues rather than price.
M02_FARN0095_03_GE_C02.INDD 48 13/08/14 1:42 PM
ChAPter 2 Demand, Supply, and Equilibrium Prices 49
Demand is defined in economics as a functional relationship between the price
of a good or service and the quantity demanded by consumers in a given period
of time, all else held constant. (The Latin phrase ceteris paribus is often used in
place of “all else held constant.”) A functional relationship means that demand
focuses not just on the current price of the good and the quantity demanded at that
price, but also on the relationship between different prices and the quantities that
would be demanded at those prices. Demand incorporates a consumer’s willing-
ness and ability to purchase a product.
Nonprice Factors Influencing Demand
The demand relationship is defined with “all else held constant” because many
other variables in addition to price influence the quantity of a product that con-
sumers demand. The following sections summarize these variables, many of which
were discussed in the opening case on the copper industry.
Tastes and Preferences Consumers must first desire or have tastes and
preferences for a good. For example, in the aftermath of the September 11, 2001,
terrorist attacks on New York and Washington, D.C., the tastes and preferences
of U.S. consumers for airline travel changed dramatically. People were simply
afraid to fly and did not purchase airline tickets regardless of the price charged. In
October 2001, most of the major airlines began advertising campaigns to increase
consumer confidence in the safety of air travel. United Airlines’ advertisements
featured firsthand employee accounts, while American Airlines encouraged
people to spend time with family and friends over the upcoming holidays and
beyond.18
Changing attitudes toward cigarette smoking have had a major impact on Zippo
Manufacturing Co., which produced “windproof” cigarette lighters for 78 years.
Annual lighter sales decreased from 18 million in 1998 to 12 million in 2010. The
company tried to influence consumer behavior with new lighter designs, includ-
ing those with images of Elvis Presley and the Playboy logo. However, the com-
pany also developed new products including a men’s fragrance, casual clothing,
watches, and camping supplies as a response to these changes in preferences.19
The U.S. pecan industry has been impacted by changing Chinese preferences for
these nuts. China bought one-quarter of the U.S. crop in 2009, whereas the country
had little demand five years earlier. A belief that eating pecans would help ward off
Alzheimer’s disease and influence the brain development of babies helped generate
this demand.20
The Japanese earthquake in March 2011 influenced the demand for luxury goods
in that country. Although Japanese consumers traditionally were willing to pay
some of the world’s highest prices for fashion and other luxury goods, surveys fol-
lowing the quake showed that many consumers believed that showing off luxury
goods was in bad taste. Sales of expensive fashion items and accessories in Japan
were second only to that in the United States before the quake.21
Socioeconomic variables such as age, gender, race, marital status, and level of
education are often good proxies for an individual’s tastes and preferences for a
particular good, because tastes and preferences may vary by these groupings and
products are often targeted at one or more of these groups. Beer brewers have tar-
geted Hispanics who will account for 23 percent of the nation’s legal-drinking-age
Demand
The functional relationship
between the price of a good or
service and the quantity demanded
by consumers in a given time
period, all else held constant.
Functional relationship
A relationship between variables,
usually expressed in an equation
using symbols for the variables,
where the value of one variable, the
independent variable, determines
the value of the other, the depen-
dent variable.
18Melanie Trottman, “Airlines Launch New Ad Campaigns Using Emotion to Restore Confidence,” Wall
Street Journal, October 24, 2001.
20David Wessel, “Shell Shock: Chinese Demand Reshapes U.S. Pecan Business,” Wall Street Journal
(Online), April 18, 2011.
21Mariko Sanchanta, “Japan Grows Leery of Luxury,” Wall Street Journal (Online), May 27, 2011.
19James R. Hagerty, “Zippo Preps for a Post-Smoker World,” Wall Street Journal (Online), March 8, 2011.
M02_FARN0095_03_GE_C02.INDD 49 13/08/14 1:42 PM
50 PArt 1 Microeconomic Analysis
population in 2030, particularly given the decline in overall sales due to high unem-
ployment among men ages 21–34. Corona developed Spanish- and English-language
advertisements focusing on luxurious beach settings to convey the brand’s pre-
mium status. It also developed a 32-ounce bottled version of the beer designed for
family gatherings that was targeted on states with large Hispanic populations, such
as Arizona and California. MillerCoors began a campaign to promote its products
to Mexican soccer fans.22
Similarly, Procter & Gamble Co. retargeted its marketing, changed its mix of
celebrity spokeswomen, and increased the amount of Spanish on its products.
This was part of its competitive strategy particularly in the U.S. toothpaste mar-
ket where Colgate-Palmolive Co. built a dominant position based on its strength in
Latin American markets. Procter & Gamble found that Hispanic customers were
more likely to use fragrances in their homes than other sociodemographic groups.
Hispanic households spent more on cleaning and beauty products and were more
loyal to their brands than the average U.S. customer. Procter & Gamble also used
actress Eva Mendes and singer-actress Jennifer Lopez as spokeswomen to promote
its products in the Hispanic community.23
Economic theory may also suggest that one or more of these socioeconomic vari-
ables influences the demand for a particular good or service. For example, persons
with more education are believed to be more knowledgeable about using preven-
tive services to improve their health. Marital status may influence the demand for
acute care and hospital services because married individuals have spouses who
may be able to help take care of them in the home.24 Thus, tastes and preferences
encompass all the individualistic variables that influence a person’s willingness to
purchase a good.
Income The level of a person’s income also affects demand, because demand
incorporates both willingness and ability to pay for the good. If the demand for
a good varies directly with income, that good is called a normal good. This defi-
nition means that, all else held constant, an increase in an individual’s income
will increase the demand for a normal good, and a decrease in that income will
decrease the demand for that good. If the demand varies inversely with income,
the good is termed an inferior good. Thus, an increase in income will cause a
consumer to purchase less of an inferior good, while a decrease in that income
will actually cause the consumer to demand more of the inferior good. Note that
the term inferior has nothing to do with the quality of the good—it refers only to
how purchases of the good or service vary with changes in income.
Normal Goods In many cases, the effect of income on particular goods and ser-
vices is related to the general level of economic activity in the economy. Although
jewelers used the transition from the year 1999 to 2000 to influence consumer
tastes and preferences for jewelry, the strong economy and the booming stock mar-
ket in 1999 also played a role in influencing demand.25 On the other hand, the loss
of both jobs and stock market wealth in fall 2008 caused retail spending to decline
below already-weak forecasts.26 This frugality continued throughout the recession
and the slow recovery in the subsequent years. Wal-Mart noted an increase in pay-
check-cycle shopping where consumers stocked up on products soon after getting
Normal good
A good for which consumers will
have a greater demand as their
incomes increase, all else held
constant, and a smaller demand
if their incomes decrease, other
factors held constant.
Inferior good
A good for which consumers will
have a smaller demand as their
incomes increase, all else held
constant, and a greater demand
if their incomes decrease, other
factors held constant.
22David Kesmodel, “Brewers Go Courting Hispanics,” Wall Street Journal (Online), July 12, 2011.
23Ellen Bryon, “Hola: P&G Seeks Latino Shoppers,” Wall Street Journal (Online), September 15, 2011.
24The demand for health and medical services is discussed in Donald S. Kenkel, “The Demand for
Preventive Medical Care,” Applied Economics 26 (April 1994): 313–25; and in Rexford E. Santerre and
Stephen P. Neun, Health Economics: Theories, Insights, and Industry Studies, 4th ed. (Mason, OH:
Thomson South-Western, 2007).
25Rebecca Quick, “Jewelry Retailers Have Gem of a Holiday Season,” Wall Street Journal, January 7, 2000.
26Ann Zimmerman, “Retailers Wallow and See Only More Gloom,” Wall Street Journal, November 7, 2008.
M02_FARN0095_03_GE_C02.INDD 50 13/08/14 1:42 PM
ChAPter 2 Demand, Supply, and Equilibrium Prices 51
paid and moved toward smaller product sizes toward the end of the month when
their cash ran low. Wal-Mart customers also demanded a return of a Depression-
era strategy, layaway, which the company had cancelled in 2005. Target, which
attracts more affluent customers than Wal-Mart, found that its sales rebounded
more quickly to prerecession patterns than did Wal-Mart.27
Both increases in income and changes in tastes and preferences have resulted
in an increased demand for gourmet pet food, especially for dogs. The head of Del
Monte’s food and pet division said in 2006 that “the humanization of pets is the
single biggest trend driving our business.”28 Changes in tastes in human food spill
over into the pet food market. However, the demand for gourmet pet food was also
driven by the change in pet ownership from parents of small children, who had nei-
ther the time nor money to spend lavishly on their pets, to childless people ranging
from gay couples to parents whose children have left home. These couples have
larger incomes and treat their pets as they would their children.
Inferior Goods Firms producing inferior goods do not benefit from a boom-
ing economy. One such example is the pawnshop industry, which suffered during
the economic prosperity of the late 1990s and 2000, as fewer people swapped jew-
elry and other items for cash to cover car payments and other debts.29 Although
pawnshops have always suffered from a somewhat disreputable image, the strong
economy provided an income effect that further hurt the business and caused
many chains to incur large losses.
Dollar stores’ sales increased during the 2007 recession and moderated only
somewhat during the slow recovery. These stores experienced increases in the
number of customers who traded down out of economic necessity and who could
have gone elsewhere but were still exercising frugality.30 Payday lenders also
increased their business during the recession. Although these companies often
charged interest rates of more than 500 percent on their loans, they developed
strategies to lure customers away from traditional banks by appealing to people
with substandard credit records. The 22 payday loan offices in West Palm Beach,
Florida made $328.9 million in loans in fiscal year 2010, an increase of 119 percent
from fiscal year 2008.31
In the health care area, it is argued that tooth extractions are an example of an
inferior good. As individuals’ incomes rise, they are able to afford more complex
and expensive dental restorative procedures, such as caps and crowns, and they
are able to purchase more regular preventive dental services. Thus, the need for
extractions decreases as income increases.32
Prices of Related Goods There are two major categories of goods or prod-
ucts whose prices influence the demand for a particular good: substitute goods
and complementary goods.
Substitute Goods Products or services are substitute goods for each other
if one can be used in place of another. Consumers derive satisfaction from either
good or service. If two goods, X and Y, are substitutes for each other, an increase
in the price of good Y will cause consumers to decrease their consumption of
Substitute goods
Two goods, X and Y, are substitutes
if an increase in the price of good Y
causes consumers to increase their
demand for good X or if a decrease
in the price of good Y causes con-
sumers to decrease their demand
for good X.27Ann Zimmerman, “Frontier of Frugality,” Wall Street Journal (Online), October 4, 2011.
28Deborah Ball, “Nothing Says, ‘I Love You, Fido’ Like Food with Gourmet Flair,” Wall Street Journal,
March 18, 2006.
29Kortney Stringer, “Best of Times Is Worst of Times for Pawnshops in New Economy,” Wall Street Journal,
August 22, 2000.
30Zimmerman, “Frontier of Frugality.”
31Jessica Silver-Greenberg, “Payday Lenders go Hunting: Operations Encroach on Banks during Loan
Crunch; ‘Here, I Feel Respected,’ ” Wall Street Journal (Online), December 23, 2010.
32Rexford E. Santerre and Stephen P. Neun, Health Economics: Theories, Insights, and Industry Studies,
rev. ed. (Orlando, FL: Dryden, 2000), 90.
M02_FARN0095_03_GE_C02.INDD 51 13/08/14 1:42 PM
52 PArt 1 Microeconomic Analysis
good Y and increase their demand for good X. If the price of good Y decreases,
the demand for substitute good X will decrease. Thus, changes in the price of
good Y and the demand for good X move in the same direction for substitute
goods. The amount of substitution depends on the consumer’s tastes and prefer-
ences for the two goods and the size of the price change.
By 2006 the abundance and relatively low prices of cell phones, iPods, and laptop
computers resulted in many teens and young adults no longer purchasing wrist-
watches. In 2005, sales of watches priced between $30 and $150, the type most
often purchased by these age groups, declined more than 10 percent from 2004.33
In response to this threat from substitute products, watchmakers developed new
models that do much more than tell time, including watches with earbuds that play
digital music files, watches with programmable channels, and models with com-
passes and thermometers.
In 2007, large increases in the price of platinum resulted in an increased demand
for palladium, a lesser-known platinum-group metal. The price of an ounce of plat-
inum was approximately $1,190 compared with $337 for an ounce of palladium.
Because the two metals have a similar look and feel, many jewelers offered pal-
ladium to customers as a less expensive alternative, particularly for wedding and
engagement rings. World demand for palladium in jewelry was 1.12 million ounces
in 2006 compared with 1.74 million ounces for platinum.34
There are many substitutes for a given brand of bottled water, including both
other types of drinks and other brands of water. Customers bought less of Nestle’s
bottled water during the 2007 recession, due to both the loss of income and the
switch to the large number of cheaper private-label brands launched by supermar-
kets. Nestle responded by pushing Pure Life, a lower-priced water derived from
purified municipal sources.35
Complementary Goods Complementary goods are products or services
that consumers use together. If products X and Y are complements, an increase in
the price of good Y will cause consumers to decrease their consumption of good
Y and their demand for good X, since X and Y are used together. Likewise, if the
price of good Y decreases, the demand for good X will increase. Changes in the
price of good Y and the demand for good X move in the opposite direction if X
and Y are complementary goods.
As prices of personal computers have dropped over time, there has been an
increased demand for printers and printer cartridges. This complementary rela-
tionship has allowed Hewlett-Packard Company to actually sell its printers at a
loss that it recouped through its new ink and toner sales. Analysts estimated that in
2005 the company earned at least a 60 percent profit margin on both ink and toner
cartridges and two-thirds of the company’s profits were derived from these sales.
In 2006, Walgreen Company, the drugstore chain, announced plans for an ink-refill
service in 1,500 of its stores with a price at less than half the cost of buying new
cartridges.36 OfficeMax and Office Depot also offered these services. This example
shows how a complementary relationship between two goods can create a profit
opportunity for a firm, which then may still be competed away by the development
of substitute goods.
Future Expectations Expectations about future prices also play a role in
influencing current demand for a product. If consumers expect prices to be lower
in the future, they may have less current demand than if they did not have those
Complementary goods
Two goods, X and Y, are comple-
mentary if an increase in the price
of good Y causes consumers to
decrease their demand for good X
or if a decrease in the price of good
Y causes consumers to increase
their demand for good X.
33Jessica E. Vascellaro, “The Times They Are a-Changin’,” Wall Street Journal, January 18, 2006.
34Elizabeth Holmes, “Palladium, Platinum’s Cheaper Sister, Makes a Bid for Love,” Wall Street Journal,
February 13, 2007.
35Deborah Ball, “Bottled Water Pits Nestle vs. Greens,” Wall Street Journal (Online), May 25, 2010.
36Pui-Wing Tam, “A Cheaper Way to Refill Your Printer,” Wall Street Journal, January 26, 2006.
M02_FARN0095_03_GE_C02.INDD 52 13/08/14 1:42 PM
ChAPter 2 Demand, Supply, and Equilibrium Prices 53
expectations. In 2011, steel prices fell due to decreased demand arising from
unrest in the Middle East, the impact of Japan’s earthquake and tsunami, and
relatively high supply. Yet some buyers, including Moscow-based Central Steel
Co., held off on further purchases, given an expectation that prices would drop
another 2–5 percent in the following weeks. A U.K.-based steel consulting firm
noted that many Western European customers with adequate stockpiles were
also waiting on the sidelines for future price decreases.37
Likewise, if prices are expected to increase, consumers may demand more of the
good at present than they would without these expectations. In fall 2007, world
grain prices were surging from major demand increases stimulated by U.S. gov-
ernment incentives encouraging businesses to turn corn and soybeans into motor
fuel, increased incomes from the growing economies of Asia and Latin America,
and a growing middle class in these areas that was eating more meat and milk,
increasing the demand for grain to feed the livestock. Even though U.S. corn farm-
ers expected a record harvest, which should have had a moderating effect on grain
prices, traders in the futures markets for corn were already betting that the price
of corn would increase from $3.25 per bushel to more than $4.00 in March 2008 and
would stay above that level until 2010.38
Number of Consumers Finally, the number of consumers in the market-
place influences the demand for a product. A firm’s marketing strategy is typi-
cally based on finding new groups of consumers who will purchase the product.
In many cases, a country’s exports may be the source of this increased demand.
Although the U.S. timber industry continued to be depressed in 2011 from the
weakness in the U.S. housing market, exports to China surged, particularly from
mills in the Pacific Northwest. Russia increased tariffs on its exports to China
in 2007, so Chinese buyers turned to the United States and Canada to satisfy the
demand arising from that country’s construction boom. The number of U.S. logs
shipped to China increased more than 10 times between 2007 and 2010.39
The effect of growing populations on demand and grain prices was discussed
above in the “Future Expectations” section of the chapter. Both increases in the
size of the population in Asian and Latin American economies and growth in the
middle-class segments of these economies had a stimulating effect on the demand
for many types of grain.
Demand Function
We can now summarize all the variables that influence the demand for a particular
product in a generalized demand function represented as follows:
2.1 QXD ∙ f(PX, T, I, PY, PZ, EXC, NC, N)
where
QXD = quantity demanded of good X
PX = price of good X
T = variables representing an individual’s tastes and preferences
I = income
PY, PZ = prices of goods Y and Z, which are related to the
consumption of good X
EXC = consumer expectations about future prices
NC = number of consumers
37Robert Guy Matthews, “Steel Price Softens as Supply Solidifies,” Wall Street Journal (Online), April 10, 2011.
38Scott Kilman, “Historic Surge in Grain Prices Roils Market,” Wall Street Journal, September 28, 2007.
39Jim Carlton, “Chinese Demand Lifts U.S. Wood Sales,” Wall Street Journal (Online), February 8, 2011.
M02_FARN0095_03_GE_C02.INDD 53 13/08/14 1:42 PM
54 PArt 1 Microeconomic Analysis
Equation 2.1 is read as follows: The quantity demanded of good X is a function (f)
of the variables inside the parentheses. An ellipsis is placed after the last variable
to signify that many other variables may also influence the demand for a specific
product. These may include variables under the control of a manager, such as the
size of the advertising budget, and variables not under anyone’s control, such as the
weather.
Each consumer has his or her own individual demand function for different
products. However, managers are usually more interested in the market demand
function, which shows the quantity demanded of the good or service by all con-
sumers in the market at any given price. The market demand function is influenced
by the prices of related goods, as well as by the tastes and preferences, income,
and future expectations of all consumers in the market. It can also change because
more consumers enter the market.
Demand Curves
Equation 2.1 shows the typical variables included in a demand function. To system-
atically analyze all of these variables, economists define demand as we did earlier
in this chapter: the functional relationship between alternative prices and the quan-
tities consumers demand at those prices, all else held constant. This relationship is
portrayed graphically in Figure 2.1, which shows a demand curve for a given prod-
uct. Price (P), measured in dollar terms, is the variable that is explicitly analyzed
and shown on the vertical axis of the graph. Quantity demanded (Q) is shown on
the horizontal axis. The other variables in the demand function are held constant
with a given demand curve, but act as demand shifters if their values change.
As we just mentioned, demand curves are drawn with the price placed on the
vertical axis and the quantity demanded on the horizontal axis. This may seem
inconsistent because we usually think of the quantity demanded of a good (depen-
dent variable) as a function of the price of the good (independent variable). The
dependent variable in a mathematical relationship is usually placed on the vertical
axis and the independent variable on the horizontal axis. The reverse is done for
demand because we also want to show how revenues and costs vary with the level
of output. These variables are placed on the vertical axis in subsequent analysis. In
mathematical terms, an equation showing quantity as a function of price is equiva-
lent to the inverse equation showing price as a function of quantity.
Demand curves are generally downward sloping, showing a negative or inverse
relationship between the price of a good and the quantity demanded at that price,
all else held constant. Thus, in Figure 2.1, when the price falls from P1 to P2, the
quantity demanded is expected to increase from Q1 to Q2, if nothing else changes.
This is represented by the movement from point A to point B in Figure 2.1. Likewise,
an increase in the price of the good results in a decrease in quantity demanded, all
else held constant. Most demand curves that show real-world behavior exhibit this
Individual demand
function
The function that shows, in
symbolic or mathematical terms,
the variables that influence the
quantity demanded of a particular
product by an individual consumer.
Market demand function
The function that shows, in sym-
bolic or mathematical terms, the
variables that influence the quantity
demanded of a particular product
by all consumers in the market and
that is thus affected by the number
of consumers in the market.
Demand curve
The graphical relationship between
the price of a good and the quantity
consumers demand, with all other
factors influencing demand held
constant.
Demand shifters
The variables in a demand function
that are held constant when defin-
ing a given demand curve, but that
would shift the demand curve if
their values changed.
Negative (inverse)
relationship
A relationship between two
variables, graphed as a downward
sloping line, where an increase in
the value of one variable causes a
decrease in the value of the other
variable.
A
BP2
P1
Q1 Q2
Demand
0 Q
PFIgure 2.1
the Demand Curve for a Product
A demand curve shows the
relationship between the price of a
good and the quantity demanded,
all else held constant.
M02_FARN0095_03_GE_C02.INDD 54 13/08/14 1:42 PM
ChAPter 2 Demand, Supply, and Equilibrium Prices 55
inverse relationship between price and quantity demanded. (We’ll later discuss the
economic model of consumer behavior that lies behind this demand relationship.)
(Chapter 3 appendix.)
Change in Quantity Demanded and Change in Demand
The movement between points A and B along the demand curve in Figure 2.1 is
called a change in quantity demanded. It results when consumers react to a
change in the price of the good, all other factors held constant. This change in
quantity demanded is pictured as a movement along a given demand curve.
It is also possible for the entire demand curve to shift. This shift results when the
values of one or more of the other variables in Equation 2.1 change. For example, if
consumers’ incomes increase, the demand curve for the particular good generally
shifts outward or to the right, assuming that the good is a normal good. This shift
of the entire demand curve is called a change in demand. It occurs when one or
more of the variables held constant in defining a given demand curve changes.
This distinction between a change in demand and a change in quantity demanded
is very important in economic analysis. The two phrases mean something differ-
ent and should not be used interchangeably. The distinction arises from the basic
economic framework, in which we examine the relationship between two variables
while holding all other factors constant.
An increase in demand, or a rightward or outward shift of the demand curve, is
shown in Figure 2.2. We’ve drawn this shift as a parallel shift of the demand curve,
although this doesn’t have to be the case. Suppose this change in demand results
from an increase in consumers’ incomes. The important point in Figure 2.2 is that
an increase in demand means that consumers will demand a larger quantity of the
good at the same price—in this case, due to higher incomes. This outcome is con-
trasted with a movement along a demand curve or a change in quantity demanded,
where a larger quantity of the good is demanded only at a lower price. This distinc-
tion can help you differentiate between the two cases.
Changes in any of the variables in a demand function, other than the price of
the product, will cause a shift of the demand curve in one direction or the other.
Thus, the relationship between quantity demanded and the first variable on the
right side of Equation 2.1 (price) determines the slope of the curve (downward
sloping), while the other right-hand variables cause the curve to shift. In Figure 2.2,
we assumed that the good was a normal good so that an increase in income would
result in an increase in demand, or a rightward shift of the demand curve. If the
good was an inferior good, this increase in income would result in a decrease in
demand, or a leftward shift of the curve. An increase in the price of a substitute
good would cause the demand curve for the good in question to shift rightward,
while an increase in the price of a complementary good would cause a leftward
Change in quantity
demanded
The change in quantity consumers
purchase when the price of the
good changes, all other factors held
constant, pictured as a movement
along a given demand curve.
Change in demand
The change in quantity purchased
when one or more of the demand
shifters change, pictured as a shift
of the entire demand curve.
P1
Q1 Q2
D1
D2
0 Q
P FIgure 2.2
Change (Increase) in Demand
A change in demand occurs when
one or more of the factors held
constant in defining a given
demand curve changes.
M02_FARN0095_03_GE_C02.INDD 55 13/08/14 1:42 PM
56 PArt 1 Microeconomic Analysis
shift of the demand curve. A change in consumer expectations could also cause
the curve to shift in either direction, depending on whether a price increase or
decrease was expected. If future prices were expected to rise, the current demand
curve would shift outward or to the right. The opposite would happen if future
prices were expected to decrease. An increase in the number of consumers in the
market would cause the demand curve to shift to the right, while the opposite
would happen for a decrease in the number of consumers.
Individual Versus Market Demand Curves
The shift in the market demand curve as more individuals enter the market is illus-
trated in Figure 2.3, which shows how a market demand curve is derived from indi-
vidual demand curves. In this figure, demand curve DA represents the demand for
individual A. If individual A is the only person in the market, this demand curve
is also the market demand curve. However, if individual B enters the market with
demand curve DB, then we have to construct a new market demand curve. As shown
in Figure 2.3, individual B has a larger demand for the product than individual A.
The demand curve for B lies to the right of the demand curve for A, indicating that
individual B will demand a larger quantity of the product at every price level.
To derive the market demand curve for both individuals, we do a horizontal
summation of individual demand curves. This means that for every price we
add the quantity that each person demands at that price to determine the market
quantity demanded at that price. At prices above P1, only individual B is in the
market, so demand curve B is the market demand curve in that price range. Below
price P1, we need to add together the quantities that each individual demands.
For example, at a zero price, individual A demands quantity Q2, and individual B
demands quantity Q3, so the quantity demanded by the market (both individuals)
at the zero price is Q4, which equals Q2 plus Q3. The market demand curve, DM, is
derived in the same manner by adding the quantities demanded at other prices.
Based on the information in Figure 2.3, we can infer that if another individual,
C, came into the market, the market demand curve would shift further to the right.
Thus, a market demand curve can shift as individuals enter or leave a market.
Linear Demand Functions and Curves
The demand curves in Figures 2.1, 2.2, and 2.3 have been drawn as straight lines, rep-
resenting a linear demand function. A linear demand function is a specific math-
ematical relationship of the generalized demand function (Equation 2.1) in which
all terms are either added or subtracted and there are no exponents in any terms
that take a value other than 1. The graph of a linear demand function has a constant
slope. This linear relationship is used both because it simplifies the analysis and
horizontal summation of
individual demand curves
The process of deriving a market
demand curve by adding the quan-
tity demanded by each individual at
every price to determine the market
demand at every price.
Linear demand function
A mathematical demand function
graphed as a straight-line demand
curve in which all the terms are
either added or subtracted and no
terms have exponents other than 1.
DA
DB
P1
P
Q1 Q2 Q3
DM = DA + DB
Q4 Q0
FIgure 2.3
Individual Versus Market
Demand Curve
A market demand curve is derived
from the horizontal summation of
individual demand curves; that is, for
every price, add the quantity each
individual demands at that price
to determine the market quantity
demanded at that price.
M02_FARN0095_03_GE_C02.INDD 56 13/08/14 1:42 PM
ChAPter 2 Demand, Supply, and Equilibrium Prices 57
because many economists believe that this form of demand function best repre-
sents individuals’ behavior, at least within a given range of prices. However, not all
demand functions are linear. We will later discuss the implications of a particular
type of demand function for consumer behavior in greater detail (Chapter 3).
Mathematical Example of a Demand Function
Although we have been discussing demand functions and demand curves in ver-
bal, symbolic, and graphical terms, these relationships can also be expressed as a
mathematical equation. In this section, we begin a hypothetical numerical example
based on the copper industry articles that we have used throughout the chapter.
For simplicity, we assume that our demand and supply functions are both linear.
Suppose that the demand function for copper at the beginning of 2010 is repre-
sented by Equation 2.2:
2.2 QD ∙ 3 ∙ 2Pc ∙ 0.2I ∙ 1.6TC ∙ 0.04E
where
QD = quantity demanded of copper (millions of pounds)
PC = price of copper ($ per pound)
I = consumer income index
TC = telecom index showing uses or tastes for copper in the
telecommunications industry
E = expectations index representing purchasers’ expectations of
a lower price over the following six months
We assume here that the quantity demanded of copper is a function only of PC, the
price of copper; I, consumer income; TC, the telecom index; and E, the expecta-
tions index. An economist or market analyst would develop this model of demand
and derive the actual values of the constant term and coefficients of the variables
in Equation 2.2 from real-world data using various empirical methods (Chapter 4).
The negative coefficient on the PC variable shows the inverse relationship between
price and quantity demanded of copper. If the price of copper increases, the quantity
demanded decreases. This represents a typical downward sloping demand curve.
We can see from this demand function that copper is a normal good because the
income variable, I, in Equation 2.2 has a positive coefficient. Increases in income
result in increases in the demand for copper. The positive coefficient on the TC vari-
able means that as improved technology and higher demand for telecom services
in North America and Europe create more uses for copper in the telecommunica-
tions industry, the overall demand for copper increases. The expectations index, E,
represents consumers’ expectations of a lower price over the following six months,
where a lower index number implies that more purchasers expect a lower price. This
expectation decreases the current demand for copper. Equation 2.2 is a mathemati-
cal representation of the conceptual relationships developed earlier in the chapter.
To define a specific demand curve for copper, we need to hold constant the level
of consumer income, the telecom index, and the expectations index. Suppose that
I = 20, TC = 2.5, and E = 100. Substituting 20 for I, 2.5 for TC, and 100 for E in
Equation 2.2 gives us Equation 2.3:
2.3 QD ∙ 3 ∙ 2PC ∙ 0.2(20) ∙ 1.6(2.5) ∙ 0.04(100)
or
QD ∙ 15 ∙ 2PC or [PC ∙ 7.5 ∙ 0.5QD]
We can clearly see the meaning of the expression all else held constant in Equation
2.3. In that equation, the effects of consumer income, the telecom index, and the
M02_FARN0095_03_GE_C02.INDD 57 13/08/14 1:42 PM
58 PArt 1 Microeconomic Analysis
expectations index are embodied in the constant term 15. If we change the values of
any of these three variables, the constant term in Equation 2.3 changes, and we have
a change in demand or a new demand equation that graphs into a different demand
curve. A change in quantity demanded in Equation 2.3 is represented by substitut-
ing different values for the price of copper and calculating the resulting quantity
demanded at those prices. Equation 2.3 also shows the inverse demand function,
with price as a function of quantity. These equations are equivalent mathematically.
Managerial rule of thumb
Demand Considerations
Managers need to understand the factors that influence consumer demand for their products.
Although product price is usually important, other factors may play a significant role. In developing
a competitive strategy, managers need to determine which factors they can influence and how to
handle the factors that are beyond their control. ■
Supply
We now examine producer decisions to supply various goods and services and the
factors influencing those decisions. Supply is the functional relationship between
the price of a good or service and the quantity that producers are willing and able
to supply in a given time period, all else held constant.
Nonprice Factors Influencing Supply
Although supply focuses on the influence of price on the quantity of a good or
service supplied, many other factors influence producer supply decisions. These
factors generally relate to the cost of production.
State of Technology The state of technology, or the body of knowledge
about how to combine the inputs of production, affects what output producers
will supply because technology influences how the good or service is actually
produced, which, in turn, affects the costs of production. For example, the dis-
cussion of the copper industry noted that a change in mining technology allowed
companies to produce copper at a lower cost, keeping more of them in business.
This change in technology contributed to a decrease in mining costs of 30 percent
between the 1980s and the 1990s.40
In the nickel industry, most of the world’s production has come from deposits
that were relatively easy to exploit. However, these deposits comprise only about
40 percent or less of the world’s remaining reserves. During the 1990s companies
tried to develop a process called “high pressure acid leaching” to remove nickel
from other rock deposits. The hope was that this new technology would open large
new deposits of nickel. Although the initial equipment failed to stand up to the
extreme heat and pressure of the process, more recent changes in the technology
have increased its reliability and usefulness.41
Pecan growers were able to respond to the increased demand of the nut described
earlier in the chapter, given the development of machines to shake the nuts from
Supply
The functional relationship
between the price of a good
or service and the quantity
supplied by producers in a given
time period, all else held constant.
40Lucchetti, “Copper Limbo: Just How Low Can It Go?”
41Patrick Barta, “With Easy Nickel Fading Fast, Miners Go After the Tough Stuff,” Wall Street Journal,
July 12, 2006.
M02_FARN0095_03_GE_C02.INDD 58 13/08/14 1:42 PM
ChAPter 2 Demand, Supply, and Equilibrium Prices 59
the trees and sweep them off the ground. In the shelling plants, upgraded technol-
ogy resulted in the use of machines to wash the nuts in hot water to kill bacteria,
crack and separate the meat from the shells, separate halves from smaller pieces,
skim out any worms, and roast, chop, and sort the pieces by size and color.42
Input Prices Input prices are the prices of all the inputs or factors of
production—labor, capital, land, and raw materials—used to produce the given
product. These input prices affect the costs of production and, therefore, the prices
at which producers are willing to supply different amounts of output. For broiler
chickens, feed costs represent 70–75 percent of the costs of growing a chicken to
a marketable size. Thus, changes in feed costs are so important that market ana-
lysts often use them as a proxy to forecast broiler prices and returns to broiler
processors.43
Although the higher prices for pecans noted earlier in the chapter benefited grow-
ers, they represented increased costs for bakers and ice cream makers. Pecans are
a key ingredient in fruitcakes, accounting for 27 percent of the weight in many
varieties. Bakers worried that consumers would react to the higher prices of their
products resulting from these increased costs of production.44 One of the factors
influencing the 2011 decrease in steel prices noted earlier in the chapter was falling
prices for major raw ingredients in steel production. Prices for iron ore decreased
9.5 percent from February to March 2011.45
Prices of Goods Related in Production The prices of other goods related
in production can also affect the supply of a particular good. Two goods are sub-
stitutes in production if the same inputs can be used to produce either of the
goods, such as land for different agricultural crops. Between 2005 and 2007, U.S.
tobacco acreage increased 20 percent with tobacco being planted in areas such
as southern Illinois that had not grown any substantial amount since the end of
World War I. Even though corn prices were at near-record levels of $4.00 per
bushel during this period, they were not high enough to compete with tobacco
planting. Even with higher labor and other costs, one farmer in Illinois estimated
that he netted $1,800 per acre from his 150 acres of tobacco compared with $250
per acre for corn and that planting tobacco had increased his annual income by
35 percent over the previous three years.46
Companies use the same type of rigs to drill for oil and natural gas. Therefore,
they allocate equipment according to the price and profitability of each fuel.
Given that the price of oil increased significantly between 2010 and 2011 from
unrest in Northern Africa and the Middle East, the number of land rigs in the U.S.
drilling for natural gas decreased 8 percent while oil rigs increased 81 percent.
One major gas producer spent 90 percent of its $5 billion budget on oil drilling.47
There can also be complementary production relationships between com-
modities. As more oil and natural gas are produced, the supply of sulfur, which
is removed from the products, also increases. Sixty-foot-high blocks of unwanted
sulfur were reported in Alberta, Canada, and Kazakhstan in 2003.48 Likewise, if the
demand for and price of white-meat chicken increases, there will be an increase in
the supply of dark-meat chicken.
42David Wessel, “Shell Shock: Chinese Demand Reshapes U.S. Pecan Business.”
43Richard T. Rogers, “Broilers: Differentiating a Commodity,” in Industry Studies, 2nd ed., ed. Larry L.
Duetsch (Armonk, NY: Shapiro, 1998), 71.
44David Wessel, “Shell Shock: Chinese Demand Reshapes U.S. Pecan Business.”
45Robert Guy Matthews, “Steel Price Softens as Supply Solidifies.”
46Lauren Etter, “U.S. Farmers Rediscover the Allure of Tobacco,” Wall Street Journal, September 18, 2007.
47Daniel Gilbert, “As Natural Gas Prices Fall, the Search Turns to Oil,” Wall Street Journal (Online), May 23,
2011.
48Alexei Barrionuevo, “A Chip off the Block Is Going to Smell Like Rotten Eggs,” Wall Street Journal,
November 4, 2003.
M02_FARN0095_03_GE_C02.INDD 59 13/08/14 1:42 PM
60 PArt 1 Microeconomic Analysis
Future Expectations Future expectations can play a role on the supply side
of the market as well. If producers expect prices to increase in the future, they may
supply less output now than without those expectations. The opposite could hap-
pen if producers expect prices to decrease in the future. These expectations could
become self-fulfilling prophecies. Smaller current supplies in the first case could
drive prices up, while larger current supplies in the second case could result in lower
prices. Expectations may not always be correct. Given the high demand and lumber
prices in summer 2004, lumber manufacturers expected that demand would start to
drop as interest rates rose. When this did not happen, prices continued to climb.49
Number of Producers Finally, the number of producers influences the total
supply of a product at any given price. The number of producers may increase
because of perceived profitability in a given industry or because of changes in
laws or regulations such as trade barriers. For example, the lumber market was
reported to be exceedingly strong in January 1999, largely due to demand from
the booming U.S. housing market. However, quotas on the amount of wood that
Canada could ship into the United States also played a role in keeping the price of
lumber high in the United States in January of that year.50
Similarly, in November 2004, tariffs on 115 Chinese producers of wooden bed-
room furniture were lowered from 12.9 to 8.6 percent. Because these companies
accounted for 65 percent of the bedroom furniture imported to the United States
from China, the resulting increased supply lowered prices for consumers and put
more competitive pressure on U.S. furniture makers who had already closed doz-
ens of factories in North Carolina and Virginia in the previous four years.51
In 2010, the Canadian lumber industry was revived by increased demand from
China. Numerous producers reopened recently shuttered mills and called back
workers to respond to the increased Chinese demand. Owners estimated that one
new mill of Western Forest Products, Inc., which was about to start production,
would send half of its output to China.52
Supply Function
A supply function for a product, which is defined in a manner similar to a demand
function, is shown in Equation 2.4:
2.4 QXS ∙ f (PX, TX, PI, PA, PB, EXP, NP, N)
where
QXS = quantity supplied of good X
PX = price of good X
TX = state of technology
PI = prices of the inputs of production
PA, PB = prices of goods A and B, which are related in production to
good X
EXP = producer expectations about future prices
NP = number of producers
Equation 2.4 shows that the quantity supplied of good X depends on the price of
good X, the other variables listed above, and possibly variables peculiar to the firm
49Avery Johnson, “Sticker Shock at the Lumberyard,” Wall Street Journal, August 11, 2004.
50Terzah Ewing, “Lumber’s Strength Defies Bearish Trend,” Wall Street Journal, January 26, 1999.
51Dan Morse, “U.S. Cuts Tariffs on Imports of China’s Bedroom Furniture,” Wall Street Journal, November
10, 2004.
52Joel Millman, “Canada’s Mills Lumber Back to Life, Fueled by Chinese,” Wall Street Journal (Online),
November 2, 2010.
M02_FARN0095_03_GE_C02.INDD 60 13/08/14 1:42 PM
ChAPter 2 Demand, Supply, and Equilibrium Prices 61
or industry that are not included in the list, as indicated by the ellipsis. As with the
demand function, we can distinguish between an individual supply function and a
market supply function. The individual supply function shows, in symbolic or
mathematical terms, the variables that influence an individual producer’s supply
of a product. The market supply function shows the variables that influence the
overall supply of a product by all producers and is thus affected by the number of
producers in the market.
Supply Curves
We graph a supply curve in Figure 2.4, showing price (P) on the vertical axis and
quantity supplied (Q) on the horizontal axis. For simplicity, all supply curves in
this chapter will represent market supply functions. The supply curve in Figure 2.4
shows the relationship between price and quantity supplied, holding constant all
the other variables influencing the supply decision (all variables beside PX on the
right side of Equation 2.4). Changes in these variables, the supply shifters, will
cause the supply curve to shift.
As you can see in Figure 2.4, a supply curve generally slopes upward, indicating a
positive or direct relationship between the price of the product and the quantity
producers are willing to supply. A higher price typically gives producers an incen-
tive to increase the quantity supplied of a particular product because higher produc-
tion is more profitable. The supply curve in Figure 2.4 represents a linear supply
function and is graphed as a straight line. Not all supply functions are linear, but we
will use this type of function for simplicity. Keep in mind that a supply curve does
not show the actual price of the product, only a functional relationship between
alternative prices and the quantities that producers want to supply at those prices.
Change in Quantity Supplied and Change in Supply
Figure 2.4 shows a given supply curve defined with all other factors held constant.
If the price increases from P1 to P2, the quantity supplied increases from Q1 to Q2.
This movement from point A to point B represents a movement along the given
supply curve, or a change in quantity supplied. Some factor has caused the price
of the product to increase, and suppliers respond by increasing the quantity sup-
plied. This supply response is by the existing suppliers, since the number of suppli-
ers is held constant when defining any given supply curve.
Figure 2.5 shows a shift of the entire supply curve. This represents a change in
supply, not a change in quantity supplied. The supply curve shifts from S1 to S2
because one or more of the factors from Equation 2.4 held constant in supply curve
S1 changes. The increase in supply, or the rightward shift of the supply curve in
Figure 2.5, shows that producers are willing to supply a larger quantity of output
at any given price. Thus, the quantity supplied at price P1 increases from Q1 to Q2.
This differs from the movement along a supply curve, or a change in quantity sup-
plied, shown in Figure 2.4, where the increase in quantity supplied is associated
Individual supply function
The function that shows, in sym-
bolic or mathematical terms, the
variables that influence the quan-
tity supplied of a particular product
by an individual producer.
Market supply function
The function that shows, in sym-
bolic or mathematical terms, the
variables that influence the quan-
tity supplied of a particular product
by all producers in the market and
that is thus affected by the number
of producers in the market.
Supply curve
The graphical relationship between
the price of a good and the quantity
supplied, with all other factors
influencing supply held constant.
Supply shifters
The other variables in a supply
function that are held constant
when defining a given supply
curve, but that would cause that
supply curve to shift if their values
changed.
Positive (direct)
relationship
A relationship between two
variables, graphed as an upward
sloping line, where an increase in
the value of one variable causes
an increase in the value of the
other variable.
Linear supply function
A mathematical supply function,
which graphs as a straight-line
supply curve, in which all terms are
either added or subtracted and no
terms have exponents other than 1.
Change in quantity
supplied
The change in amount of a good
supplied when the price of the
good changes, all other factors held
constant, pictured as a movement
along a given supply curve.
P1
P2
Q1 Q2
A
B
Supply
0 Q
P FIgure 2.4
the Supply Curve for a Product
A supply curve shows the relation-
ship between the price of a good and
the quantity supplied, all else held
constant.
M02_FARN0095_03_GE_C02.INDD 61 13/08/14 1:42 PM
62 PArt 1 Microeconomic Analysis
with a higher price for the product. This distinction between a change in quantity
supplied and a change in supply is analogous to the distinction between a change
in quantity demanded and a change in demand. We use the same framework—the
relationship between two variables (price and quantity), all else held constant—on
both the demand and the supply sides of the market.
Developing new technology typically causes an increase in supply, or a rightward
shift of the supply curve, because technology changes usually lower the costs of
production. The same result holds for a decrease in the price of any of the inputs of
production, which lowers the costs of production and causes the supply curve to
shift to the right. Any increase in the price of inputs increases the costs of produc-
tion and causes the supply curve of the product to shift to the left.
The effect of a change in the price of a related good on the supply of a given good
depends on whether the related good is a substitute or complement in production.
An increase in the price of a substitute good causes the supply curve for the given
good to shift to the left. A decrease in the price of a substitute good causes an
increase in the supply of the given good. The opposite set of relationships holds for
goods that are complements in production. If the price of the complementary good
increases, the supply of the given good increases.
Producer expectations of lower prices cause the supply curve of a good to shift
to the right. The supply increases in anticipation of lower prices in the future. The
opposite holds if producers expect prices to increase. There would be a smaller
current supply than without those expectations.
Finally, an increase in the number of producers results in a rightward shift of
the supply curve, while a decrease results in a leftward shift of the supply curve.
A given supply curve shows how prices induce the current number of producers to
change the quantity supplied. Any change in the number of producers in the market
is represented by a shift of the entire curve.
Mathematical Example of a Supply Function
To continue the mathematical example we began in the demand section, we assume
that the supply function for copper is represented by Equation 2.5. (Note that real-
world supply functions are empirically estimated from data in different firms and
industries.)
2.5 QS ∙ ∙5 ∙ 8PC ∙0.5W ∙ 0.4T ∙ 0.5N
where
QS = quantity supplied of copper (millions of pounds)
PC = price of copper ($ per pound)
W = an index of wage rates in the copper industry
T = technology index
N = number of active mines in the copper industry
Change in supply
The change in the amount of a
good supplied when one or more of
the supply shifters change, pictured
as a shift of the entire supply curve.
P1
Q1 Q2
S1
S2
0 Q
PFIgure 2.5
Change (Increase) in Supply
A change in supply occurs when one
or more of the factors held constant
in defining a given supply curve
changes.
M02_FARN0095_03_GE_C02.INDD 62 13/08/14 1:42 PM
ChAPter 2 Demand, Supply, and Equilibrium Prices 63
In Equation 2.5, we assume that the quantity supplied of copper is a function only
of the price of copper, wage rates in the copper industry (the price of an input of
production), the technology index, and the number of firms in the industry. The
positive coefficient on the PC variable shows the positive relationship between
the price of copper and the quantity supplied. A higher price will elicit a larger
quantity supplied. This relationship represents a normal, upward sloping supply
curve. The other variables in Equation 2.5 cause the supply curve to shift. The
wage rate index, W, has a negative coefficient. As wage rates increase, the supply
of copper decreases because an increase in this input price represents an increase
in the costs of production. The technology index (T) and the number of active
mines variable (N) both have positive coefficients, indicating that an increase in
technology or in the number of active mines will increase the supply of copper.
To define a specific supply curve, we need to hold constant the wage rate index,
the technology index, and the number of firms in the copper industry. Suppose
that W = 100, T = 50, and N = 20. Substituting these values into Equation 2.5 gives
Equation 2.6.
2.6 QS ∙ ∙5 ∙ 8PC ∙0.5(100) ∙ 0.4(50) ∙ 0.5(20)
or
QS ∙ ∙25 ∙ 8PC or [PC ∙ 3.125 ∙ 0.125QS]
As with the demand curve in Equation 2.3, the supply curve in Equation 2.6 shows
the relationship between the price of copper and the quantity supplied, all else
held constant. The constant term, –25, incorporates the effect of the wage and
technology indices and the number of firms in the industry. Any changes in these
variables change the size of the constant term, which results in a different supply
curve.53
Summary of Demand and Supply Factors
Table 2.1 summarizes the factors influencing both the demand and the supply sides
of the market. Notice the symmetry in that some of the factors—including the
prices of related goods, future expectations, and the number of participants—influ-
ence both sides of the market.
53If we rewrite the supply equation with price as a function of quantity supplied, we get P = 3.125 + 0.125QS,
as shown in Equation 2.6. This equation implies that producers must receive a price of at least $3.125 per
pound to induce them to supply any copper.
tAbLe 2.1 Factors Influencing Market Demand and Supply
DeMAND SuPPLY
Price of the product Price of the product
Consumer tastes and preferences State of technology
Consumer income: Input prices
Normal goods
Inferior goods
Price of goods related in consumption: Prices of goods related in production
Substitute goods Substitute goods
Complementary goods Complementary goods
Future expectations Future expectations
Number of consumers Number of producers
M02_FARN0095_03_GE_C02.INDD 63 13/08/14 1:42 PM
64 PArt 1 Microeconomic Analysis
Managerial rule of thumb
Supply Considerations
In developing a competitive strategy, managers must examine the technology and costs of produc-
tion, factors that influence the supply of the product. Finding ways to increase productivity and lower
production costs is particularly important in gaining a strategic advantage in a competitive market
where managers have little control over price. ■
Demand, Supply, and Equilibrium
As we discussed earlier in the chapter, demand and supply are both functional
relationships between the price of a good and the quantity demanded or supplied.
Neither function by itself tells us what price will actually exist in the market. That
price will be determined when the market is in equilibrium.
Definition of Equilibrium Price and Equilibrium Quantity
In a competitive market, the interaction of demand and supply determines the
equilibrium price, the price that will actually exist in the market or toward which
the market is moving. Figure 2.6 shows the equilibrium price (PE) for good X. The
equilibrium price is the price at which the quantity demanded of good X by consum-
ers equals the quantity that producers are willing to supply. This quantity is called
the equilibrium quantity (QE). At any other price, there will be an imbalance
between quantity demanded and quantity supplied. Forces will be set in motion to
push the price back toward equilibrium, assuming no market impediments or gov-
ernmental policies exist that would prevent equilibrium from being reached.
Lower-Than-Equilibrium Prices
The best way to understand equilibrium is to consider what would happen if some
price other than the equilibrium price actually existed in a market. Suppose P1 is
the actual market price in Figure 2.7. As you see in the figure, price P1 is lower
than the equilibrium price, PE. You can also see that the quantity of the good
demanded by consumers at price P1 is greater than the quantity producers are
willing to supply. This creates a shortage of the good, shown in Figure 2.7 as the
amount of the good between QD and QS. At the lower-than-equilibrium price, P1,
consumers demand more of the good than producers are willing to supply at that
equilibrium price
The price that actually exists in the
market or toward which the market
is moving where the quantity
demanded by consumers equals
the quantity supplied by producers.
equilibrium quantity (QE)
The quantity of a good, determined
by the equilibrium price, where the
amount of output that consumers
demand is equal to the amount that
producers want to supply.
PE
QE
S
D
Q
P
0
FIgure 2.6
Market equilibrium
Market equilibrium occurs at that
price where the quantity demanded
by consumers equals the quantity
supplied by producers.
M02_FARN0095_03_GE_C02.INDD 64 13/08/14 1:42 PM
ChAPter 2 Demand, Supply, and Equilibrium Prices 65
price. Because there is an imbalance between quantity demanded and quantity
supplied at this price, the situation is not stable. Some individuals are willing to
pay more than price P1, so they will start to bid the price up. A higher price will
cause producers to supply a larger quantity. This adjustment process will continue
until the equilibrium price has been reached and quantity demanded is equal to
quantity supplied.
Price and rent controls are examples of the imbalances between demand and
supply that result from lower-than-equilibrium prices. In New York City, which
used rent controls for many years for certain apartments, the excess demand for
these apartments meant that many of them never actually appeared on the mar-
ket. They were either kept by the current occupants or transferred to those with
connections to the occupants.54 Rent controls often led to decreased maintenance
in the controlled units and increased rents in the noncontrolled sector.55 You can
also observe lower-than-equilibrium prices being charged for tickets to the Super
Bowl and many other sporting and entertainment events where scalpers sell tickets
for prices far exceeding the stated price. The quantity demanded of tickets at the
stated price is greater than the quantity supplied at that price, so people will pay
much more than the stated price for these tickets. Recognizing this excess demand,
the producers of the hit Broadway show The Producers began, in October 2001,
setting aside at least 50 seats at every performance to sell at $480 per ticket, a price
far exceeding the top regular charge of $100. This was a strategic move to tap into
the excess demand and ensure that the creators of the play, and not the scalpers,
received a bigger share of the royalties.56 This trend increased both on Broadway
and with other performing arts organizations over the next decade.57
Notre Dame University conducts a lottery every year to parcel out the 30,000
seats available to contributors, former athletes, and parents for football games in
its 80,000-seat stadium. In September 2006, the university had 66,670 ticket requests
for the 30,000 seats for the Notre Dame–Penn State game. Individuals and busi-
nesses take advantage of the excess demand during game weekends. Houses for
visitors rent for $3,000 or more, motel rooms normally renting for $129 per night
sell out at $400, parking passes are sold on eBay for $500, and $59 tickets can fetch
a price of $1,600.58
P1
PE
QEQS QD
S
D
Shortage
0 Q
P FIgure 2.7
A Lower-than-equilibrium Price
A shortage of a good results when the
market price, P1, is below the equilib-
rium price, PE.
54Richard Arnott, “Time for Revisionism on Rent Control?” Journal of Economic Perspectives 9(1) (Winter
1995): 99–120.
55Blair Jenkins, “Rent Control: Do Economists Agree?” Econ Journal Watch 6(1) (January 2009): 73–112.
56Jessie McKinley, “For the Asking, $480 a Seat for ‘The Producers,’ ” New York Times, October 26, 2001.
57Patrick Healy, “Broadway Hits Make Most of Premium Pricing,” The New York Times (Online), November
24, 2011.
58Ilan Brat, “Notre Dame Football Introduces Its Fans to Inflationary Spiral,” Wall Street Journal,
September 6, 2006.
M02_FARN0095_03_GE_C02.INDD 65 13/08/14 1:42 PM
66 PArt 1 Microeconomic Analysis
Higher-Than-Equilibrium Prices
Figure 2.8 shows the opposite case, a higher-than-equilibrium price. At price P2,
the quantity supplied, QS, is greater than the quantity demanded, QD, at that price.
This above-equilibrium price creates a surplus of the good and sets into motion
forces that will cause the price to fall. As the price falls, the quantity demanded
increases and the quantity supplied decreases until a balance between quantity
demanded and quantity supplied is restored at the equilibrium price. Thus, the exis-
tence of either shortages or surpluses of goods is an indication that a market is not
in equilibrium.
In this chapter, we have used several agricultural examples to illustrate the forces
shifting demand and supply curves because agricultural markets exhibit many
competitive characteristics. These markets also provide good examples of non-
equilibrium prices and imbalances between demand and supply, given the exten-
sive government agricultural subsidization programs that have been in operation
over the years. These crop subsidy programs kept the price of many agricultural
products above equilibrium, which resulted in an excess quantity supplied com-
pared with quantity demanded. During the 1990s, the U.S. government began elimi-
nating or cutting back many of these subsidy programs. As prices for their crops
have fallen, many farmers have gone out of business.
The choices among crops to be planted are often influenced by the pattern of
federal subsidies. The Wall Street Journal reported in April 1999 that U.S. farm-
ers intended to plant a record 73.1 million acres of soybeans that spring, even in
the face of declining prices for this product.59 Although this move was likely to
cause soybean prices to fall even further, farmers were responding to a soybean
subsidy that was higher than those for other crops. This increased level of planting
by U.S. farmers, combined with large harvests from other countries, was expected
to push soybean prices to under $4 a bushel, the lowest level since the 1980s.
However, under a U.S. Department of Agriculture marketing-loan program, U.S.
farmers could expect a price of $5.26 per bushel of soybeans. In response, they
were expected to produce 2.9 billion bushels of soybeans, up 5 percent from the
previous year’s harvest.
The tobacco subsidy program guaranteed farmers a minimum price for their
crops and allocated quotas stating how many acres could be planted. Growers who
did not own a quota had to purchase or rent one from current owners. This system
increased prices and limited production to narrow geographic areas and to plots of
land that were typically not larger than 10 acres. When the system was disbanded
in 2004, thousands of farmers stopped growing tobacco. In 2005, acreage dropped
P2
PE
QEQD QS
S
D
Surplus
0 Q
P
59Scott Kilman, “Farmers to Plant Record Soybean Acres Despite Price Drop, as a Result of a Subsidy,” Wall
Street Journal, April 4, 1999.
FIgure 2.8
A higher-than-equilibrium Price
A surplus of a good results when the
market price, P2, is above the equilib-
rium price, PE.
M02_FARN0095_03_GE_C02.INDD 66 13/08/14 1:42 PM
ChAPter 2 Demand, Supply, and Equilibrium Prices 67
27 percent from the previous year, and tobacco prices fell from $1.98 to $1.64 per
pound. However, over time tobacco production has increased again, given that
farmers no longer have to purchase quotas and can plant much more acreage.
Tobacco production has shifted to large tracts of land where the crop can be grown
more efficiently.60
By 2011, many crop prices had increased so significantly that they were too high to
trigger payouts under a price support formula. Global grain markets shifted in 2006
when the federal government required that the oil industry mix billions of gallons of
corn-derived ethanol with gasoline each year. The increase in middle-class consum-
ers in emerging economies such as China also increased the demand for corn. In
2011, corn sold for $7 per bushel, far above the target price of $2.63, and soybeans
sold for $13 per bushel compared with a target price of $6. Economists estimated that
commodity prices would not fall back to target levels for at least another decade.61
Mathematical Example of Equilibrium
We can illustrate the concept of equilibrium with the mathematical example of
the copper industry we have been using throughout the chapter. So far, we have
defined the demand and supply curves for copper in Equations 2.3 and 2.6:
2.3 QD ∙ 15 ∙ 2PC
2.6 QS ∙ ∙25 ∙ 8PC
Equilibrium in a competitive market occurs at the price where quantity demanded
equals quantity supplied. Since Equation 2.3 represents quantity demanded as a
function of price and Equation 2.6 represents quantity supplied as a function of
price, we need an equilibrium condition to find the solution in the market. The equi-
librium condition is shown in Equation 2.7 where we set the two equations equal to
each other and solve for the equilibrium price and quantity.
2.7 QD ∙ QS
15∙2PC ∙ ∙25 ∙ 8PC
40 ∙ 10PC
PC ∙ PE ∙ $4.00 and QE ∙ 7 (by substituting $4.00 into
either equation)
Thus, the equilibrium price of copper in this example is $4.00 per pound, and the
equilibrium quantity is 7 million pounds. This is the only price–quantity combina-
tion where quantity demanded equals quantity supplied. At a price lower than $4.00
per pound, the quantity demanded from Equation 2.3 will be greater than the quan-
tity supplied from Equation 2.6, and a shortage of copper will result. At a price
higher than $4.00 per pound, the quantity demanded will be less than the quantity
supplied, and a surplus of copper will occur.
Changes in Equilibrium Prices and Quantities
Changes in equilibrium prices and quantities occur when market forces cause either
the demand or the supply curve for a product to shift or both curves shift. These
shifts occur when one or more of the factors held constant behind a given demand
60Etter, “U.S. Farmers Rediscover the Allure of Tobacco.”
61Scott Kilman, “Crop Prices Erode Farm Subsidy Program,” Wall Street Journal (Online), July 25, 2011.
M02_FARN0095_03_GE_C02.INDD 67 13/08/14 1:42 PM
68 PArt 1 Microeconomic Analysis
or supply curve change. Much economic analysis focuses on examining the changes
in equilibrium prices and quantities that result from shifts in demand and supply.
Change in Demand Figure 2.9 shows the effect of a change in demand in a
competitive market. The original equilibrium price, P0, and quantity, Q0, arise from
the intersection of demand curve D0 and supply curve S0. An increase in demand
is shown by the rightward or outward shift of the demand curve from D0 to D1.
This increase in demand could result from a change in one or more of the follow-
ing nonprice variables: tastes and preferences, income, prices of related goods,
expectations, or number of consumers in the market, as we discussed earlier in the
chapter. This increase in demand results in a new higher equilibrium price, P1, and
a new larger equilibrium quantity, Q1, or in the movement from point A to point B in
Figure 2.9. This change represents a movement along the supply curve or a change
in quantity supplied. Thus, a change in demand (a shift of the curve on one side of
the market) results in a change in quantity supplied (movement along the curve on
the other side of the market).
The opposite result occurs for a decrease in demand. In this case, the demand
curve shifts from D0 to D2 in Figure 2.9, and the equilibrium price and quantity fall
to P2 and Q2. This change in demand also causes a change in quantity supplied, or a
movement along the supply curve from point A to point C.
Change in Supply Figure 2.10 shows the effect of a change in supply on equi-
librium price and quantity. Starting with the original demand and supply curves,
D0 and S0, and the original equilibrium price and quantity, P0 and Q0, an increase
in supply is represented by the rightward or outward shift of the supply curve
from S0 to S1. As we discussed earlier in the chapter, this shift could result from a
P2
P0
P1
Q0Q2 Q1
S0
D0
D1
0
D2
A
C
B
Q
PFIgure 2.9
Change in Demand
A change in demand, represented by
a shift of the demand curve, results in
a movement along the supply curve.
P2
P0
P1
Q0Q2 Q1
S0
S1
S2
D0
0
A
C
B
Q
P
FIgure 2.10
Change in Supply
A change in supply, represented by a
shift of the supply curve, results in a
movement along the demand curve.
M02_FARN0095_03_GE_C02.INDD 68 13/08/14 1:42 PM
ChAPter 2 Demand, Supply, and Equilibrium Prices 69
change in technology, input prices, prices of goods related in production, expecta-
tions, or number of suppliers. The result of this increase in supply is a new lower
equilibrium price, P1, and a larger equilibrium quantity, Q1. This change in supply
results in a movement along the demand curve or a change in quantity demanded
from point A to point B.
Figure 2.10 also shows the result of a decrease in supply. In this case, the supply
curve shifts leftward or inward from S0 to S2. This results in a new higher equi-
librium price, P2, and a smaller equilibrium quantity, Q2. This decrease in supply
results in a decrease in quantity demanded or a movement along the demand curve
from point A to point C.
Changes on Both Sides of the Market As in the copper case discussed
at the beginning of this chapter, most outcomes result from changes on both
sides of the market. The trends in equilibrium prices and quantities will depend
on the size of the shifts of the curves and the responsiveness of either quantity
demanded or quantity supplied to changes in prices.
In some cases, we know the direction of the change in equilibrium price, but
not the equilibrium quantity. This result is illustrated in Figures 2.11 and 2.12,
which show a decrease in supply (the shift from point A to point B) combined
with an increase in demand (the shift from point B to point C). Both shifts cause
the equilibrium price to rise from P0 to P2. However, the direction of change
for the equilibrium quantity (Q0 to Q2) depends on the magnitude of the shifts
in the curves. If the decrease in supply is less than the increase in demand, the
equilibrium quantity will rise, as shown in Figure 2.11. The equilibrium quantity
will fall if the increase in demand is less than the decrease in supply, as shown
in Figure 2.12.
P2
P0
P
P1
Q0 QQ2Q1
S0
S1
D0
D1
0
A
C
B
FIgure 2.12
Decrease in Supply and Increase
in Demand: Decrease in
equilibrium Quantity
These changes in demand and supply
result in a higher equilibrium price
and a smaller equilibrium quantity.
P2
P0
P1
Q0 Q2Q1
S0
S1
D0
D1
0
A
C
B
Q
P FIgure 2.11
Decrease in Supply and
Increase in Demand: Increase in
equilibrium Quantity
These changes in demand and supply
result in a higher equilibrium price
and a larger equilibrium quantity.
M02_FARN0095_03_GE_C02.INDD 69 13/08/14 1:42 PM
70 PArt 1 Microeconomic Analysis
In other cases, we know the direction of the change in the equilibrium quantity,
but not the equilibrium price. Figures 2.13 and 2.14, which illustrate this situation,
show an increase in supply (from point A to point B) combined with an increase
in demand (from point B to point C). Both of these shifts in the curves result in
a larger equilibrium quantity (an increase from Q0 to Q2). However, the direction
of the price change depends on the magnitude of the shift in each curve. If the
increase in demand is less than the increase in supply, the equilibrium price will
fall, as shown in Figure 2.13. The equilibrium price will rise if the increase in supply
is less than the increase in demand, as shown in Figure 2.14.
Mathematical Example of an Equilibrium Change
In Equation 2.7, we solved for the equilibrium price and quantity of copper with
demand Equation 2.3 and supply Equation 2.6. This resulted in an equilibrium price
of $4.00 per pound and an equilibrium quantity of 7 million pounds. We now show
how a change in the equilibrium price and quantity from the beginning of 2010 to
the end of 2011 resulted from changes in the factors discussed in the beginning of
this chapter.
Suppose the recession in the United States and the turmoil in European finan-
cial conditions resulted in the cancellation of copper-using projects and there
was no offsetting increase in the demand for copper from China. Assume that
this change caused the income index (I) in demand Equation 2.3 to decrease
from 20 to 14. The uncertain demand for telecommunications services in North
America and Europe caused the telecom index to decrease from 2.5 to 1.875.
The expectations index (E) decreased over the period from 100 to 80, given that
P0
S0
D1
D0
S1
A
C
B
P2
P1
Q1 Q2Q00 Q
P
P0
S0
D1
D0
S1
A
C
B
P2
P1
Q1 Q2Q00 Q
PFIgure 2.13
Increase in Supply and Increase in
Demand: Lower equilibrium Price
These changes in demand and supply
result in a lower equilibrium price and
a larger equilibrium quantity.
FIgure 2.14
Increase in Supply and Increase
in Demand: higher equilibrium
Price
These changes in demand and supply
result in a higher equilibrium price
and a larger equilibrium quantity.
M02_FARN0095_03_GE_C02.INDD 70 13/08/14 1:42 PM
ChAPter 2 Demand, Supply, and Equilibrium Prices 71
a larger number of purchasers expected a lower price over the following six
months. These changes give a new demand function, as shown in Equation 2.8.
2.8 QD2 ∙ 3 ∙ 2PC ∙ 0.2I ∙ 1.6TC ∙ 0.04E
∙ 3 ∙ 2PC ∙ 0.2(14) ∙ 1.6(1.875) ∙ 0.04(80)
∙ 12 ∙ 2PC or [PC ∙ 6 ∙ 0.5QD]
Also suppose that the wage index, W, decreased slightly from 100 to 98 over this
period due to the economic slowdown. Assume that improvements in physical
capital increased the technology index (T ) from 50 to 55 and that China released
some of its previously undiscovered stockpiles of copper, which had the effect of
increasing the value of N from 20 to 28. These changes gave a new supply function,
as shown in Equation 2.9.
2.9 QS2 ∙ ∙5 ∙ 8PC ∙ 0.5W ∙ 0.4T ∙ 0.5N
∙ ∙5 ∙ 8PC ∙ 0.5(98) ∙ 0.4(55) ∙ 0.5(28)
∙ ∙18 ∙ 8PC or [PC ∙ 2.25 ∙ 0.125QS]
The new equilibrium price and quantity are derived in Equation 2.10 by setting the
new demand function, Equation 2.8, equal to the new supply function, Equation 2.9.
2.10 QD2 ∙ QS2
12 ∙ 2PC ∙ ∙18 ∙ 8PC
30 ∙ 10PC
PC ∙ PE ∙ $3.00 and QE ∙ 6
The resulting equilibrium price is $3.00 per pound of copper, and the equilibrium
quantity is 6 million pounds.
Figure 2.15 shows the original and final equilibrium in the copper industry
example (original PE = $4.00 per pound, QE = 7 million pounds; final PE = $3.00
per pound, QE = 6 million pounds, respectively). Both the demand and the supply
curves are graphed from the equations showing price as a function of quantity. We
can see that both the demand curve shift and the supply curve shift resulted in a
lower equilibrium price.
0 5 6 7
$
/lb
0
1
2
3
4
5
6
7
8
10 15 20
A
B
S2010
D2011
S2011
D2010
millions of lbs
FIgure 2.15
Copper Industry example
This figure illustrates the changes in
demand and supply in the copper
industry discussed in the opening
case of the chapter. Both the demand
and the supply shifts resulted in a
downward trend in copper prices.
M02_FARN0095_03_GE_C02.INDD 71 13/08/14 1:42 PM
72 PArt 1 Microeconomic Analysis
Summary
In this chapter, we discussed how the forces of demand and supply determine
prices in competitive markets. In the case of the copper industry, we saw how both
demand- and supply-side factors influenced copper prices. We also discussed how
both microeconomic factors, such as a change in the technology of copper produc-
tion, and macroeconomic factors, including changes in the Chinese economy and
the U.S. recession that began in 2007, affected the prices charged, the profitability,
and the competitive strategies of firms in the copper industry.
We examined these changes with the economic model of demand and supply.
Demand is defined as the relationship between the price of the good and the quan-
tity demanded by consumers in a given period of time, all other factors held con-
stant. Supply is defined as the relationship between the price of the good and the
quantity supplied by producers in a given period of time, all other factors held con-
stant. The equilibrium price, or the price that actually exists in the market, is that
price where quantity demanded equals quantity supplied and is represented by the
intersection of given demand and supply curves. When the factors held constant
behind a particular demand or supply curve change, equilibrium prices respond to
these demand and supply shifters. We provided numerous examples of these shift-
ers throughout the chapter and discussed the effect of these demand and supply
changes on prices in the copper industry.
We’ll later examine the quantitative concept of elasticity, which economists have
developed to measure the amount of consumer response to changes in the vari-
ables in market demand functions. We’ll also examine what impact elasticity has on
a firm’s revenues and pricing policies (Chapter 3).
Key Terms
change in demand, p. 55
change in quantity demanded, p. 55
change in quantity supplied, p. 61
change in supply, p. 61
complementary goods, p. 52
demand, p. 49
demand curve, p. 54
demand shifters, p. 54
equilibrium price, p. 64
equilibrium quantity (QE), p. 64
functional relationship, p. 49
horizontal summation of individual
demand curves, p. 56
individual demand function, p. 54
individual supply function, p. 61
inferior good, p. 50
linear demand function, p. 56
linear supply function, p. 61
market demand function, p. 54
market supply function, p. 61
negative (inverse) relationship, p. 54
normal good, p. 50
positive (direct) relationship, p. 61
substitute goods, p. 51
supply, p. 58
supply curve, p. 61
supply shifters, p. 61
Exercises
Technical Questions
1. Consider the demand for computers. For each of
the following, state the effect on demand:
a. An increase in consumer incomes
b. An increase in the price of computers
c. A decrease in the price of Internet service
providers
d. A decrease in the price of semiconductors
e. It is October, and consumers expect that com-
puters will go on sale just before Christmas.
M02_FARN0095_03_GE_C02.INDD 72 13/08/14 1:42 PM
ChAPter 2 Demand, Supply, and Equilibrium Prices 73
2. Consider the supply of computers. For each of the
following, state the effect on supply:
a. A change in technology that lowers production
costs
b. An increase in the price of semiconductors
c. A decrease in the price of computers
d. An increase in the wages of computer assembly
workers
e. An increase in consumer incomes
3. The demand curve is given by
QD ∙ 500 ∙ 5PX ∙ 0.5I ∙ 10PY ∙ 2PZ
where
QD = quantity demanded of good X
PX = price of good X
I = consumer income, in thousands
PY = price of good Y
PZ = price of good Z
a. Based on the demand curve above, is X a nor-
mal or an inferior good?
b. Based on the demand curve above, what is the
relationship between good X and good Y?
c. Based on the demand curve above, what is the
relationship between good X and good Z?
d. What is the equation of the demand curve if
consumer incomes are $30,000, the price of
good Y is $10, and the price of good Z is $20?
e. Graph the demand curve that you found in (d),
showing intercepts and slope.
f. If the price of good X is $15, what is the quantity
demanded? Show this point on your demand
curve.
g. Now suppose the price of good Y rises to $15.
Graph the new demand curve.
4. The supply curve is given by
QS ∙ ∙200 ∙ 20PX ∙ 5PI ∙ 0.5PZ
where
QS = quantity supplied of good X
PX = price of good X
PI = price of inputs to good X
PZ = price of good Z
a. Based on the supply curve above, what is the
relationship between good X and good Z?
b. What is the equation of the supply curve if input
prices are $10 and the price of Z is $20?
c. Graph the supply curve that you found in (b),
showing intercepts and slope.
d. What is the minimum price at which the firm
will supply any of good X at all?
e. If the price of good X is $25, what is the quantity
supplied? Show this point on your supply curve.
f. Now suppose the price of inputs falls to $5.
Graph the new supply curve.
5. Consider the market for Good X.
a. Suppose that consumers do not buy any of
Good X at the price of $120, and for every $10
decrease in price, the quantity consumed in-
creases by 20. Write the equation for the de-
mand curve of Good X.
b. Suppose that producers do not produce any of
Good X at the price of $50, and for every $10
increase in price, the producers increase the
quantity produced by 30. Write the equation for
the supply curve of Good X.
c. Find the equilibrium price and quantity.
d. At what price does this market have a shortage
of 40?
e. At what price does this market have a surplus
of 60?
6. Graph representative supply and demand curves
for the breakfast cereal market, labeling the cur-
rent equilibrium price and quantity. Then show the
effect on equilibrium price and quantity of each of
the following changes (consider each separately):
a. The price of muffins rises, assuming muffins
and breakfast cereals are substitutes.
b. The price of wheat, an input to cereal produc-
tion, rises.
c. Consumers expect that cereal prices will be
higher in the future.
d. There is a change in technology that makes pro-
duction less expensive.
e. New medical reports indicate that eating break-
fast is less important than had previously been
thought.
7. Consider the following markets, and draw repre-
sentative supply and demand curves.
a. What is the effect of bad weather on the equilib-
rium price and quantity of coffee beans?
b. What is the effect of the change in the market
for coffee beans on the equilibrium price and
quantity of coffee?
c. What is the effect of the change in the market
for coffee on the equilibrium price and quantity
of a powdered non-dairy creamer?
d. What is the effect of the change in the market
for coffee on the equilibrium price and quantity
of tea?
8. Consider the market for hamburger, and draw rep-
resentative supply and demand curves.
a. Assume that hamburger is an inferior good.
Suppose that consumer incomes fall, and at the
M02_FARN0095_03_GE_C02.INDD 73 13/08/14 1:42 PM
74 PArt 1 Microeconomic Analysis
same time, an improvement in technology low-
ers production costs. Show this on your graph.
If you have no other information, what can you
say about the change in equilibrium price and
quantity?
b. Now suppose that you have the additional in-
formation that the change in consumer incomes
has been relatively small, while the reduction in
production costs has been relatively large. How
would this change your answer to (a)?
Application Questions
1. Using the facts in the opening case, the discussion
in the chapter, and demand and supply curves, show
the impacts of the events in the case on the price
and quantity of copper. Clearly distinguish between
changes in demand and supply and changes in the
quantity demanded and the quantity supplied.
2. Using data sources from business publications
and the Internet, discuss significant trends in both
demand and supply in the copper industry that have
influenced the price of copper since September
2011. What are the implications of these trends for
managerial decision making in the copper industry?
3. According to leading coffee merchants, there will be
a shortage in the global coffee market due to Brazil’s
declining coffee production in 2014-2015.62 Brazil is
the world’s largest producer of coffee beans.
a. Using demand and supply analysis, illustrate
the effect of Brazil’s declining coffee produc-
tion on the global coffee market in 2014. Show
the shortage of coffee beans in your graph.
b. With the shortage, how would you expect the
price of coffee beans to change in 2014?
c. Arabica coffee beans were being traded at
seven-year lows in November 2013. If the
demand for coffee beans decreases further in
2014, along with the change mentioned above,
how would you expect the equilibrium price
and quantity of Arabica coffee beans to change?
Draw a graph to explain your answer.
4. Consider the following discussion.63
It has been a tough year in the poultry business,
with supply outpacing demand and feed-grain
prices rising substantially. But producers are
hoping all that changes when the summer
cook-out season starts.
The seasonal upswing in chicken consumption,
along with the anticipated jump in spot-market
poultry prices, could bring some relief to produc-
ers whose profit margins have been slashed by
surging corn and soybean-meal costs.
Rising feed-grain prices, accelerated by the diver-
sion of corn to make ethanol, have pushed up the
cost of producing a live chicken by as much as 65
percent over the past two years.
Three factors make analysts more optimistic:
Companies are cutting production, weekly egg-set
numbers are declining (egg sets are fertile eggs
placed in incubators), and prices are responding
positively to the decreasing supply.
The production slowdown is a response to the
surge in feed-grain prices last fall.
Profit margins at producers will not improve
unless spot-market prices for chicken move up
fast enough to cover costs paid for corn and
soybean meal to feed chicken flocks.
62William Jones, “Brazil Coffee Production to Decrease: Daily,” The
Rio Times (Online), January 7, 2014.
63“Chicken Producers in Price Pinch,” Wall Street Journal, May 21,
2008.
M02_FARN0095_03_GE_C02.INDD 74 13/08/14 1:42 PM
ChAPter 2 Demand, Supply, and Equilibrium Prices 75
Production cutbacks and seasonal demand have
helped fuel a 20-cent increase in boneless, skinless
breast-meat prices to $1.46 a pound. Prices are
expected to reach at least $1.80 by summer 2008.
a. Use demand and supply analysis to illustrate
the changes in chicken prices described in the
above article.
b. Describe what has happened in the corn and
soybean-meal markets and how that has influ-
enced the chicken market.
5. In early 2014, China decided to cancel the 1.2
million ton rice import contract with Thailand.64
Using demand and supply analysis, answer the fol-
lowing questions with supporting graphs.
a. What will be the effect of the cancellation of the
import contract on the equilibrium price and
quantity of rice in Thailand?
b. What will be the effect of the cancellation of the
import contract on the equilibrium price and
quantity of rice in China?
c. If China increases the import of rice from
Vietnam in order to stabilize the domestic price
of rice, how will it affect the equilibrium price
and quantity of rice in Vietnam?
d. Suppose the Vietnamese rice producers have
built the storehouses and expect the increasing
demand for rice from China will keep pushing
the price up in the future. How would you ex-
pect the current equilibrium price and quantity
of rice in Vietnam to change, compared to those
before China increases the import of rice from
Vietnam?
64Dat Viet, “China Refuses Thai Rice, Bringing Opportunities to Vietnam,” VietNamNet Bridge, February 12,
2014.
M02_FARN0095_03_GE_C02.INDD 75 13/08/14 1:42 PM
In this chapter, we explore the concept of demand in more detail. We focus on the downward sloping demand curve, which shows an inverse relationship between the price of the good and the quantity demanded by consumers, all else held constant. This demand curve applies to the
entire market or industry in a perfectly competitive market structure, even
though individual firms in this market are price-takers who cannot influence
the product price.
All firms in the other market structures—monopolistic competition, oli-
gopoly, and monopoly—face downward sloping demand curves because they
have varying degrees of market power. These firms must lower the price at
which they are willing to sell their product if they want to sell more units. If
the product price is higher, consumers will buy fewer units. Thus, product
price is a strategic variable that managers in all real-world firms must choose.
Managers must also develop strategies regarding the other variables influenc-
ing demand, including tastes and preferences, consumer income, the price of
related goods, and future expectations. This chapter focuses on the quanti-
tative measure—demand elasticity—that shows how consumers respond to
changes in the different variables influencing demand.
We begin this chapter with a case that discusses Procter & Gamble Co.’s
pricing strategies and consumer responsiveness to the demand for its prod-
ucts as the company navigated the economic downturn from 2009 to 2011. We
then formally present the concept of price elasticity of demand and develop
a relationship among changes in prices, changes in revenues that a firm
receives, and price elasticity. Next we illustrate all elasticities with examples
drawn from both the economics and the marketing literature.
The chapter appendix presents the formal economic model of consumer
behavior, which shows how both consumer tastes and preferences and the
constraints of income and product prices combine to influence the consumer’s
choice of different products.
3 Demand Elasticities
76
M03_FARN0095_03_GE_C03.INDD 76 13/08/14 1:41 PM
Like many other companies, Procter & Gamble Co. (P&G) had
to constantly alter its pricing strategies as it faced declining
and shifting consumer demand for many of its products from
2009 to 2011. Although the recession that began in December
2007 officially ended in June 2009, P&G managers continued
to face consumer cutbacks even on basic household staples.
Rather than purchase P&G premium-priced brands, such as
Tide detergent and Pampers diapers, consumers chose less-
expensive brands, including Gain detergent and Luvs diapers.
The P&G chief executive noted at the time that consumers were
trying more private-label and retailer brands than they would in
more normal economic times.1
Because the company also faced higher commodity prices
and global currency swings, P&G officials raised prices in the
first quarter of 2009, developed new products, and increased
advertising to emphasize why their brands offered more value
than the competition. Officials reported that the higher prices
hurt sales volume but increased total sales revenues by 7 per-
cent. However, industry analysts wondered if the deceased
sales volume would eventually cause the company to lower
prices and increase promotions.2
By spring 2010, P&G had reversed course and was engaged
in a market-share war by cutting prices, increasing product
launches and spending more on advertising. The company’s
goal was to win back market share lost during the recession to
lower-priced rivals even at the expense of profitability. P&G
lowered prices on almost all of its product categories during
early 2010.3
This strategy continued into the summer of 2010,
although there were concerns at that point that the com-
pany had missed industry analyst profit estimates even
though it had increased market share. Although the company
announced that it intended to raise prices in the first half
of 2011, officials debated whether consumers had become
accustomed to the lower prices. Industry analysts argued
that the company needed to sell more products in the lower-
priced categories.4
The ongoing discounting reduced P&G’s profits, which
decreased 12 percent in the second quarter of 2010, because
sales revenue rose less than P&G expected. To offset the nega-
tive effects of the lower prices, P&G introduced new products
including Gillette razors that promised a less irritating shave,
Crest toothpaste with a “sensitive shield,” and Downy fabric
softener that advertised keeping sheets smelling fresh for a
week. The company also began moving into emerging mar-
kets such as Brazil, where its research showed that Brazillians
took more showers, used more hair conditioner, and brushed
their teeth more often than residents of any other country. The
company planned to enter the Brazillian market in several new
product categories at once, such as Oral B toothpaste and Olay
skin cream.5
Given continued lower-than-expected revenue and slow
sales in early 2011, P&G announced that it would cut costs
but would also try to raise prices on goods to offset the
higher costs. P&G announced initiatives to eliminate some
manufacturing lines and sell off smaller brands. However,
private-label brands continued to post larger sales gains than
brand names.6
In April 2011, the company announced a 7 percent
increase in prices for its Pampers diapers and a 3 percent
increase in the price of wipes. Surveys indicated that cus-
tomers were less likely to switch to a cheaper baby product
than for items such as bleach, bottled water, and liquid soap.
The company hoped that parents would be willing to pay
higher prices for diapers, even if they cut back elsewhere,
in the belief that the higher-priced products were better for
their baby’s comfort or development. P&G also raised the
price of its Charmin toilet paper and Bounty paper towels.
One industry analyst concluded that brands that had the
highest market share, were purchased infrequently (such as
sunscreen or light bulbs), were necessities, had few competi-
tors, or where it would be difficult to reduce consumption
Case for Analysis
Demand Elasticity and Procter & Gamble’s Pricing Strategies
77
1Ellen Byron, “P&G, Colgate Hit by Consumer Thrift—
Household Products Makers See Sales Weakening, Raise Prices
to Keep Quarterly Profits from Plunging,” Wall Street Journal
(Online), May 1, 2009.
2Byron, “P&G, Colgate Hit by Consumer Thrift—Household
Products Makers See Sales Weakening, Raise Prices to Keep
Quarterly Profits from Plunging.”
3Ellen Byron, “P&G Puts Up Its Dukes Over Pricing—Consumer-
Products Makers Risk Margins to Grab Market Share from
Rivals and Cheap Store Brands,” Wall Street Journal (Online),
April 30, 2010.
4John Jannarone, “The Hefty Price of Procter’s Gambet,” Wall
Street Journal (Online), August 12, 2010.
5Ellen Byron, “P&G Chief Wages Offensive Against Rivals, Risks
Profits,” Wall Street Journal (Online), August 19, 2010.
6Ellen Byron, “Earnings: P&G Feels the Pinch of Rising Costs,”
Wall Street Journal (Online), January 28, 2011.
M03_FARN0095_03_GE_C03.INDD 77 13/08/14 1:41 PM
78 PArt 1 Microeconomic Analysis
Demand Elasticity
A demand elasticity is a quantitative measurement (coefficient) showing the
percentage change in the quantity demanded of a particular product relative to
the percentage change in any one of the variables included in the demand func-
tion for that product. Thus, an elasticity can be calculated with regard to product
price, consumer income, the prices of other goods and services, advertising bud-
gets, education levels, or any of the variables included in a demand function.9
The important point is that an elasticity measures this responsiveness in terms
of percentage changes in both variables. Thus, an elasticity is a number, called
a coefficient, that represents the ratio of two percentage changes: the percent-
age change in quantity demanded relative to the percentage change in the other
variable.
Percentage changes are used so that managers and analysts can make com-
parisons among elasticities for different variables and products. If absolute
changes were used instead of percentage changes and the quantities of prod-
ucts were measured in different units, elasticities could vary by choice of the
unit of measurement. For example, using absolute values of quantities, man-
agers would find it difficult to compare consumer responsiveness to demand
variables if the quantity of one product is measured in pounds and another is
measured in tons, because they would be comparing changes in pounds with
changes in tons.
Demand elasticity
A quantitative measurement
(coefficient) showing the
percentage change in the quantity
demanded of a particular product
relative to the percentage change
in any one of the variables included
in the demand function for that
product.
(toilet paper) were most likely to be the products whose
prices could be increased. P&G, with its distinctive items,
including beauty products, pet food, and toothpaste, was
likely to be better able to raise prices than Kimberly-Clark
and Clorox that operated in highly competitive product cat-
egories with large commodity cost pressures.7
By fall 2011, P&G reported solid sales growth and that it
had successfully raised prices even though some of its com-
petitors held back on their price increases. P&G had more abil-
ity to raise prices on its premium products because company
officials observed that higher-end consumer spending had held
up better than that of lower-income shoppers, who were still
affected by continuing unemployment. P&G lost some mar-
ket share in North America and Western Europe because its
competitors did not immediately follow its price increases.
However, company officials expected that the competitors
would soon follow P&G on its higher prices.8
This case illustrates how a company’s pricing policies
depend on how consumers respond to price changes. In the first
quarter of 2009, P&G raised prices and then reported declin-
ing sales volume but increased sales revenues. In subsequent
years, the company lowered prices, which increased sales vol-
ume, but did not increase revenue as much as expected so that
there was a negative effect on profits. Because the company
was concerned about consumer adjustment to lower prices over
time, it also adopted other strategies to increase profitability,
such as developing new products and entering new markets.
Thus, it appears from the above case that consumer respon-
siveness to a company’s price changes is related to
1. Tastes and preferences for various quality characteristics of
a product as compared to the impact of price
2. Consumer income and the amount spent on a product in
relation to that income
3. The availability of substitute goods and perceptions about
what is an adequate substitute
4. The amount of time needed to adjust to change in prices
To examine these issues in more detail, we first define demand
elasticity, and we relate this discussion to the variables influ-
encing demand.
9Although we can also calculate supply elasticities from a product supply function in a comparable manner,
we will postpone our discussion of this issue until we present the model of perfect competition (Chapter 7).
8Paul Ziobro, “P&G Says Costs Will Curb Current Quarter,” Wall
Street Journal (Online), October 28, 2011.
7Ellen Byron and Paul Ziobro, “Whoa Baby, Prices Are Jumping
for Diapers, Other Family Basics,” Wall Street Journal (Online),
April 25, 2011.
M03_FARN0095_03_GE_C03.INDD 78 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 79
Price Elasticity of Demand
The price elasticity of demand (ep) is defined as the percentage change in
the quantity demanded of a given good, X, relative to a percentage change in its
price, all other factors assumed constant, as shown in Equation 3.1.10 A percentage
change in a variable is the ratio of the absolute change (Q2 – Q1 or ΔQ; P2 – P1 or
ΔP) in that variable to a base value of the variable, as shown in Equation 3.2.
3.1 eP ∙
%∆QX
%∆PX
3.2 eP ∙
∆QX
QX
∆ PX
PX
∙
Q2 ∙ Q1
QX
P2 ∙ P1
PX
where
eP = price elasticity of demand
Δ = the absolute change in the variable: (Q2 − Q1) or (P2 − P1)
QX = the quantity demanded of good X
PX = the price of good X
Price elasticity of demand is illustrated by the change in quantity demanded
from Q1 to Q2 as the price changes from P1 to P2, or the movement along the
demand curve from point A to point B in Figure 3.1. Because we are moving along
a demand curve, all other factors affecting demand are assumed to be constant,
and we are examining only the effect of price on quantity demanded. All demand
elasticities are defined with the other factors influencing demand assumed
constant so that the effect of the given variable on demand can be measured
independently.
Price elasticity of demand
(eP)
The percentage change in the
quantity demanded of a given
good, X, relative to a percentage
change in its own price, all other
factors assumed constant.
10Price elasticity is sometimes called the “own price elasticity of demand” because it shows the ratio of the
percentage change in the quantity demanded of a product to the percentage change in its own price.
P2
P1
Q2
∆Q
∆P
Q1
Demand
0
A
B
Q
P FiGurE 3.1
Price Elasticity and the
Movement Along a Demand
Curve
Price elasticity is measured as a
movement along a demand curve
from point A to point B.
M03_FARN0095_03_GE_C03.INDD 79 13/08/14 1:41 PM
80 PArt 1 Microeconomic Analysis
The Influence of Price Elasticity on Managerial
Decision Making
Price elasticity of demand is an extremely important concept for a firm because it
tells managers what will happen to revenues if the price of a product changes. It
can also help firms develop a pricing strategy that will maximize their profits.
The demand for airline travel changed substantially between 1999 and 2006 due
to the following factors. Although business travelers have always been less price
sensitive than tourists because they have less ability to postpone a trip or search
for alternatives, they have become more price sensitive due to improvements in
electronic communications and increased restrictions on travel reimbursement.
Tightened security restrictions at airports have resulted in travelers having an
increased preference for direct flights. The option of purchasing tickets on the
Internet has reduced customer search costs and increased their knowledge
about alternative fares. It has been estimated that the price elasticity of demand
for tourists increased in absolute value from 0.78 to 1.05 and for business trav-
elers from 0.07 to 0.10 over the period 1999 to 2006. The price elasticity was
31 percent larger for tourists and 43 percent larger for business travelers. The
connection semi-elasticity, or the percentage reduction in quantity demanded if
a direct flight became a connecting flight, increased in absolute value from 0.55
to 0.75 for business travelers and from 0.75 to 0.80 for tourists. Thus, both busi-
ness travelers and tourists exhibited a stronger preference for direct flights in
2006.11 The airlines use knowledge about these different elasticities to develop
a complex schedule of prices for different groups of travelers and varying types
of flights.
Information on the price elasticity of demand for gasoline affects managerial
decisions in the automobile industry. In response to a price increase, consumers
may travel less by car, either switching to alternative modes of transport or by trav-
eling less in general. Consumers may also sell their cars, buy more efficient models,
or change the usage of various household models. Thus, changes in gasoline prices
affect the quantity demanded of gasoline though fuel efficiency, mileage per car,
and car ownership. Consumers will typically drive less in the short run and then
consider ownership changes in the long run.12
In an analysis based on data from 43 other studies, researchers estimated a short-
run price elasticity of gasoline demand of –0.34 and a long-run elasticity of –0.84.
Consumers have more options to adjust to changes in gasoline prices in the long
run than in the short run. The long-run elasticity estimate can be decomposed into
estimates of the price elasticities of fuel efficiency (0.31), mileage per car (–0.29),
and car ownership (–0.24). Thus, in the long run the response to changes in gaso-
line prices is driven by a similar size of response in terms of fuel efficiency, mile-
age per car, and car ownership. These relatively small price elasticities in both the
short run and the long run indicate that the use of gasoline taxes to decrease the
demand for gasoline may not be a very effective policy.13
Elasticities are also important for management in the public sector. For example,
a manager at a public transit agency needs to know how much decrease in ridership
will result if the agency raises transit fares and the impact of this fare increase on
the total revenue the agency receives from its passengers (the amount of money
received by a producer for the sale of its product, calculated as the price per unit
times the quantity sold).
total revenue
The amount of money received by
a producer for the sale of its prod-
uct, calculated as the price per unit
times the quantity sold.
11Steven Berry and Panle Jia, “Tracing the Woes: An Empirical Analysis of the Airline Industry,” American
Economic Journal: Microeconomics 2 (August 2010): 1–43.
12Martijn Brons, Peter Nijkamp, Eric Pels, and Piet Rietveld, “A Meta-Analysis of the Price Elasticity of
Gasoline Demand: A SUR Approach,” Energy Economics 30 (2008): 2105–22.
13Brons et al., “A Meta-Analysis of the Price Elasticity of Gasoline Demand: A SUR Approach.”
M03_FARN0095_03_GE_C03.INDD 80 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 81
Price Elasticity Values
The calculated value of all price elasticities for downward sloping demand curves
is a negative number, given the inverse relationship between price and quan-
tity demanded. If price increases, quantity demanded decreases and vice versa.
Therefore, it is easier to drop the negative sign and examine the absolute value
(|eP|) of the number to determine the size of the price elasticity. This procedure
leads to the definitions shown in Table 3.1.
As shown in Table 3.1, demand is elastic if the coefficient’s absolute value is
greater than 1 and inelastic if the coefficient’s absolute value is less than 1. For
elastic demand, the percentage change in quantity demanded by consumers is
greater than the percentage change in price. This implies a larger consumer respon-
siveness to changes in prices than does inelastic demand, in which the percent-
age change in quantity demanded by consumers is less than the percentage change
in price. In the case of unitary elasticity, where 0 eP 0 ∙1, the percentage change in
quantity demanded is exactly equal to the percentage change in price.
Elasticity and Total Revenue
The fourth column of Table 3.1 shows the relationship among price elasticity,
changes in prices, and total revenue received by the firm, which, as noted above, is
defined as price times quantity [(P)(Q)]. If demand is elastic, higher prices result in
lower total revenue, while lower prices result in higher total revenue. This outcome
arises because the percentage change in quantity is greater than the percentage
change in price. If the price increases, enough fewer units are sold at the higher
price that total revenue actually decreases. Likewise, with elastic demand, if price
decreases, total revenue increases. Even though each unit is now sold at a lower
price, there are enough more units sold that total revenue increases. Thus, for elas-
tic demand, changes in price and the resulting total revenue move in the oppo-
site direction. A higher price causes total revenue to decrease, while a lower price
causes total revenue to increase.
These relationships for elastic demand are illustrated for the demand curve
shown in Figure 3.2.14 For this demand curve, at a price of $10, 2 units of the prod-
uct are demanded, and the total revenue the firm receives is $10 × 2 units, or $20. If
the price decreases to $9, the quantity demanded increases to 3 units, and the total
revenue increases to $27. Demand is elastic in this range because total revenue
increases as the price decreases.
Elastic demand
The percentage change in quantity
demanded by consumers is greater
than the percentage change in
price and 0 eP 0 7 1.
inelastic demand
The percentage change in quantity
demanded by consumers is less
than the percentage change in
price and 0 eP 0 6 1.
unitary elasticity (or unit
elastic)
The percentage change in quantity
demanded is exactly equal to the
percentage change in price and
0 eP 0 ∙ 1.
VAluE oF ElAStiCity
CoEFFiCiEnt
ElAStiCity
DEFinition
rElAtionShiP AMonG
VAriAblES

iMPACt on totAl rEVEnuE
0 eP 0 7 1 Elastic demand % ∆Qx 7 % ∆Px Price increase results in lower total revenue.
Price decrease results in higher total revenue.
0 eP 0 6 1 Inelastic demand % ∆Qx 6 % ∆Px Price increase results in higher total revenue.
Price decrease results in lower total revenue.
0 eP 0 =1 Unit elastic or unitary
elasticity
% ∆Qx = % ∆Px Price increase or decrease has no impact on
total revenue.
tAblE 3.1 Values of Price Elasticity of Demand Coefficients
14This demand curve is also the basis for the numerical example in the next section of the chapter.
M03_FARN0095_03_GE_C03.INDD 81 13/08/14 1:41 PM
82 PArt 1 Microeconomic Analysis
This change in total revenue is illustrated graphically in Figure 3.2. If the price of
$10 is labeled P1 and the quantity of 2 units is labeled Q1, the total revenue of $20
is represented by the area of the rectangle 0P1AQ1. Likewise, if the price of $9 is
labeled P2 and the quantity of 3 units is labeled Q2, the total revenue of $27 is rep-
resented by the area of the rectangle 0P2BQ2. The change in revenue is represented
by a comparison of the size of the rectangle P1ACP2 (rectangle Y) with that of the
rectangle Q1CBQ2 (rectangle X). The first rectangle, Y, represents the loss in rev-
enue from selling the original 2 units at the lower price of $9 instead of the original
price of $10. This loss of revenue is 2 units times $1 per unit, or $2. The second
rectangle, X, represents the gain in revenue from selling more units at the lower
price of $9. This gain in revenue is 1 unit times $9 per unit, or $9. We can see both
numerically and graphically that the gain in revenue (rectangle X) is greater than
the loss in revenue (rectangle Y). Therefore, total revenue increases as the price is
lowered when demand is elastic.
The opposite result holds for inelastic demand. In this case, if the price increases,
total revenue also increases because the percentage decrease in quantity is less
than the percentage increase in price. With a price increase, enough units are still
sold at the higher price to cause total revenue to increase because each unit is
sold at the higher price. Likewise, if price decreases, total revenue will decrease.
All units are now being sold at a lower price, but the quantity demanded has not
increased proportionately, so total revenue decreases. Thus, for inelastic demand,
changes in price and the resulting total revenue move in the same direction.
A higher price causes total revenue to increase, while a lower price causes total
revenue to decrease.
Figure 3.3 illustrates this relationship for inelastic demand. For this demand
curve, at a price of $4, 8 units are demanded, and the firm receives $32 in revenue.
If the price falls to $3 per unit, the quantity demanded is 9 units, and the firm
takes in $27 in revenue. Thus, as the price decreases, total revenue decreases,
illustrating inelastic demand. In Figure 3.3, as in Figure 3.2, you can see the
change in total revenue by comparing rectangle P1ACP2 (rectangle Y) with rect-
angle Q1CBQ2 (rectangle X). When the price is lowered from $4 to $3, the 8 units
that were formerly sold at the price of $4 are now sold for $3 each. The associ-
ated revenue loss is $1 per unit times 8 units, or $8. The revenue gain is the one
additional unit that is now sold at a price of $3, or $3. It can be seen both graphi-
cally and numerically that the revenue gain (rectangle X) is less than the revenue
loss (rectangle Y). Therefore, as the price decreases with inelastic demand, total
revenue decreases.
If demand is unit elastic, changes in price have no impact on total revenue
because the percentage change in price is exactly equal to the percentage change
12
(P1) 10
(P2) 9
Demand
0 2 3
(Q1) (Q2)
12
A
BY
C
X
Area X = Q1CBQ2; Area Y = P1ACP2
Q
PFiGurE 3.2
Elastic Demand and total
revenue
If demand is elastic, a decrease in
price results in an increase in total
revenue, and an increase in price
results in a decrease in total revenue.
M03_FARN0095_03_GE_C03.INDD 82 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 83
Area X = Q1CBQ2; Area Y = P1ACP212
(P1) 4
(P2) 3
Demand
0 8 9
(Q1) (Q2)
12
A
BY
C X
Q
P FiGurE 3.3
inelastic Demand and total
revenue
If demand is inelastic, a decrease in
price results in a decrease in total
revenue, and an increase in price
results in a increase in total revenue.
15This discussion is drawn from Shlomo Maital, Executive Economics (New York: Free Press, 1994), 186–88.
Managerial rule of thumb
Estimating Price Elasticity
The examples of point elasticity and changes in revenue can be converted into managerial rules of
thumb for estimating price elasticity.15 Managers can get a ballpark estimate of price elasticity by ask-
ing their customers two questions:
1. What do you currently pay for my product? (Call this price P1.)
2. At what price would you stop buying my product altogether? (Call this price P2.)
Price elasticity can then be calculated as P1/(P1 – P2). The intuition behind this rule is that the higher the
value of P2, the higher the price the customer is willing to pay rather than do without the product, and
the lower the price elasticity. This rule of thumb is based on an implicit linear demand function and the
point price elasticity formula given in Equation 3.5 later in this chapter.
For the second rule, managers should ask themselves the following questions regarding a proposed
10 percent drop in the price of the firm’s product:
1. By how much will the sales revenue increase as a result of the higher volume of sales? (Call this
amount X.)
2. By how much will the sales revenue decrease as a result of a lower price on each unit sold? (Call this
amount Y.)
The price elasticity of demand is the ratio of X/Y. This rule of thumb is based on the changes in revenue
with elastic and inelastic demand illustrated in Figures 3.2 and 3.3. A large price elasticity coefficient
means that X will be large relative to Y, whereas a small price elasticity coefficient means that Y will be
large relative to X. ■
Determinants of Price Elasticity of Demand
Three major factors influence the price elasticity of demand and cause it to differ
among products:
1. The number of substitute goods
2. The percent of a consumer’s income that is spent on the product
3. The time period under consideration
in quantity. The effects on price and quantity are equal and offsetting. Rectangles
X and Y in Figures 3.2 and 3.3, representing the gain and loss of revenue, would be
exactly the same size if demand was unit elastic.
M03_FARN0095_03_GE_C03.INDD 83 13/08/14 1:41 PM
84 PArt 1 Microeconomic Analysis
All else held constant, demand is generally more inelastic or less responsive to price.
•    The fewer the number of substitutes or perceived substitutes available
•    The smaller the percent of the consumer’s income that is spent on the product
•    The shorter the time period under consideration
We’ll look at each of these factors in turn.
Number of Substitute Goods
If there are few substitute goods for a given product or, more important, if con-
sumers perceive there are few substitute goods for the product, managers have
more ability to raise prices without fear of losing sales than if a greater number of
substitute goods are available. Coke and Pepsi engage in extensive advertising to
convince their customers that the other product is not an adequate substitute. Each
company wants to shift out the demand curve for its product and make it relatively
more inelastic. This is a constant struggle, given the availability of a wide range of
substitute drinks: other soft drinks, teas, fruit drinks, sports beverages, and even
water. Coke and Pepsi have, of course, expanded into these other markets, so that
each company owns a number of substitute products for the basic cola.
In response to the expansion in the speciality-coffee market by McDonald’s Corp.,
which impacted consumers’ price sensitivity for Starbucks’ coffee, Starbucks
announced a plan in May 2010 to sell Seattle’s Best Coffee in 30,000 fast-food
outlets, supermarkets, and coffee houses. During the recession of 2007 to 2009,
Starbucks suffered a decline in same-store sales and closed hundreds of stores,
while McDonald’s expanded with its lower-priced coffees. Seattle’s Best was a for-
mer competitor that Starbucks acquired in 2003.16
In addition to the differences between business and leisure travel noted previ-
ously in the chapter, airline demand elasticities also depend on the length of the
trip. Because cars, buses, and trains are substitutes for shorter airline flights, these
flights should have a larger price elasticity of demand. Research studies have esti-
mated the following price elasticities: long-haul international business, –0.26; long-
haul international leisure, –0.99; long-haul domestic business, –1.15; and long-haul
domestic leisure, –1.52.17 These estimates confirm expectations about the role of
distance (international vs. domestic flights) and the differing elasticities for busi-
ness versus leisure travel.
The role of substitutes in influencing price elasticity of demand means that the
price elasticity of demand for the product of a specific producer will be larger than
the price elasticity for the product in general. All other producers of that same
product are substitutes for the specific producer. We discuss several examples
later in the chapter where the price elasticity of demand for the product is inelastic,
whereas it is elastic for the output of a specific producer.
Percent of Consumer’s Income Spent on the Product
Items that cost little tend to have more inelastic demands. If the price of your local
newspaper doubles tomorrow, going from 50 cents to $1, you may not even notice
the price increase, or perhaps you will choose to buy the paper four rather than
five times per week. If the price of the European vacation you have planned for
next summer doubles, you may consider traveling to a destination closer to home.
16Kevin Helliker, “Starbucks Targets Regular Joes: Firm to Offer Second Coffee Brand—Its Seattle’s Best—
in Fast-Food Outlets, Supermarkets, Machines,” Wall Street Journal (Online), May 11, 2010.
17David W. Gillen, William G. Morrison, and Christopher Stewart, Air Travel Demand Elasticities:
Concepts, Issues, and Measurement (Ottawa: Department of Finance, 2002).
M03_FARN0095_03_GE_C03.INDD 84 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 85
In this case, your quantity demanded decreases to zero, whereas there was only a
slight decrease for the newspaper case. As you would guess, consumers tend to be
more sensitive to changes in the prices of goods that represent a large percent of
their incomes.
Time Period
The shorter the time period, the less chance consumers have of finding acceptable
substitutes for a product whose price has risen, and the more inelastic the demand.
Over time, consumers can find a greater number of substitutes, and elasticities
tend to be larger. We noted these differences in the earlier discussion of short- and
long-run price elasticities of demand for gasoline.
Numerical Example of Elasticity,
Prices, and Revenues
We are now ready to explore the issues presented in Table 3.1 in more detail
through the use of a numerical example that illustrates the relationships among
elasticities, changes in prices, and changes in revenues to a firm. However, we first
discuss a problem that arises in the calculation of price elasticities.
Calculating Price Elasticities
A problem occurs during the calculation of price elasticities because there are dif-
ferent sources of data available for these calculations. We may have data on actual
quantities and prices, or we may have a demand equation that shows the functional
relationship between price and quantity demanded.
Arc Price Elasticity We first analyze the case with data on quantities and
prices. In Figure 3.1, we illustrated a large price change that resulted in a large
change in quantity demanded. If the price falls from P1 to P2, all else assumed
constant, the quantity demanded increases from Q1 to Q2. Because points Q1 and
Q2 may be significantly different from each other, a different value for the per-
centage change in quantity may result, depending on whether Q1 or Q2 is used
for the base quantity in Equation 3.2. If we are measuring the effect of a price
decrease from P1 to P2, which causes the quantity demanded to increase from Q1
to Q2, we will tend to use Q1 as the base because that is our beginning quantity.
If we are measuring the decrease in quantity demanded resulting from a price
increase from P2 to P1, we will tend to use quantity Q2 as the base quantity. The
same problem occurs when we are measuring the percentage change in price. We
will tend to use P1 as the base for price decreases and P2 as the base for price
increases because these are the current prices of the product.
Because an elasticity coefficient is just a number, it is useful to have that coef-
ficient the same for an increase or a decrease in quantity demanded. However, that
result might not occur with the example in Figure 3.1 because dividing the absolute
change in quantity (ΔQ) by Q1 could result in a quite different number than divid-
ing it by Q2. For example, if Q1 = 10 and Q2 = 20, ΔQ = 10. ΔQ/Q1 = 10/10 = 1.0, or
a 100 percent increase in quantity. However, ΔQ/Q2 = 10/20 = 0.5, or a 50 percent
decrease in quantity. The percentage increase in quantity is substantially different
from the percentage decrease in quantity.
This issue is not a problem with the definition of price elasticity; instead, it is
a numerical or calculation problem that arises for elasticity of demand when the
starting and ending quantities and prices are significantly different from each other,
as in Figure 3.1. We are calculating elasticity over a region or arc on the demand
M03_FARN0095_03_GE_C03.INDD 85 13/08/14 1:41 PM
86 PArt 1 Microeconomic Analysis
curve (point A to point B in Figure 3.1). The calculation problem can also arise if a
manager does not know the shape of the entire demand curve, but simply has data
on several prices and quantities.18
The conventional solution to this problem is to calculate an arc price elasticity
of demand, where the base quantity (or price) is the average value of the starting
and ending points, as shown in Equation 3.3.
3.3 eP =
(Q2 ∙ Q1)
(Q1 ∙ Q2)
2
(P2 ∙ P1)
(P1 ∙ P2)
2
Point Price Elasticity A price elasticity is technically defined for very tiny
or infinitesimal changes in prices and quantities. In Figure 3.1, if point B is moved
very close to point A, the starting and ending prices and quantities are also very
close to each other. We can then think of calculating an elasticity at a particular
point on the demand curve (such as point A). This can be done in either of two
ways: using calculus or using a noncalculus approach.
Equation 3.4 shows the formula for point price elasticity of demand where
d is the derivative from calculus showing an infinitesimal change in the variables.
3.4 eP ∙
dQX
QX
dPX
PX
∙
dQXPX
dPXQX
If you have a specific demand function, you can use calculus to compute the appro-
priate derivative (dQX/dPX) for Equation 3.4.
However, because we do not require calculus in this text, we’ll use a simpler
approach for a linear demand function. The point price elasticity of demand can be
calculated for a linear demand function as shown in Equation 3.5.
3.5 eP ∙
P
(P ∙ a)
where
P = the price charged
a = the vertical intercept of the plotted demand curve (the P-axis)19
Thus, for any linear demand curve, a point price elasticity can be calculated for
any price by knowing the vertical intercept of the demand curve (as plotted on the
P-axis) and using the formula in Equation 3.5.
Arc price elasticity of
demand
A measurement of the price
elasticity of demand where the base
quantity or price is calculated as the
average value of the starting and
ending quantities or prices.
Point price elasticity of
demand
A measurement of the price
elasticity of demand calculated at
a point on the demand curve using
infinitesimal changes in prices and
quantities.
18If a manager has data only on prices and quantities, he or she needs to be certain that all other factors
are constant as these prices and quantities change to be able to correctly estimate the price elasticity of
demand. This is the major problem in estimating demand functions and elasticities (Chapter 4).
19Following S. Charles Maurice and Christopher R. Thomas, Managerial Economics, 7th ed. (McGraw-Hill
Irwin, 2002), 92, the derivation of this result is as follows. For a linear demand curve,
P = a ∙ bQ or Q = [(P ∙ a)/b]
b = (∆P/∆Q) and 1/b = (∆Q/∆P)
eP = (∆Q/Q)/(∆P/P) = (∆Q/∆P)(P/Q) = (1/b)[P/(P ∙ a)/b] = [P/(P ∙ a)]
M03_FARN0095_03_GE_C03.INDD 86 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 87
Numerical Example
Table 3.2 presents a numerical example using a linear, or straight-line, downward
sloping demand curve. Demand curves may be either straight or curved lines,
depending on how people actually behave. Throughout most of this text, we use
linear downward sloping demand curves for our examples. These are the simplest
types of curves to illustrate mathematically. They are also good representations of
consumer behavior in different markets (Chapter 4).
The Demand Function
The demand function in Table 3.2 shows a relationship between quantity demanded
(Q) and price (P), with all other factors held constant. The effect of all the other
variables influencing demand is summarized in the constant term of 12.20 Demand
functions such as the one in Table 3.2 are estimated from data on real-world con-
sumer behavior (Chapter 4).
The first row in Table 3.2 shows that the demand function can be stated either as
quantity as a function of price or as price as a function of quantity. Mathematically, the
two forms of the relationship are equivalent. In a behavioral sense, we usually think
of quantity demanded as being a function of the price of the good. However, we use
the inverse form of the relationship, price as a function of quantity, to plot a demand
curve and to calculate the point price elasticity of demand, as shown in Equation 3.5.
Other Functions Related to Demand
Given the demand function in Table 3.2, we can derive a total revenue function,
which shows the total revenue (price times quantity) received by the producer as a
function of the level of output. To find total revenue, we can calculate the quantity
demanded at different prices and multiply the terms together, or we can use the
formal total revenue function given in Table 3.2.
Average revenue is defined as total revenue per unit of output. The average
revenue function shows how average revenue is related to the level of output.
Because total revenue equals (P)(Q), average revenue equals the price of the prod-
uct by definition. This is shown in the third line of Table 3.2. Thus, at any level of
output, the average revenue received by the producer equals the price at which that
output is sold.
total revenue function
The functional relationship that
shows the total revenue (price
times quantity) received by a
producer as a function of the level
of output.
Average revenue
Total revenue per unit of output.
Average revenue equals the price of
the product by definition.
Average revenue function
The functional relationship that
shows the revenue per unit of
output received by the producer at
different levels of output.
20If we had a demand function that explicitly included another variable such as income, once we put in a specific
value for income (to hold it constant), that number would become part of the constant term of the equation.
tAblE 3.2 numerical Example of Demand, total revenue, Average
revenue, and Marginal revenue Functions
Demand function Q ∙ 12 ∙ P or P ∙ 12 ∙ Q
Total revenue function TR ∙ (P)(Q) ∙ (12 ∙ Q)(Q) ∙ 12Q ∙ Q2
Average revenue function
AR ∙
TR
Q
∙
(P)(Q)
Q
∙ P
Marginal revenue function
MR ∙
∆TR
∆Q
∙
TR2 ∙ TR1
Q2 ∙ Q1
MR ∙
dTR
dQ
∙ 12 ∙ 2Q
M03_FARN0095_03_GE_C03.INDD 87 13/08/14 1:41 PM
88 PArt 1 Microeconomic Analysis
Marginal revenue is defined as the additional revenue that a firm receives from
selling an additional unit of output or the change in total revenue divided by the
change in output. It can be calculated in discrete terms if you have data on the total
revenue associated with different levels of output, as shown in the fourth line of
Table 3.2. If you have a mathematical total revenue function, the marginal rev-
enue function can be calculated by taking the derivative of the total revenue func-
tion with respect to output. (Because calculus is not required in this text, we will
supply any marginal revenue functions that you need.)
The numerical values for the functional relationships in Table 3.2 are given in
Table 3.3. The first two columns of Table 3.3 show the values of the demand func-
tion and the inverse relationship between price and quantity demanded. Column 3
presents total revenue for the different levels of output. Column 4 shows marginal
revenue calculated in discrete terms, which represents the change in total revenue
between one and two units of output, between two and three units of output, and
so on. Column 5 shows marginal revenue calculated from the marginal revenue
function presented in the last line of Table 3.2. In this case, marginal revenue is cal-
culated for an infinitesimal change in output that occurs at a given level of output.
Thus, Column 5 shows marginal revenue calculated precisely at a given level of
output compared with the Column 4 calculations of marginal revenue between dif-
ferent levels of output. You will notice that the values in Columns 4 and 5 are very
similar. The differences between Columns 4 and 5 are similar to the differences
between the arc and point price elasticities of demand we discussed earlier in the
chapter. Remember that these are differences in the calculation of the numbers,
not in the definition of the concepts.
Calculation of Arc and Point Price Elasticities
Table 3.4 illustrates arc and point price elasticity calculations from the demand
functions in Tables 3.2 and 3.3. Table 3.4 illustrates both the differences in the
calculation methods for arc and point price elasticities and the similarities in the
Marginal revenue
The additional revenue that a firm
takes in from selling an additional
unit of output or the change in total
revenue divided by the change in
output.
Marginal revenue function
The functional relationship that
shows the additional revenue a
producer receives by selling an
additional unit of output at differ-
ent levels of output.
tAblE 3.3 numerical Values for the Functional relationships
in table 3.2
(1) (2) (3) (4) (5)
Q P TR = (P)(Q) MR = ΔTR/ΔQ MR = dTR/dQ
0 12 0 12
1 11 11 11 10
2 10 20 9 8
3 9 27 7 6
4 8 32 5 4
5 7 35 3 2
6 6 36 1 0
7 5 35 −1 −2
8 4 32 −3 −4
9 3 27 −5 −6
10 2 20 −7 −8
11 1 11 −9 −10
12 0 0 −11 −12
M03_FARN0095_03_GE_C03.INDD 88 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 89
tAblE 3.4 Arc Price Elasticity Versus Point Price Elasticity
Calculations (Data from tables 3.2 and 3.3)
ArC ElAStiCity: ElAStiC DEMAnD
P1 = $10; Q1 = 2; TR1 = $20
P2 = $9; Q2 = 3; TR2 = $27
eP =
Q2 ∙ Q1
Q1 ∙ Q2
2
P2 ∙ P1
P1 ∙ P2
2
=
3 ∙ 2
2 ∙ 3
2
9 ∙ 10
10 ∙ 9
2
eP =
1
5
2
-1
19
2
=
2
5
-2
19
=
-19
5
= - 3.80
ArC ElAStiCity: inElAStiC DEMAnD
P1 = $4; Q1 = 8; TR1 = $32
P2 = $3; Q2 = 9; TR2 = $27
eP =
Q2 ∙ Q1
Q1 ∙ Q2
2
P2 ∙ P1
P1 ∙ P2
2
=
9 ∙ 8
8 ∙ 9
2
3 ∙ 4
4 ∙ 3
2
eP =
1
17
2
-1
7
2
=
2
17
-2
7
=
-7
17
= ∙0.41
Point ElAStiCity: ElAStiC DEMAnD
eP =
p
(P ∙ a)
where a = 12
P = $10
ep =
10
(10 ∙ 12)
=
10
-2
= ∙5.00
Point ElAStiCity: unit ElAStiC DEMAnD
ep =
P
(P ∙ a)
where a = 12
P = $6
ep =
6
(6 ∙ 12)
=
6
-6
= ∙1.00
Point ElAStiCity: inElAStiC DEMAnD
ep =
P
(P ∙ a)
where a = 12
P = $4
eP =
4
(4 ∙ 12)
=
4
-8
= ∙ 0.50
results. In this example, the arc price elasticity of demand between a price of $10
and a price of $9 is –3.80, while the point price elasticity calculated precisely at $10
is –5.00. The arc price elasticity calculated between a price of $4 and a price of $3 is
–0.41, while the point price elasticity at $4 is –0.50.
Price Elasticity Versus Slope of the Demand Curve
We can see in Table 3.4 that the price elasticity of demand is not constant along this
linear demand curve. At prices above $6, the demand is elastic, whereas the demand
is inelastic at prices below $6. The demand is unit elastic at a price of $6. We’ll explore
these relationships in more detail in the next section of the chapter. However, this
analysis does show us that elasticity and slope are not the same concepts. A linear
demand curve, like any straight line, has a constant slope, but the price elasticity of
demand varies along this demand curve. Thus, for a linear demand function, the price
M03_FARN0095_03_GE_C03.INDD 89 13/08/14 1:41 PM
90 PArt 1 Microeconomic Analysis
elasticity coefficient must be calculated for a specific price and quantity demanded
on that curve because the coefficient is smaller at lower prices than at higher prices.21
Demand Elasticity, Marginal Revenue, and Total Revenue
The relationships among demand, total revenue, and marginal revenue in Tables
3.2 and 3.3 are summarized in Figures 3.4 and 3.5. The inverse demand curve,
P = 12 – Q, is plotted in Figure 3.4 along with the corresponding marginal revenue
curve. Values of the total revenue function are plotted in Figure 3.5.
A firm is always constrained by its demand curve. In the case of the linear demand
curve in Figure 3.4, a price of $12 drives the quantity demanded to 0, and, at a price
of $0, the quantity demanded is 12 units. Thus, total revenue (price times quantity)
in Figure 3.5 begins and ends at zero at each end of the demand curve in Figure 3.4.
In the top half of the demand curve in Figure 3.4, when managers lower the price,
total revenue in Figure 3.5 increases. This means that demand is elastic in this range
of the demand curve, as a decrease in price results in an increase in total revenue.
At a price of $6 and a quantity demanded of 6 units, total revenue is maximized at
$36, as shown in Figure 3.5. Demand at this point is unit elastic. In the bottom half
of the demand curve, below a price of $6, a decrease in price causes total revenue
to fall as quantity demanded increases from 6 to 12 units of output. This means that
demand is inelastic for this portion of the demand curve. The decrease in total rev-
enue between 6 and 12 units of output is illustrated in Figure 3.5.
The marginal revenue curve is also plotted with the demand curve in Figure 3.4. The
marginal revenue curve begins at the point where the demand curve intersects the
price axis and then has a slope twice as steep as the demand curve. This can be seen
in the equations in Table 3.2, where the demand function is expressed as P = 12 – Q
and the marginal revenue function is MR = 12 – 2Q. This relationship between the
demand and the marginal revenue function holds for all linear downward sloping
demand curves. Once you draw the demand curve, you can draw the corresponding
marginal revenue curve, even if you do not have specific equations for the curves.
21Equation 3.2 can be simplified to show price elasticity as follows:
3.2 ep ∙
∆QX
QX
∆PX
PX
∙
(∆QX)(PX)
(∆PX)(QX)
The first ratio of variables in Equation 3.2 (QX/PX) is a slope term. It shows the absolute change in quantity
divided by the absolute change in price and is constant for a linear demand function. To calculate price
elasticity, however, we must multiply this slope term by the ratio of a given price and quantity demanded,
the second ratio of variables in Equation 3.2 (PX/QX). While the slope term remains constant along the
demand curve, the second term does not. As you move down the demand curve, price decreases and
quantity demanded increases, so the ratio and, thus, the price elasticity of demand decrease.
Marginal
Revenue
ep > 1
ep< 1
ep = 1
12
6
Demand
0 6 12 Q
$FiGurE 3.4
Demand and Marginal revenue
Functions
The demand, marginal revenue,
and total revenue functions are
interrelated, as shown in Figures 3.4
and 3.5.
M03_FARN0095_03_GE_C03.INDD 90 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 91
We can also see a relationship between marginal revenue and price elasticity in
Figure 3.4. Marginal revenue is positive, but decreasing in value, between a price of
$12 and a price of $6 (or between 0 and 6 units of output). This means that as price is
lowered in that range, total revenue increases, but at a decreasing rate.22 Figure 3.5
shows that total revenue increases from $0 to $36 as output increases from 0 to 6
units. However, the rate of increase lessens and the total revenue curve becomes
flatter as output approaches 6 units. Because the top half of the demand curve is the
elastic portion, marginal revenue must be a positive number when demand is elastic.
Decreases in the price below $6 cause the marginal revenue curve in Figure 3.4 to
become negative. The additional revenue that the firm takes in from selling an addi-
tional unit of output is negative. The total revenue function in Figure 3.5 starts to
decrease after 6 units of output are sold. We already established that the bottom half of
the demand curve is the inelastic portion of that curve. Thus, when demand is inelas-
tic, lowering the price decreases total revenue, so that marginal revenue is negative.
At the exact midpoint of the demand curve, marginal revenue equals zero. This is
also the point where total revenue reaches its maximum value. In Figure 3.5, total
revenue is at a maximum of $36 at a quantity of 6 units of output and a price of $6.
And as we established, demand is unit elastic at this price. Any small change in
price at this point will have no impact on total revenue. Table 3.5 summarizes all of
these relationships for a linear downward sloping demand curve.
22This can be explained mathematically because marginal revenue is the slope of the total revenue function.
The slope of the total revenue curve in Figure 3.5 decreases as output increases to 6 units.
tAblE 3.5 relationships for a linear Downward Sloping Demand Curve
ElAStiCity iMPACt on totAl rEVEnuE MArGinAl rEVEnuE
Elastic ↓P ⇒ ↑TR Positive (for increases in Q)
|eP| > 1
Upper half of demand curve
↑P ⇒ ↓TR
inelastic ↓P ⇒ ↓TR Negative (for increases in Q)
|eP| < 1
Lower half of demand curve
↑P ⇒ ↑TR
unit Elastic ↓P ⇒ No change in TR Zero
|eP| = 1
Midpoint of demand curve
↑P ⇒ No change in TR
TR is at its maximum value
Q
$
36
18
0 6 12
FiGurE 3.5
the total revenue Function
M03_FARN0095_03_GE_C03.INDD 91 13/08/14 1:41 PM
92 PArt 1 Microeconomic Analysis
Vertical and Horizontal Demand Curves
The previous discussion focused on linear downward sloping demand curves. We
use these examples to represent all downward sloping demand curves that exhibit
an inverse relationship between price and quantity demanded. These demand
curves are important to managers because they reflect typical consumer behav-
ior, with the price elasticity measuring how responsive quantity demanded is to
changes in price. There are, however, two polar cases of demand curves that we
should also consider: vertical and horizontal demand curves.
Vertical Demand Curves
Figure 3.6 presents a vertical demand curve. This curve shows that the quantity
demanded of the good is the same regardless of the price—in other words, there
is no consumer responsiveness to changes in the price of the good. This vertical
demand curve represents perfectly inelastic demand, where the elasticity coef-
ficient is zero (eP = 0).
Can you guess what, if any, types of goods would have such a demand curve?
Students often suggest products that are produced by only one supplier, such as
the electricity supplied by a local power utility in a state where there has been
no deregulation of electricity. Yet this answer is incorrect. Even if people can buy
their electric power from only one source, and even if they usually will not be very
responsive to price, they typically will not be totally unresponsive to changes in
price. If the price of electricity increases, people may choose to run their air condi-
tioners less in the summer or be more careful about how many lights they light up
in their houses. Thus, they are decreasing the quantity demanded of electricity in
response to a higher price and therefore do not have a vertical demand curve for
electricity.
A vertical demand curve would pertain to a product that is absolutely necessary
for life and for which there are no substitutes. Insulin for a diabetic might be a rea-
sonable example, although this answer relates to the product insulin in general and
not to a particular type of insulin produced by a specific drug company. You would
think that illegal, addictive drugs or other addictive substances would have very low
elasticities of demand, even if they are not zero. However, the evidence is not clear
even for these products. Researchers have estimated the price elasticity of demand
for marijuana to lie between –1.0 and –1.5, while that for opium to be approximately
–0.7 over shorter time periods and around –1.0 over longer periods. In a study of per-
sons arrested for cocaine and heroin use in 42 large cities from 1988 to 2003, cocaine
price elasticity estimates ranged from –0.07 to –0.17 and that for heroin use was
estimated at –0.1. The estimates were slightly lower for drug offenders compared
with nondrug offenders (–0.12 to –0.13 vs. –0.16 to –0.19 for cocaine use and –0.09
Perfectly inelastic demand
Zero elasticity of demand, illus-
trated by a vertical demand curve,
where there is no change in quan-
tity demanded for any change in
price.
Demand
0 Q1 Q
PFiGurE 3.6
Vertical Demand Curve
A vertical demand curve represents
perfectly inelastic demand.
M03_FARN0095_03_GE_C03.INDD 92 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 93
to –0.14 vs. –0.12 to –0.28 for heroin use). Cigarette smoking price elasticities have
been estimated at –0.75 for adults, while teenage smoking elasticities may be greater
than 1 in absolute value.23 Thus, even for addictive substances, the price elasticities
may not be close to zero. The key issues for perfectly inelastic demand are that the
product is necessary for life and there are no substitutes.24
Horizontal Demand Curves
The other polar case, the horizontal demand curve, is shown in Figure 3.7. This
is the example of perfectly (or infinitely) elastic demand (eP = ∞). Any
increases in price above P1 in Figure 3.7 would cause the quantity demanded to
decrease to zero, while any price decreases below P1 would cause the quantity
demanded to increase tremendously. This demand curve does not have any exact
applications in reality, although estimates of the price elasticity of demand for
the output of individual farmers are extremely large. Estimated absolute values
of the demand elasticities for individual producers of common fruits and veg-
etables range from 500 to 21,000, with most values greater than 2,000.25 These
values are not infinite in size, but they are extremely large compared with normal
elasticity values.
The perfectly elastic demand curve plays a very important role in economic theory
because it represents the demand curve facing an individual firm in the model of per-
fect competition. In this model, the individual firm is one of a large number of firms
producing a product such that no single firm can influence the price of the prod-
uct. If such a firm tried to raise its price, its quantity demanded would fall to zero.
Perfectly (or infinitely)
elastic demand
Infinite elasticity of demand,
illustrated by a horizontal
demand curve, where the
quantity demanded would vary
tremendously if there were any
changes in price.
23Charles T. Nisbet and Firouz Vakil, “Some Estimates of Price and Expenditure Elasticities of Demand for
Marijuana Among U.C.L.A. Students,” Review of Economics and Statistics 54 (November 1972): 473–75; Jan
C. Van Ours, “The Price Elasticity of Hard Drugs: The Case of Opium in the Dutch East Indies, 1923–1938,”
Journal of Political Economy 103 (1995): 261–79; Dhaval Dave, “Illicit Drug Use Among Arestees, Prices,
and Policy,” Journal of Urban Economics 63 (2008): 694–714; Gary S. Becker, Michael Grossman, and Kevin
M. Murphy, “An Empirical Analysis of Cigarette Addiction,” American Economic Review 84 (June 1994):
396–418; Frank J. Chaloupka and Michael Grossman, Price, Tobacco Control, and Youth Smoking, NBER
Working Paper Series, no. 5740 (Cambridge, MA: National Bureau of Economic Research, 1996); Frank
J. Chaloupka and Henry Wechsler, “Price, Tobacco Control Policies, and Smoking Among Young Adults,”
Journal of Health Economics 16 (June 1997): 359–73; David P. Hopkins, Peter A. Briss, Connie J. Ricard,
Corinne G. Husten, Vilma G. Carande-Kulis, Jonathan E. Fielding, Mary O. Alao et al., “Review of Evidence
Regarding Interventions to Reduce Tobacco Use and Exposure to Environmental Tobacco Smoke,”
American Journal of Preventive Medicine 20 Suppl 1 (2001): 16–66.
24Although the individual demand curve for insulin might be perfectly inelastic, the market demand curve
would have a nonzero elasticity coefficient. For every user of insulin, there is some maximum price they are
willing and able to pay. When this price is exceeded, these users drop out of the market, causing quantity
demanded to vary with price.
25Dennis W. Carlton and Jeffrey M. Perloff, Modern Industrial Organization, 4th ed. (New York: Pearson
Addison-Wesley, 2005).
Demand
0
P1
Q
P FiGurE 3.7
horizontal Demand Curve
A horizontal demand curve
represents perfectly or infinitely
elastic demand.
M03_FARN0095_03_GE_C03.INDD 93 13/08/14 1:41 PM
94 PArt 1 Microeconomic Analysis
Thus, each firm is a price-taker and faces a horizontal demand curve. Individual agri-
cultural producers come close to fitting this definition. That is why the estimated
demand elasticities presented above, while not infinite, are very large in size.
Income and Cross-Price Elasticities of Demand
Although price elasticity of demand is of great importance, managers also need to
know the size of the other elasticities in the demand function for a given product.
Two other common elasticities are the income elasticity and the cross-price elastic-
ity of demand.
Income Elasticity of Demand
The income elasticity of demand shows how consumers change their demand
for a particular product in response to changes in income. The elasticity coefficient
is defined as the percentage change in the quantity demanded of the good relative
to the percentage change in income, holding all other factors constant. This change
in income could be a change for an individual consumer resulting from a raise or
new job, or it could arise from a change in the general level of economic activity in
the overall economy affecting all consumers.
If an increase in income results in an increase in the demand for the good or
if declining income causes consumers to decrease their demand, the good has a
positive income elasticity of demand and is called a normal good. Thus, changes
in income and the demand for normal goods move in the same direction. If an
increase in income results in a decrease in demand or vice versa, the good has a
negative income elasticity and is termed an inferior good. As you’ve learned, this
term has nothing to do with the quality of the product; it simply denotes a negative
income elasticity of demand. Changes in income and the demand for inferior goods
move in opposite directions. Thus, the mathematical sign of the income elasticity
of demand coefficient (positive or negative) is as important as the size of the elas-
ticity coefficient (magnitude of the number). The sign tells a manager whether the
good is normal or inferior, while the size of the coefficient measures the respon-
siveness of the demand to changes in income.
For goods with positive income elasticities, we often make a distinction between
necessities and luxuries. Necessities are defined as goods with an income elastic-
ity between 0 and 1 (0 < eI < 1), while luxuries are defined as goods with an income
elasticity greater than 1 (eI > 1). Consumer spending on necessities does not change
substantially as income changes, whereas spending on luxury goods changes more
than proportionately with changes in income.
Table 3.6 summarizes these concepts. For income elasticity of demand, the percent-
age change in quantity is the change between the two quantities demanded divided
by the base quantity; the same is true for the percentage change in income. As with
price elasticity, income elasticities can be calculated either for discrete changes in
income and quantities (arc elasticity) or for infinitesimal changes (point elasticity).
A study based on scanner data for wine purchases in U.S. retail outlets over
the years 2002 to 2005 estimated that the income elasticities of demand for white
wines (2.996 to 3.003) were twice as large as those for red wines (1.285 to 1.312).
A similar result occurred when wines under and over $10 per bottle were analyzed
separately. When wines were analyzed by both price point and varietal, all income
elasticities were estimated to be positive with Rielsing over $10 per bottle having
a significantly large elasticity value of 9.9. Thus, wines are generally luxury goods
with some fairly large income elasticities.26
income elasticity of
demand
The percentage change in the
quantity demanded of a given
good, X, relative to a percent-
age change in consumer income,
assuming all other factors constant.
necessity
A good with an income elasticity
between 0 and 1, where the expen-
diture on the good increases less
than proportionately with changes
in income.
luxury
A good with an income elasticity
greater than 1, where the expen-
diture on the good increases more
than proportionately with changes
in income.
26Steven S. Cuellar, Tim Colgan, Heather Hunnicutt, and Gabriel Ransom, “The Demand for Wine in the
USA,” International Journal of Wine Business Research 22 (2010): 178–90.
M03_FARN0095_03_GE_C03.INDD 94 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 95
Cross-Price Elasticity of Demand
The cross-price elasticity of demand measures how the demand for one good,
X, varies with changes in the price of another good, Y. The elasticity coefficient is
defined as the percentage change in the quantity demanded of good X relative to
the percentage change in the price of good Y, holding all other factors constant.
Two goods with a positive cross-price elasticity of demand coefficient are said
to be substitute goods. An increase in the price of good Y causes consumers to
demand more of good X because they are substituting good X for good Y. Coffee
and tea are substitute goods, as an increase in the price of coffee will cause some
people to switch to drinking tea. If two goods have a negative cross-price elasticity
of demand coefficient, they are called complementary goods. An increase in the
price of good Y results in a decrease in the demand for good X if the two goods are
used together or are complements. Coffee and cream are complements because an
Cross-price elasticity of
demand
The percentage change in the
quantity demanded of a given
good, X, relative to the percentage
change in the price of good Y, all
other factors held constant.
tAblE 3.6 income Elasticity and Cross-Price Elasticity of Demand Coefficients
ElAStiCity
nAME
ElAStiCity
DEFinition
VAluE oF ElAStiCity
CoEFFiCiEnt

iMPACt on DEMAnD
Income elasticity: eI
%∆QX
%∆l
=
∆QX
QX
∆l
∆l
eI > 0: Normal good
0 < eI < 1: Necessity
eI > 1: Luxury
eI < 0: Inferior good
Increase in income results in increase in demand
Decrease in income results in decrease in demand
Increase in income results in decrease in demand
Decrease in income results in increase in demand
Cross-price %∆QX
%∆PY
∙
∆Qx
Qx
∆PY
∆PY
eC > 0: Substitute good Increase in the price of good Y results in increase in
the demand for good X
Decrease in the price of good Y results in decrease in
the demand for good X
Elasticity: eC eC < 0: Complementary good Increase in the price of good Y results in decrease in
the demand for good X
Decrease in the price of good Y results in increase in
the demand for good X
Managerial rule of thumb
Calculating income Elasticity
The following is a simple rule of thumb for calculating the income elasticity of demand for a product
based on two questions for a consumer:
1. What fraction of your total budget do you spend on Product X?
2. If you earned a bonus of an additional $1,000, what part of that bonus would you spend on Product X?
The ratio of the answer to question 2 to the answer to question 1 is the income elasticity of demand.27
Applying this rule to different products will give managers a quick means of determining how changes
in income will affect the demand for various products. ■
27This example is drawn from Shlomo Maital, Executive Economics (New York: Free Press, 1994), 195.
The answer to question 1 is X/Y, where X is the amount of good X purchased and Y is income. The answer
to question 2 is (ΔX)/(ΔY). The ratio of answer 2 to answer 1 is (ΔX/ΔY)/(X/Y), which can be converted to
(ΔX/X)/(ΔY/Y), the definition for the income elasticity of demand.
M03_FARN0095_03_GE_C03.INDD 95 13/08/14 1:41 PM
96 PArt 1 Microeconomic Analysis
increase in the price of coffee causes people to drink less coffee and, therefore, use
less cream. Goods that have a zero cross-price elasticity of demand are unrelated
in terms of consumption.
Thus, both the mathematical sign and the magnitude or size of the cross-price
elasticity coefficient are important concepts for managers. The sign of the coef-
ficient tells whether the goods are substitutes or complements, and the size of the
cross-price elasticity measures the extent of the relationship between the goods.
These relationships are summarized in the bottom part of Table 3.6.
In 1986, the Federal Trade Commission (FTC) filed suit to block a merger
between the Coca-Cola Company and the Dr. Pepper Company in order to main-
tain competition and, thus, lower prices in the carbonated soft drink market.28 The
size of the relevant market, the number of substitutes, and, therefore, the implied
cross-price elasticities of demand between Coke and other beverages were key
issues in these proceedings. The FTC’s argument that the carbonated soft drink
market was the relevant market was based on evidence that soft drink pricing
and marketing strategies focused on the producers of other soft drinks, not fruit
juices, milk, coffee, tea, or other beverages. Documents indicated that Coke offi-
cials gathered information on the prices and sales of other carbonated soft drink
producers, not producers of other beverages. Although Coca-Cola argued that the
company competed against all other beverages, which were, therefore, actual or
potential substitutes for carbonated soft drinks, the judge in the case ruled for
the FTC and accepted its argument regarding the narrower number of relevant
substitutes.
Estimates of the cross-price elasticity of demand influenced the decision by the
Federal Aviation Administration (FAA) in 2005 not to mandate the use of child
safety seats on commercial airlines. An FAA analysis indicated that if families were
forced to purchase additional seats for infants under two years of age, they might
opt to drive rather than fly. The increase in the price of airline fares could cause
families to substitute relatively risky automobile travel for relatively safe air travel.
It has been estimated that such a mandate would save 0.3 infant lives per year in
the air, but, given a positive cross-price elasticity of 0.356 between the price of air
travel and the quantity of road travel, would result in an additional 11.5 deaths on
the nation’s roads.29
The cross-price elasticity between land- or wireline and wireless telephone ser-
vices has also played a role in the communications industry. Regulation of land-
line services regarding price increases is influenced by the amount of substitution
between these services and other modes of communication. The U.S. Department
of Justice and the Federal Communications Commission have generally been skep-
tical that the availability of wireless phones constrains the ability of landline carri-
ers to raise prices. Because many U.S. households have a landline connection and
at least one wireless telephone, it is also not clear whether these two services are
substitutes or complements.
A study of state data from 2001 to 2007 indicated that a 1 percent increase in the
price of wireline service was estimated to increase the demand for wireless service by
approximately 0.48 to 0.69 percent, while the cross-price elasticity of wireline demand
with respect to the wireless price was estimated to lie between 1.25 and 1.32. The
cross-price elasticity between the demand for cable television and the price of direct
broadcast satellite has been estimated to be between 0.3 and 0.5, so that consumers
appear to view wireless telephones to be at least as interchangeable with wirelines as
28This discussion is based on Lawrence J. White, “Application of the Merger Guidelines: The Proposed
Merger of Coca-Cola and Dr. Pepper (1986),” in The Antitrust Revolution: The Role of Economics, eds. John
E. Kwoka Jr. and Lawrence J. White, 2nd ed. (New York: HarperCollins, 1994), 76–95.
29Shane Sanders, Dennis L. Weisman, and Dong Li, “Child Safety Seats on Commercial Airliners: A
Demonstration of Cross-Price Elasticities,” Journal of Economic Education 39 (Spring 2008): 135–44.
M03_FARN0095_03_GE_C03.INDD 96 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 97
cable television is with direct broadcast satellite service. Thus, wireless telephones
appear to have evolved as a strong substitute for traditional landline service.30
In the previously discussed study of the demand for wine, the cross-price elastic-
ity of white wine for red wine was estimated to be positive but less than one, while
the cross-price elasticity of red wine for white wine was positive and greater than
two. Thus, red wine drinkers are more likely to switch to white wine than white
wine drinkers are to switch to reds. The cross-price elasticities of most wine vari-
etals were positive, indicating substitutability among varietals.31
Elasticity Estimates: Economics Literature
Table 3.7 presents estimates of elasticity of demand coefficients derived in the eco-
nomics literature for various products. These estimates show how elasticities dif-
fer among products, groups of consumers, and over time. Remember that price
elasticity coefficients are reported as negative numbers even though we look at
their absolute values to determine the size of the coefficients.
30Kevin W. Caves, “Quantifying Price-Driven Wireless Substitution in Telephony,” Telecommunications
Policy 35 (2011): 984–98.
31Cuellar et al., “The Demand for Wine in the USA.”
tAblE 3.7 Estimates of Demand Elasticities
(continued)

ProDuCt
PriCE ElAStiCity
CoEFFiCiEnt
inCoME ElAStiCity
CoEFFiCiEnt
CroSS-PriCE ElAStiCity
CoEFFiCiEnt
ChiCken/AgriCulturAl ProduCts/
other Food
Broiler chickens –0.2 to –0.4 +1.0 (1950)
+0.38 (1980s)
+0.20 (beef)
+0.28 (pork)
Cabbage –0.25 N.A.
Potatoes –0.27 +0.15
Eggs –0.43 +0.57
Oranges –0.62 +0.83
Cream –0.69 +1.72
Apples –1.27 +1.32
Fresh tomatoes –2.22 +0.24
Lettuce –2.58 +0.88
Fresh peas –2.83 +1.05
Individual producer
Food away from home
Soft drinks
Juice
Cereals
Milk
Fish
Cheese
Sweets/sugars
–500 to –21,000
–0.81
–0.79
–0.76
–0.60
–0.59
–0.50
–0.44
–0.34
M03_FARN0095_03_GE_C03.INDD 97 13/08/14 1:41 PM
98 PArt 1 Microeconomic Analysis

ProDuCt
PriCE ElAStiCity
CoEFFiCiEnt
inCoME ElAStiCity
CoEFFiCiEnt
CroSS-PriCE ElAStiCity
CoEFFiCiEnt
Beer
Commodity –0.7 to –0.9
72-oz. packages
144-oz. packages
288-oz. packages
–5.07
–5.008
–4.543
WAter demAnd
the toBACCo industry (CigArettes)
–0.33 +0.13
College students –0.906 to –1.309
Secondary school students –0.846 to –1.450
Adults, long-run, permanent
change in price
–0.75
Adults, short-run, permanent
change in price
–0.40
Adults, temporary change in price –0.30
heAlth CAre
Primary care –0.1 to –0.7 0.0 < eI < +1.0
Total/elective surgery –0.14 to –0.17
Physician visits –0.06
Dental care –0.5 to –0.7
Nursing homes –0.73 to –2.40
Inpatient/outpatient hospital
services
N.A. +0.85 to +1.46
Individual physicians –2.80 to –5.07
higher eduCAtion tuition
Any university
Individual university
–0.05 to –0.40
–2.17 to –5.38
Sources: Richard T. Rogers, “Broilers: Differentiating a Commodity,” in Industry Studies, ed. Larry L. Duetsch, 2nd ed. (Armonk, NY: Sharpe, 1998); Tatiana
Andreyeva, Michael W. Long, and Kelly D. Brownell, “The Impact of Food Prices on Consumption: A Systematic Review of Research on the Price Elasticity of Demand
for Food,” American Journal of Public Health 100 (February 2010): 216–22; Daniel B. Suits, “Agriculture,” and Kenneth G. Elzinga, “Beer,” in The Structure of
American Industry, eds. Walter Adams and James W. Brock, 11th ed. (Upper Saddle River, NJ: Prentice Hall, 2005); Dennis W. Carlton and Jeffrey M. Perloff, Modern
Industrial Organization, 4th ed. (New York: Pearson Addison-Wesley, 2005); Jeremy W. Bray, Brett R. Loomis, and Mark Engelen, “You Save Money When You Buy
in Bulk: Does Volume-Based Pricing Cause People to Buy More Beer?” Health Economics 18 (2009): 607–18; Sheila M. Olmstead, W. Michael Hanemann, and Robert
N. Stavins, “Water Demand Under Alternative Price Structures,” Journal of Environmental Economics and Management 54 (2007): 181–98; Frank J. Chaloupka and
Henry Wechsler, “Price, Tobacco Control Policies and Smoking Among Young Adults,” Journal of Health Economics 16 (June 1997): 359–73; Frank J. Chaloupka and
Michael Grossman, Price, Tobacco Control, and Youth Smoking, NBER Working Paper Series, no. 5740 (Cambridge, MA: National Bureau of Economic Research,
1996); Gary S. Becker, Michael Grossman, and Kevin M. Murphy, “An Empirical Analysis of Cigarette Addiction,” American Economic Review 84 (June 1994): 396–418;
Rexford E. Santerre and Stephen P. Neun, Health Economics: Theories, Insights, and Industry Studies, 4th ed. (Mason, OH: Thomson SouthWestern, 2007); Sherman
Folland, Allen C. Goodman, and Miron Stano, The Economics of Health and Health Care, 3rd ed. (Upper Saddle River, NJ: Prentice Hall, 2001); Robert E. Carter and
David J. Curry, “Using Student-Choice Behavior to Estimate Tuition Elasticity in Higher Education,” Journal of Marketing Management 27 (2011): 1186–207.
tAblE 3.7 Continued
Elasticity and Chicken and Agricultural/Food Products
As shown in Table 3.7, broiler chickens have a low price elasticity of demand, as
do many other agricultural products. This low demand elasticity accounts for the
wide swings in the income of farmers, particularly in response to bumper crops,
M03_FARN0095_03_GE_C03.INDD 98 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 99
or large increases in supply. Farm production is subject to many factors outside
producer control, such as the weather and attacks by insects. Crops are grown and
then thrown on the market for whatever price they will bring. If there is a bumper
crop, this increase in supply drives farm product prices down. Because quantity
demanded does not increase proportionately, given the inelastic demand, total rev-
enue to the producers decreases. This is the essence of the “farm problem” that has
confronted U.S. policy makers for many years.32
Table 3.7 shows that not all farm products have inelastic demands, however.
Customers are much more responsive to the price of fresh tomatoes, lettuce, and
fresh peas, with the elasticity of demand exceeding 2.00 in absolute value for these
products. Table 3.7 also shows the difference between the elasticity of demand for
the product as a whole and that for an individual producer of the product. While
the elasticity of demand for many agricultural products is inelastic or less than 1
in absolute value, the elasticities of demand for individual producers are extremely
large, ranging from –500 to –21,000. Farming can be considered a perfectly com-
petitive industry, given the huge elasticities of demand for the individual producers
of farm products. This is why we use the infinitely elastic or horizontal demand
curve to portray the individual firm in perfect competition and the downward slop-
ing demand curve for the output in the entire market.
Agricultural products are generally necessities, with income elasticities less
than 1. However, the larger income elasticities for apples and cream mean that con-
sumption will increase more than proportionately with increases in income. Broiler
chickens have changed from a luxury good in the 1950s to a necessity today, as
evidenced by the decrease in the size of their income elasticity. And, as expected,
chicken is a substitute good with beef and pork, because chicken has a positive
cross-price elasticity of demand with both of these products.
The price elasticity estimates in Table 3.7 for the categories “food away from home”
to “sweets/sugars” were derived from a review of U.S. studies published from 1938
to 2007 with average values of the most recent estimates reported.33 Although all of
these estimates indicated inelastic demand, estimates for “soft drinks” and “juice”
were relatively less inelastic while those for “sweets/sugars” were relatively more
inelastic. “Food away from home” had the highest elasticity among these categories.
Although the average price elasticity estimate for the “milk” category was –0.59,
the study also reported estimates for “skim milk,” 1 percent, and for “whole milk”
ranging from –0.75 to –0.79, whereas the average elasticity for 2 percent milk was
–1.22. Cross-price elasticities indicated that consumers were more likely to switch
to reduced- or low-fat milk than skim milk when the price of whole milk increased.
The authors of the study did not find any studies of price elasticities or cross-price
elasticities that would predict how consumers would shift between healthier and
less healthy food categories, such as whole grain and refined flour breads, brown
and white rice, baked and regular chips, and reduced-fat and regular cheese.
Elasticity and Beer
The price elasticities of demand for beer also differ for the overall commodity and
individual brands. Price elasticity estimates for beer as a commodity are less than 1
in absolute value, whereas estimates for individual brands are reported to be quite
elastic, as there are many more substitutes among brands of beer than for beer as
a product. The package size price elasticities in Table 3.7 represent substitution
32This discussion is drawn from Daniel B. Suits, “Agriculture,” in The Structure of American Industry, eds.
Walter Adams and James Brock, 11th ed. (Upper Saddle River, NJ: Prentice Hall, Inc., 2005), 1–22.
33Tatiana Andreyeva, Michael W. Long, and Kelly D. Brownell, “The Impact of Food Prices on Consumption:
A Systematic Review of Research on the Price Elasticity of Demand for Food,” American Journal of Public
Health 100 (February 2010): 2216–22.
M03_FARN0095_03_GE_C03.INDD 99 13/08/14 1:41 PM
100 PArt 1 Microeconomic Analysis
with a different package size of the same brand of beer or with another brand in
response to price changes, so they would be expected to be large.34 Results of the
same study indicated that consumers are more likely to buy a larger-volume pack-
age in response to a price change than they are to buy a smaller-volume package
or to switch brands. Although this study did not explicitly examine the extent of
substitution from beer to wine or liquor, other research has indicated that many
wine and liquor drinkers will switch to beer when faced with a price increase, but
that few beer drinkers will switch to wine or liquor. They are more likely to switch
package size for their beer.
Water Demand
Because water managers have often argued that consumers do not respond to chang-
ing prices, demand management has often been implemented through restrictions
on water uses or requirements for the adoption of specific technologies.35 However,
a review of studies between 1963 and 1993 indicated that the average price elastic-
ity of demand was –0.51, the short-run median was –0.38, and the long-run median
was –0.64, with 90 percent of all estimates between zero and –0.75. Estimating the
demand for water is complicated because water is typically priced either with a uni-
form marginal price, increasing block prices where the marginal price increases with
the quantity consumed, or decreasing block prices. One study of 11 urban areas in the
United States and Canada that took these pricing strategies into account estimated
a price elasticity of –0.33. This somewhat lower estimate may be related to consum-
ers keeping consumption at a kink point of one of the blocks or reacting to changes
in the block prices. The estimated income elasticity of demand (+0.13) was also low
compared with other studies where results ranged from 0.2 to 0.6. This difference
may have resulted from the statistical estimation techniques used in the study.
Elasticity and the Tobacco Industry
The price elasticity of demand for cigarettes is of interest to the tobacco industry,
state and federal policy makers, and public health advocates.36 Legislators and pub-
lic health advocates have long used cigarette taxation as a policy to attempt to limit
smoking, particularly among teenagers. We noted earlier in the chapter that ciga-
rette price elasticity of demand for adults is inelastic, but not zero. The estimates in
Table 3.7 also show that teenagers and college students have a larger price elastic-
ity of demand for cigarettes than adults. This result is expected for several reasons.
Teenagers are likely to spend a greater proportion of their disposable income on
cigarettes than adults. There are also substantial peer pressure effects operating on
young people. Increased cigarette taxes and prices have a direct negative effect on
consumption, as shown by the elasticity estimates.37 Using taxes to reduce teenage
smoking is an effective policy overall because few people begin smoking after the
age of 20. The tobacco industry has long been aware of these price effects on smok-
ing behavior and has lobbied to limit cigarette tax increases.
34Jeremy W. Bray, Brett R. Loomis, and Mark Engelen, “You Save Money When You Buy in Bulk: Does
Volume-Based Pricing Cause People to Buy More Beer?” Health Economics 18 (2009): 607–18.
35This discussion is based on Sheila M. Olmstead, W. Michael Hanemann, and Robert N. Stavins, “Water
Demand Under Alternative Price Structures,” Journal of Environmental Economics and Management 54
(2007): 181–98.
36This discussion is based on George M. Guess and Paul G. Farnham, Cases in Public Policy Analysis, 3rd
ed. (Washington, DC: Georgetown University Press, 2011).
37David P. Hopkins, Peter A. Briss, Connie J. Ricard, Corinne G. Husten, Vilma G. Carande-Kulis, Jonathan
E. Fielding, Mary O. Alao et al., “Reviews of Evidence Regarding Interventions to Reduce Tobacco Use
and Exposure to Environmental Tobacco Smoke,” American Journal of Preventive Medicine 20 Suppl 1
(2001): 16–66.
M03_FARN0095_03_GE_C03.INDD 100 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 101
The cigarette data also illustrate the differences between behavior in the near
future and behavior over longer periods of time and between temporary and perma-
nent price changes. Consumers are more price sensitive if they believe a price change
is going to be permanent. As noted earlier in the chapter, consumers also have larger
price elasticities over longer periods of time because they are able to search out more
substitutes for the product in question. In response to higher cigarette prices, smok-
ers may also switch to brands with higher tar and nicotine contents, inhale more
deeply, reduce idle burn time, or switch to the use of smokeless tobacco.38
Elasticity and Health Care
The price elasticity estimates for health care are important because arguments are
often made that the demand for these services is medically driven (people “need”
health care). Table 3.7 shows that consumers are price sensitive to medical care
goods and services. Although the demand is relatively inelastic, it is not perfectly
inelastic, as the “needs” argument suggests. As the table shows, the demand for pri-
mary care is more inelastic than the demand for more discretionary services, such
as dental care and nursing homes. The income elasticity of demand for health care
services is generally less than +1.00, indicating that most consumers consider these
services to be necessities. Inpatient and outpatient hospital services are generally
thought to be substitute goods, as shown by the positive cross-price elasticities in
Table 3.7, particularly because there has been a trend to perform many services on
an outpatient basis that previously had been done in the hospital. However, some
studies have derived negative cross-price elasticity estimates, indicating that these
services might be complements in certain cases because some procedures done
in the hospital may require follow-up outpatient visits. This example shows that
economic theory alone may not be able to predict the sign of an elasticity and that
elasticity coefficients need to be estimated from data on consumer behavior.
The differences in health care elasticities for overall primary care (–0.1 to –0.7)
and for services provided by individual physicians (–2.80 to –5.07) again illustrate
the principle that demand can be much more elastic for the individual producer
of a product than for the product in general. Although these differences between
product and individual producer elasticities are not as large as those between agri-
cultural products and their producers, they still indicate that individual physicians
are considered substitutes for one another.
Tuition Elasticity in Higher Education
Tuition pricing decisions in higher education are complex strategies based on pres-
sures to maintain student enrollment; hire and retain faculty, administrators, and ath-
letic coaches; and to maintain and build physical plant.39 University administrators
need to understand student response to tuition pricing decisions. Most studies have
estimated tuition price elasticities ranging from –0.05 to –0.40. However, these stud-
ies were typically based on aggregate or national data that showed student response
in terms of their decision to attend any college or university in their home country.
One study analyzed how students make decisions between a school’s tuition
level and the “non-tuition elements” including available housing, food service,
38Micheal Grossman and Frank J. Chaloupka, “Cigarette Taxes: The Straw to Break the Camel’s Back,”
Public Health Reports 112 (July/August 1997): 291–97; Matthew C. Farrelly, Christina T. Nimsch, Andrew
Hyland, and K. Michael Cummings, “The Effects of Higher Cigarette Prices on Tar and Nicotine Consumption
in a Cohort of Adult Smokers,” Health Economics 13 (2004): 49–58; Jerome Adda and Francesca Cornaglia,
“Taxes, Cigarette Consumption, and Smoking Intensity,” American Economic Review 96 (2006): 1013–28.
39This discussion is based on Robert E. Carter and David J. Curry, “Using Student-Choice Behaviour to Estimate
Tuition Elasticity in Higher Education, “ Journal of Marketing Management 27 (October 2011): 1186–207.
M03_FARN0095_03_GE_C03.INDD 101 13/08/14 1:41 PM
102 PArt 1 Microeconomic Analysis
recreational facilities, and perceptions of campus life. This study used an online
experiment to manipulate tuition levels at multiple universities to see which schools
students at the focal university selected. Researchers estimated tuition elasticities
for each of the 11 colleges in the focal university, which ranged from –2.17 in phar-
macy to –5.38 in business. As expected, these elasticities for individual university
choice were elastic and much larger compared with the aggregate national level
estimates. The study also found that in-state residents were relatively more price
sensitive than nonresidents and that undergraduates were more price sensitive
than graduate students. Elasticity estimates varied substantially by academic divi-
sion, indicating that university administrators could implement differential pricing
policies based on these elasticities.
Managerial rule of thumb
Price Elasticity Decision Making
Which demand elasticity, the one for the entire product or the one for the individual producer, is
appropriate for decision making by the firm? In markets where firms have some degree of market
power, that answer depends on the assumption made about the reaction of other firms to the price
change of a given firm. If all firms change prices together, the product demand elasticity is relevant.
However, if one firm changes price without the other firms following, the larger elasticities for indi-
vidual producers shown in Table 3.7 are appropriate. ■
Elasticity Issues: Marketing Literature
Marketing brings greater detail to the basic economic analysis of price elasticity
by examining such issues as the demand for specific brands of products and the
demand at the level of individual stores. Marketing also takes a somewhat differ-
ent approach to pricing and consumer behavior than economics, where buyers’
responses to price changes are based on the economic model of rational consumer
choice, which is described in Appendix 3A.40 Although behavioral economics,
which relaxes many consumer choice assumptions, is becoming more important in
the economics discipline, marketing has traditionally relied on insights borrowed
from psychology. These include the common belief that consumers treat price as an
indicator of product quality, that buyers tend to perceive price differences in pro-
portional rather than in absolute terms, that buyers perceive prices from left to right
and calculate differences between pairs of prices taking into account only the more
important digits on the left, and that the presentation of prices can affect consum-
ers’ reference prices or those prices that buyers’ consider to be reasonable or fair.
Marketers are also concerned about the size of the price elasticity of demand com-
pared with the advertising elasticity of demand, as both price and advertising are
strategic variables under the control of managers. Changes in product price cause
a movement along a given demand curve, while increases in advertising can cause
changes in consumer preferences and bring new consumers into the market, thus
shifting the entire demand curve. A key issue for managers is which strategy has the
greatest impact on product sales. Table 3.8 presents the results of three major market-
ing studies, all of which we’ll look at in more detail in the remainder of this section.
We will also discuss a more recent study that updates the results of this earlier work.
Advertising elasticity of
demand
The percentage change in the
quantity demanded of a good
relative to the percentage change
in advertising dollars spent on
that good, all other factors held
constant.
40This discussion is based on Thanos Skouras, George J. Avlonitis, and Kostis A. Indounas, “Economics and
Marketing on Pricing: How and Why Do They Differ?” Journal of Product & Brand Management 14 (2005):
362–74.
M03_FARN0095_03_GE_C03.INDD 102 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 103
tAblE 3.8 Elasticity Coefficients from Marketing literature
StuDy ProDuCt PriCE ElAStiCity
CoEFFiCiEnt
ADVErtiSinG ElAStiCity
CoEFFiCiEnt
Tellis (1988) Detergents –2.77
Durable goods –2.03
Food –1.65
Toiletries –1.38
Others –2.26
Pharmaceutical –1.12
Sethuraman and Tellis (1991) All products –1.61 0.11
Durables –2.01 0.23
Nondurables –1.54 0.09
Product life cycle—early –1.10 0.11
Product life cycle—mature –1.72 0.11
Hoch et al. (1995) Soft drinks –3.18
Canned seafood –1.79
Canned soup –1.62
Cookies –1.60
Grahams/saltines –1.01
Snack crackers –0.86
Frozen entrees –0.77
Refrigerated juice –0.74
Dairy cheese –0.72
Frozen juice –0.55
Cereal –0.20
Bottled juice –0.09
Bath tissue –2.42
Laundry detergent –1.58
Fabric softener –0.79
Liquid dish detergent –0.74
Toothpaste –0.45
Paper towels –0.05
Sources: Gerard J. Tellis, “The Price Elasticity of Selective Demand: A Meta-Analysis of Econometric Models of Sales,” Journal of Marketing Research 25 (November
1988): 331–41; Raj Sethuraman and Gerald J. Tellis, “An Analysis of the Tradeoff Between Advertising and Price Discounting,” Journal of Marketing Research 28 (May
1991): 160–74; Stephen J. Hoch, Byung-Do Kim, Alan L. Montgomery, and Peter E. Rossi, “Determinants of Store-Level Price Elasticity,” Journal of Marketing Research
32 (February 1995): 17–29.
Marketing Study I: Tellis (1988)
The first group of elasticities, analyzed by Tellis, is from a meta-analysis or survey
of other econometric studies of selective demand. Selective demand was defined
by Tellis as “demand for a particular firm’s branded product, measured as its sales
M03_FARN0095_03_GE_C03.INDD 103 13/08/14 1:41 PM
104 PArt 1 Microeconomic Analysis
or market share.”41 It differs from the demand for the overall product category,
which is the focus of most of the economic studies of price elasticity in Table 3.7.
The term brand was used generically by Tellis “to cover the individual brand, busi-
ness unit, or firm whose sales or market share is under investigation.”42
Tellis’s study included 367 elasticities from 220 different brands or markets for
the period 1961 to 1985. For all products in his study, Tellis found a mean price
elasticity of –1.76. Therefore, on average for these firms and products, a 1 percent
change in price results in a 1.76 percent change in sales in the opposite direction.
The mean price elasticities for all product groups were also greater than 1 in abso-
lute value. Tellis found that the demand for pharmaceutical products is relatively
more inelastic than those for the other categories, given that safety, effectiveness,
and timing considerations may be more important than price in influencing con-
sumer demand. Pharmaceuticals requiring prescriptions are likely to be covered
by health insurance for many consumers, which would make these individuals less
price sensitive because part of that price is paid by a third party. The results of the
Tellis analysis are important because they are based on numerous empirical stud-
ies of the price elasticity of demand of many different brands of products.
Marketing Study II: Sethuraman and Tellis (1991)
The second group of elasticities in Table 3.8 is derived from a meta-analysis of
16 studies conducted from 1960 to 1988 that estimated both price and advertising
elasticities. This sample was different from the first meta-analysis described in the
table because the studies surveyed in this analysis were required to have estimated
both price and advertising elasticities. There were 262 elasticity estimates repre-
senting more than 130 brands or markets in this survey.
For this sample, Sethuraman and Tellis found a mean price elasticity of –1.609
(rounded to –1.61) and a mean short-term advertising elasticity of +0.109 (rounded
to +0.11). Thus, the ratio of the two elasticities was 14.76, which means that the
size of the average price elasticity is about 15 times the size of the average advertis-
ing elasticity. Pricing is obviously a powerful tool influencing consumer demand.
In Table 3.8, the price elasticity for durable goods is greater than that for non-
durables. However, Sethuraman and Tellis argued that these estimates were not
significantly different, while the differences in advertising elasticity estimates
between durable and nondurable goods (0.23 vs. 0.09) were significant. These
researchers also argued that durable goods might have a lower price elasticity than
nondurables because consumers will pay a higher price for the higher perceived
quality associated with known brands of durable goods, while customers will be
more likely to shop around for a low price on nondurable goods. Advertising elas-
ticities are likely to be higher for durable goods because consumers generally seek
much information before the purchase of these goods.
Marketers have also focused on the stages in a product’s life cycle—Introduction,
Growth, Maturity, and Decline.43 When a new product is introduced, there is usu-
ally a period of slow growth with low or nonexistent profits, given the large fixed
costs often associated with product introduction. In the Growth period, the product
is more widely accepted, and profits increase. When the product reaches Maturity,
sales slow because the product has been accepted by most potential customers,
and profits slow or decline because competition has increased. Finally, in the
period of Decline, both sales and profits decrease. Marketers have hypothesized
41Gerard J. Tellis, “The Price Elasticity of Selective Demand: A Meta-Analysis of Econometric Models of
Sales,” Journal of Marketing Research 25 (November 1988): 331.
42Ibid.
43This discussion is based on Philip Kotler, Marketing Management: The Millennium Edition (Upper
Saddle River, NJ: Prentice-Hall, 2000), 303–16.
M03_FARN0095_03_GE_C03.INDD 104 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 105
that the price elasticity of demand increases over the product life cycle. Consumers
are likely to be better informed and more price conscious about mature products,
and there is likely to be more competition at this stage. Those who adopt a product
early, during the Introduction and Growth stages, are likely to focus more on the
newness of the product rather than its price. These expected differences in elastici-
ties are shown in the estimates in Table 3.8.
Tellis also found that the ratio of the median price elasticity to advertising elas-
ticity is three times higher for the United States than for Europe (19.5 vs. 6.2).44
Thus, consumers may be more price sensitive than advertising sensitive in the
United States compared with Europe. This difference could mean that the level of
advertising is too high in the United States or that there is less opportunity for price
discounting in Europe than in the United States.
Marketing Study III: Hoch et al. (1995)
The third set of elasticities in Table 3.8 comprises store-level price elasticities
estimated from scanner data from Dominick’s Finer Foods, a major chain in the
Chicago metropolitan area. Table 3.8 shows large differences in price elasticities
among product categories. Hoch et al. also found that the elasticities differed by
store location. They analyzed how the elasticities were related to both the charac-
teristics of the consumers in the market area and the overall competitive environ-
ment. Their results are summarized as follows:45
•    More-educated consumers have higher opportunity costs, so they devote less at-
tention to shopping and, therefore, are less price sensitive.
•    Large families spend more of their disposable income on grocery products, and,
therefore, they spend more time shopping to garner their increased returns to
search; they are also more price sensitive.
•    Households with larger, more expensive homes have fewer income constraints,
so they are less price sensitive.
•    Black and Hispanic consumers are more price sensitive.
•    Store volume relative to the competition is important, suggesting that consumers
self-select for location and convenience or price and assortment.
•    Distance from the competition also matters. Isolated stores display less price
sensitivity than stores located close to their competitors. Distance increases
shopping costs.
Marketing Study Update
Bijmolt et al. updated the 1988 study by Tellis in a 2005 meta-analysis, which
included 1,851 price elasticities from 1961 to 2004.46 This study examined brand and
stockkeeping-unit (SKU) elasticities only and excluded sales elasticities, analyzed
elasticities for a single brand or SKU and not averages across items, considered only
price elasticities based on actual purchase or sales data as opposed to experimental
or judgment data, and limited the analysis to business-to-consumer markets.
These researchers found an average price elasticity of –2.62 that was substantially
larger than the value of –1.76 in the Tellis study. This difference could have resulted
44Gerald J. Tellis, Effective Advertising: Understanding When, How, and Why Advertising Works
(Thousand Oaks, CA: Sage Publications, Inc., 2004).
45Stephen J. Hoch, Byung-Do Kim, Alan L. Montgomery, and Peter E. Rossi, “Determinants of Store-Level
Price Elasticity,” Journal of Marketing Research 32 (February 1995): 28.
46Tammo H. A. Bijmolt, Harald J. Van Heerde, and Rik G. M. Pieters, “New Empirical Generalizations on the
Determinants of Price Elasticity,” Journal of Marketing Research XLII (May 2005): 1441–56.
M03_FARN0095_03_GE_C03.INDD 105 13/08/14 1:41 PM
106 PArt 1 Microeconomic Analysis
from differences in sample sizes and estimation methods in the two studies and
from changes in the underlying determinants of elasticities over time. However,
Bijmolt et al. reconfirmed that the price elasticity of demand for durable goods was
greater than for other products. They no longer found significant price elasticity
differences between countries or between estimation methods. They did find that
price elasticities were significantly larger in the Introduction/Growth stage than in
the Mature/Decline stage, a result opposite to that of Tellis. Bijmolt et al. also found
that inflation led to substantially larger price elasticities, especially in the short
run, and that there were differences between the price elasticities for the short-run
promotion of products versus longer-run price changes.
These researchers concluded that consumers seemed to base their purchase tim-
ing and quantity decisions increasingly on price promotions. If price elasticities
were largest in the growth stage of product categories, starting with a low-price
strategy might be the most effective approach for managers. The lack of effect of
brand ownership on price elasticities meant that manufacturing brands were not
necessarily more differentiated than private labels. Finally, the absence of price
elasticity differences by country might mean that similar pricing strategies could
be developed across developed countries. Bijmolt et al. noted that all these results
pertained to bricks-and- mortar selling, and that price elasticities on the Internet
could be significantly different.
Managerial rule of thumb
Elasticities in Marketing and Decision Making
You will most likely pursue these issues raised by the marketing literature in more depth in your marketing
courses. The major point for business students is to recognize the importance of price elasticity of demand
and the linkages between economics and marketing. Managers must be familiar with the fundamentals
of demand and consumer responsiveness to all variables in a demand function because their marketing
departments will build on these concepts to design optimal promotion and pricing strategies. ■
Summary
In this chapter, we explored the concept of elasticity of demand, noting that an elas-
ticity coefficient can be calculated with regard to any variable in a demand function
to determine how consumer demand responds to that variable. We focused most
of our attention on the price elasticity of demand because this concept shows the
relationship between price and revenue changes for the firm. We also showed how
elasticity is a fundamental concept in marketing and serves as the basis for most
pricing and promotion strategies.
We can now see how price elasticity relates to Procter & Gamble’s pricing strate-
gies discussed in the opening case of the chapter. When P&G increased prices in
the first quarter of 2009, sales volume or quantity demanded decreased but sales
revenues increased. Thus, the company faced inelastic demand for its products at
that time. In the next few years when P&G cut prices to regain market share, con-
sumer response and price elasticity were less than expected, so there was a negative
effect on profits. The case discussed varying elasticities among P&G products, where
consumers had more elastic demand for products with more substitutes and greater
competition. We also noted that, in addition to changing its pricing strategy, P&G
attempted to increase demand (shift the demand curve) by developing new products
and entering emerging markets.
M03_FARN0095_03_GE_C03.INDD 106 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 107
In the following appendix, we present the standard economic model of consumer
choice. This model incorporates the concepts of consumer tastes and preferences,
income, and the market prices of goods and services to show how consumer deci-
sions change in the face of changing economic variables. The underlying assumption
is that consumers maximize the utility they receive from consuming different goods
and services subject to a budget constraint that incorporates both income and the
prices of these goods and services. The end products of this model are the consumer
demand curve and its relevant elasticities that we have been studying in this chapter.
Economists have developed a formal model of consumer choice that focuses on
the major factors discussed in this chapter that influence consumer demand: tastes
and preferences, consumer income, and the prices of the goods and services. We
briefly review this model to show how it can be used to derive a consumer demand
curve and to illustrate reactions to changes in income and prices.
Consumer Tastes and Preferences
In this model, we assume that consumers are faced with the choice of different
amounts of two goods, X and Y (although the model can be extended mathemati-
cally to incorporate any number of goods or services). We need to develop a
theoretical construct that reflects consumers’ preferences between these goods.
Consumers derive utility, or satisfaction, from consuming different amounts of
these goods. We use an ordinal measurement of utility in which consumers indi-
cate whether they prefer one bundle of goods to another, but there is no precise
measurement of the change in utility level or how much they prefer one bundle to
another. Ordinal measurement allows our utility levels to be defined as one “being
greater than another,” but not one “being twice as great as another.”
We also make the following assumptions about consumer preference orderings
over different amounts of the goods:
1. Preference orderings are complete. Consumers are able to make compari-
sons between any combinations or bundles of the two goods and to indicate
whether they prefer one bundle to another or whether they are indifferent be-
tween the bundles.
2. More of the goods are preferred to less of the goods (i.e., commodities are “goods”
and not “bads”). Preferences are transitive. If bundle A is preferred to bundle B,
and bundle B is preferred to bundle C, then bundle A is preferred to bundle C.
3. Consumers are selfish. Their preferences depend only on the amount of the
goods they directly consume.
4. The goods are continuously divisible so that consumers can always purchase
one more or one less unit of the goods.
From these assumptions, we develop a consumer’s indifference curve that shows
alternative combinations of the two goods that provide the same level of satisfac-
tion or utility. We show in Figure 3.A1 that such an indifference curve must be
downward sloping if the above assumptions about preferences hold.
In Figure 3.A1, point A represents an initial bundle of goods X and Y, with X1
amount of good X and Y1 amount of good Y. All combinations of the two goods are
represented as points in Figure 3.A1. According to the above preference assump-
tions, the bundle of goods represented by point A must be preferred to any bundle of
Appendix 3A Economic Model of Consumer Choice
M03_FARN0095_03_GE_C03.INDD 107 13/08/14 1:41 PM
108 PArt 1 Microeconomic Analysis
goods in the shaded rectangle to the southwest of point A because point A contains
either more of both goods or more of one good and no less of the other. Likewise,
any bundle of goods in the shaded area northeast of point A must be preferred to the
bundle of goods at point A. We can, therefore, conclude that the indifference curve
through point A must lie in the nonshaded areas of Figure 3.A1. Only in these areas
of the figure will there be other combinations of goods X and Y that provide the con-
sumer with the same satisfaction or utility as that provided at point A.
Figure 3.A2 illustrates such an indifference curve through point A. This indiffer-
ence curve must be downward sloping, given the above discussion. Facing a choice
between the bundle of goods represented by point A (X1, Y1) and point B (X2, Y2),
the consumer is indifferent between these bundles because they both provide the
same level of utility (U1). Other bundles of goods, such as point C (X3, Y3), provide
greater levels of utility because they lie on indifference curves farther from the ori-
gin. Thus, utility levels increase as we move in the direction of the arrow (northeast
from the origin).
Looking at points B and A on indifference curve U1, we can see that if the con-
sumer gives up a certain amount of good Y, he needs an additional amount of good
X to keep the utility level constant. If he gives up the amount of good Y represented
by Y2 – Y1 or ΔY, he needs X1 – X2 or ΔX to keep his utility level constant. The ratio
ΔY/ΔX, which shows the rate at which the consumer is willing to trade off one good
for another and still maintain a constant utility level, is called the marginal rate
of substitution (MRSXY). Mathematically, it is the slope of the indifference curve.
An indifference curve is typically drawn convex to the origin or as shown in
Figures 3.A2 and 3.A3.
The slope of the convex indifference curve in Figure 3.A3 decreases as you move
down the curve. This result implies that an individual has a diminishing marginal
rate of substitution. When the individual is at point A (X1, Y1), with only a small
amount of good X, he is willing to trade off a large amount of good Y, or Y1 – Y2,
to gain an additional amount of X and move to point B (X2, Y2). However, starting
Y1
Y
X1 X
0
A
Y2
Y3
Y1
Y
X2 X1 X3
U1
U2
U3
X
∆X
∆Y
0
A
B
C
FiGurE 3.A1
Derivation of a Consumer
indifference Curve
The indifference curve through point
A must lie in the nonshaded areas of
the quadrant.
FiGurE 3.A2
illustration of Consumer
indifference Curves
The consumer is indifferent between
the bundles of goods, X and Y,
represented by points A and B. These
points lie on the same indifference
curve, U1. Point C represents a bundle
of goods with a greater level of
utility, U3. Consumer preferences are
represented by the marginal rate of
substitution (ΔY/ΔX) or the rate at
which the consumer is willing to trade
off one good for the other.
M03_FARN0095_03_GE_C03.INDD 108 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 109
at point C (X3, Y3), the individual is willing to give up a much smaller amount of
good Y, or Y3 – Y4, to obtain the additional amount of good X and move to point
D (X4, Y4). This diminishing marginal rate of substitution reflects the principle of
diminishing marginal utility. The additional or marginal utility that an individual
derives from another unit of a good decreases as the number of units the individual
already has obtained increases. When an individual has only X1 units of good X, he
is willing to trade off more units of good Y to obtain an additional unit of good X
than when he already has a large amount of good X (X3).
The behavioral assumption behind the consumer choice model is that the con-
sumer wants to maximize the utility derived from the goods and services consumed
(goods X and Y, in this case). However, the consumer is constrained by his level of
income and by the prices he faces for the goods. We need to illustrate the effect of
this constraint and then show how the consumer solves this constrained maximi-
zation problem.
The Budget Constraint
The consumer’s budget constraint is represented by Equation 3.A1:
3.A1 I ∙ PXX ∙ PYY
where
I = level of consumer,s income
PX = price of good X
X = quantity of good X
PY = price of good Y
Y = quantity of good Y
Equation 3.A1 shows that a consumer’s income (I) can be spent either on good X
[(PX)(X)] or on good Y [(PY)(Y)]. For simplicity, we assume that all income is spent
on the two goods.
With given values of I, PX, and PY, we can graph a budget line, as shown in
Figure 3.A4. The budget line shows alternative combinations of the two goods that
can be purchased with a given income and with given prices of the two goods.
The budget line intersects the X-axis at the level of good X that can be pur-
chased (X1) if the consumer spends all his income on good X. The level of income
(I) divided by the price of good X (PX) gives this maximum amount of good X.
Likewise, the budget line intersects the Y-axis at the level of good Y that can be
Y2
Y3
Y4
Y1
Y
X2 X4X1 X3
U1
X0
A
B
C
D
FiGurE 3.A3
A Convex indifference Curve
An indifference curve is typically
drawn convex to the origin,
representing a diminishing marginal
rate of substitution.
M03_FARN0095_03_GE_C03.INDD 109 13/08/14 1:41 PM
110 PArt 1 Microeconomic Analysis
purchased (Y1) if all income is spent on good Y. This amount of good Y is deter-
mined by dividing the level of income (I) by the price of good Y (PY). The slope of
the budget line is distance 0Y1/0X1 = (I/PY)/(I/PX) = PX/PY. Thus, the slope of the
budget line is the ratio of the relative prices of the two goods.
We illustrate a change in income, holding prices constant, in Figure 3.A5. Because
the slope of the budget line is the ratio of the prices of the two goods and because
prices are being held constant, a change in income is represented by a parallel shift
of the budget line. If income increases from level I1 to I2, the budget line shifts out
from B1 to B2, as shown in Figure 3.A5. This increase in income allows the con-
sumer to purchase more of both goods, more of one good and the same amount of
the other, or more of one good and less of the other.
We illustrate a decrease in the price of good X, holding both income and the
price of good Y constant, in Figure 3.A6. The budget line swivels out, pivoting on
the Y-axis. Because the price of good Y has not changed, the maximum quantity of
Y that can be purchased does not change either. However, the price of good X has
decreased, so good X has become cheaper relative to good Y. Budget line B2 has a
flatter slope because the slope of the line is the ratio of the prices of the two goods,
which has changed.
Y1
Y2
Y
X2X1
B2
B1
X0
FiGurE 3.A5
Change (increase) in income (Prices Constant)
A change in income, assuming prices are constant, is
represented by a parallel shift of the budget line.
Y1 = I/PY
Y
X1 = I/PX X0
FiGurE 3.A4
the budget line
The budget line shows alternative combinations of the two
goods, X and Y, that can be purchased with a given income and
with given prices of the goods.
Y1
Y
X2X1
B2
B1
X0
FiGurE 3.A6
Change (Decrease) in the Price of
Good X (All Else Constant)
A decrease in the price of good X is
represented by a swiveling of the
budget line around the intercept on
the Y-axis.
M03_FARN0095_03_GE_C03.INDD 110 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 111
The Consumer Maximization Problem
We illustrate the solution to the consumer problem of maximizing utility subject
to the budget constraint in Figure 3.A7. Point A, with X1 amount of good X and Y1
amount of good Y, is the solution to the consumer maximization problem. This point
gives the consumer the highest level of utility (the indifference curve farthest from
the origin), while still allowing the consumer to purchase the bundle of goods with
the current level of income and relative prices shown in the budget line. Compare
point A with point B (X2 amount of good X and Y2 amount of good Y). Point B lies on
the budget line, so it represents a bundle of goods that the consumer could purchase.
However, it would lie on an indifference curve closer to the origin (not pictured).
This curve would represent a lower level of utility, so the consumer would not be
maximizing the level of utility. Point C, corresponding to X3 amount of good X and
Y3 amount of good Y, lies on the same indifference curve as point A and, therefore,
provides the same level of utility as the goods represented by point A. However,
point C lies outside the current budget line. It is not possible for the consumer to
purchase this bundle of goods with the given income and prices of the goods.
Point A is characterized by the tangency of the indifference curve farthest from the
origin with the budget line. The slopes of two lines are equal at a point of tangency.
The slope of the indifference curve is the marginal rate of substitution between
goods X and Y, while the slope of the budget line is the ratio of the prices of the two
goods. Thus, point A, or consumer equilibrium, occurs where MRSXY = (PX/PY).
Changes in Income
We can now use the concept of consumer equilibrium to illustrate changes in con-
sumer behavior in response to changes in economic variables. Figure 3.A8 shows
an increase in income, all else held constant. The original point of consumer equi-
librium is point A, with X1 amount of good X and Y1 of good Y. The increase in
income is represented by an outward parallel shift of the budget line. To maximize
utility with the new budget line, the consumer now moves to point B and consumes
X2 of good X and Y2 of good Y. In this case, we can see that both X and Y are nor-
mal goods because the consumer increases the quantity demanded of each good in
response to an increase in income, all else held constant.
Figure 3.A9 is a similar example showing an increase in income. However, we
can see in this example that good X is an inferior good. In Figure 3.A9, the original
Y2
Y3
Y1
Y
X2X3 X1
U1
X0
A
B
C
FiGurE 3.A7
the Consumer Maximization Problem
Point A represents the combination of goods where the
consumer maximizes utility subject to the budget constraint.
Y2
Y1
Y
X2X1
U1
U2
X0
A
B
FiGurE 3.A8
Consumer Equilibrium with a Change in income (two
normal Goods)
To maximize utility, the consumer moves from point A to point B
as income increases, consuming more of both goods.
M03_FARN0095_03_GE_C03.INDD 111 13/08/14 1:41 PM
112 PArt 1 Microeconomic Analysis
equilibrium occurs at point A (X1, Y1), while the new equilibrium after the income
increase occurs at point B (X2, Y2). As income increases, the quantity demanded
of good Y increases, but the quantity demanded of good X decreases. Thus, Y is a
normal good in this example, while X is an inferior good.
Changes in Price
Figure 3.A10 shows a decrease in the price of good X, all else held constant. The
original consumer equilibrium in Figure 3.A10 occurs at point A, with X1 of good
X and Y1 of good Y. The decrease in the price of good X, all else held constant, is
represented by the swiveling of the budget line with a new consumer equilibrium
at point B (X2, Y2). Thus, the movement from point A to point B represents a move-
ment along the consumer’s demand curve for good X because a decrease in the
price of good X results in an increase in the quantity demanded.
If we think of good Y as a composite good representing all other goods and services,
we can see in Figure 3.A10 that the decrease in the price of good X caused expendi-
ture on good X to decrease because spending on the composite good Y increased.
Decreased expenditure on good X is equivalent to decreased total revenue for the
producers of good X. Thus, in Figure 3.A10, a decrease in the price of good X results
in a decrease in the total revenue for good X, so that the demand for good X must be
inelastic. We illustrate the opposite case of elastic demand for good X in Figure 3.A11.
In Figure 3.A11, the original consumer equilibrium occurs at point A (X1, Y1),
and the equilibrium after the decrease in the price of good X occurs at point B
(X2, Y2). In this case, the quantity demanded of good X has increased so much that
total spending on the composite good, Y, has decreased, so spending on good X
Y2
Y1
Y
X2 X1
U1
U2
X0
A
B
FiGurE 3.A9
Consumer Equilibrium with a
Change in income (one normal
and one inferior Good)
To maximize utility, the consumer
moves from point A to point B as
income increases, consuming more of
good Y and less of good X.
Y2
Y1
Y
X2X1
U1
U2
X0
A
B
FiGurE 3.A10
Consumer Equilibrium with a
Decrease in the Price of Good X
(inelastic Demand for X)
A decrease in the price of good X
results in an increase in the quantity
demanded but a decrease in the
total revenue for good X, so that the
demand for good X must be inelastic.
M03_FARN0095_03_GE_C03.INDD 112 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 113
Y2
Y1
Y
X2X1
U1
U2
X0
A
B
Key Terms
advertising elasticity of demand,
p. 102
arc price elasticity of demand, p. 86
average revenue, p. 87
average revenue function, p. 87
cross-price elasticity of demand, p. 95
demand elasticity, p. 78
elastic demand, p. 81
income elasticity of demand, p. 94
inelastic demand, p. 81
luxury, p. 86
marginal revenue, p. 88
marginal revenue function, p. 88
necessity, p. 94
perfectly (or infinitely) elastic
demand, p. 93
perfectly inelastic demand, p. 92
point price elasticity of
demand, p. 86
price elasticity of demand (ep), p. 79
total revenue, p. 80
total revenue function, p. 87
unitary elasticity (or unit
elastic), p. 81
Exercises
Technical Questions
1. For each of the following cases, calculate the arc
price elasticity of demand, and state whether de-
mand is elastic, inelastic, or unit elastic.
a. When the price of milk increases from $2.25 to
$2.50 per gallon, the quantity demanded falls
from 100 gallons to 90 gallons.
b. When the price of paperback books falls from
$7.00 to $6.50, the quantity demanded rises
from 100 to 150.
c. When the rent on apartments rises from $500
to $550, the quantity demanded decreases from
1,000 to 950.
has increased. A decrease in the price of good X has resulted in an increase in
expenditure on X and, therefore, in the total revenue to the firm producing X. When
a decrease in price results in an increase in total revenue, demand is elastic.
If we think of good Y as simply another good and not a composite good, Figures
3.A10 and 3.A11 illustrate the cross-price elasticity of demand. In Figure 3.A10, a
decrease in the price of good X results in an increase in the quantity demanded of
good Y. Thus, the two goods in this figure must be complements in consumption
because the cross-price elasticity of demand is negative. The opposite case holds
in Figure 3.A11, where a decrease in the price of good X results in a decrease in the
quantity demanded of good Y. Here goods X and Y are substitute goods with a posi-
tive cross-price elasticity of demand.
FiGurE 3.A11
Consumer Equilibrium with a
Decrease in the Price of Good X
(Elastic Demand for X)
A decrease in the price of good X
results in an increase in the quantity
demanded and in the total revenue
from good X, so that the demand for
good X must be elastic.
M03_FARN0095_03_GE_C03.INDD 113 13/08/14 1:41 PM
114 PArt 1 Microeconomic Analysis
Application Questions
1. In March 2010, Mc Donald’s Corp. announced a
policy to increase summer sales by selling all soft
drinks, no matter the size, for $1.00. The policy
would run for 150 days starting after Memorial
Day. The $1.00 drink prices were a discount from
the suggested price of $1.39 for a large soda. Some
franchisees worried that discounting drinks,
whose sales compensate for discounts on other
products, could hurt overall profits, especially
if customers bought other items from the Dollar
Menu. McDonald’s managers expected this pro-
motion would draw customers from other fast-
food chains and from convenience stores such as
7-Eleven. Additional customers would also help
McDonald’s push its new beverage lineup that in-
cluded smoothies and frappes. Discounted drinks
did cut into McDonald’s coffee sales in previous
years as some customers chose the drinks rather
than pricier espresso beverages. Other chains with
new drink offerings, such as Burger King and Taco
Bell, could face pressure from the $1.00 drinks at
McDonald’s.47
a. Given the change in price for a large soda
from $1.39 to $1.00, how much would quantity
demanded have to increase for McDonald’s rev-
enues to increase? (Use the arc elasticity for-
mula for any percentage change calculations.)
b. What is the sign of the implied cross-price elas-
ticity with drinks from McDonald’s competitors?
c. What are the other benefits and costs to
McDonald’s of this discount drink policy?
2. For each of the following cases, calculate the
point price elasticity of demand, and state whether
demand is elastic, inelastic, or unit elastic. The
demand curve is given by
QD ∙ 5,000 ∙ 50PX
a. The price of the product is $50.
b. The price of the product is $75.
c. The price of the product is $25.
3. For each of the following cases, what is the
expected impact on the total revenue of the firm?
Explain your reasoning.
a. Price elasticity of demand is known to be –0.5,
and the firm raises price by 10 percent.
b. Price elasticity of demand is known to be –2.5,
and the firm lowers price by 5 percent.
c. Price elasticity of demand is known to be –1.0,
and the firm raises price by 1 percent.
d. Price elasticity of demand is known to be 0, and
the firm raises price by 50 percent.
4. The demand curve is given by
QD ∙ 500 ∙ 2PX
a. What is the total revenue function?
b. The marginal revenue function is MR = 250 – Q.
Graph the total revenue function, the demand
curve, and the marginal revenue function.
c. At what price is revenue maximized, and what
is revenue at that point?
d. Identify the elastic and inelastic regions of the
demand curve.
5. You have the following information for your
product:
• The price elasticity of demand is –2.0.
• The income elasticity of demand is 1.5.
• The cross-price elasticity of demand between
your good and a related good is –3.5.
What can you determine about consumer demand
for your product from this information?
6. Suppose each of the following cases increases
your quantity demanded for Good X by 20 percent.
What can you determine about your demand for
Good X from the information?
a. The price of Good X decreases by 22 percent.
b. The price of Good Y increases by 10 percent.
c. Your income increases by 25 percent.
47Paul Ziobro, “McDonald’s Bets Pricing Drinks at $1 Will Heat Up Summer Sales,” Wall Street Journal
(Online), March 18, 2010.
M03_FARN0095_03_GE_C03.INDD 114 13/08/14 1:41 PM
ChAPtEr 3 Demand Elasticities 115
49Kenny Chee, “China’s Xiaomi sells out 5,000 smartphones in 8 minutes, reveals units sold for first time,” The
Straits Times (Online), March 13, 2014.
2. In the second half of 2002, several major U.S. airlines
began running market tests to determine if they
could cut walk-up or unrestricted business fares and
maintain or increase revenues. Continental Airlines
offered an unrestricted fare between Cleveland
and Los Angeles of $716, compared with its usual
$2,000 fare, and found that it earned about the same
revenue as it would have collected with the higher
fare. Making similar changes on its routes from
Cleveland to Houston, Continental found that the
new fare structure yielded less revenue, but greater
market share. On the Houston–Oakland route, the
new fare structure resulted in higher revenue.48
a. What did these test results imply about busi-
ness traveler price elasticity of demand on the
Cleveland–Los Angeles, Cleveland–Houston,
and Houston–Oakland routes for Continental
Airlines?
b. How did these results differ from the discus-
sion of airline elasticity in this chapter?
c. What factors caused these differences?
3. Based on the elasticity data in Table 3.7, discuss
why public health officials generally advocate the
use of cigarette taxes to reduce teenage smoking,
while state and local governments often use these
taxes to raise revenue to fund their services.
4. Xiaomi, a Chinese smartphone company, sold
5,000 of its Redmi phones in Singapore in eight
minutes after its online sale kicked off.49
a. What does the above information tell you about
the price elasticity of demand for Redmi phones
in Singapore?
b. What is the effect of the online sale on the
total revenue received by Xiaomi for selling its
Redmi phones?
c. If more Chinese smartphone makers expand
their business to Singapore, how do you ex-
pect the price elasticity of demand for Xiaomi
phones to change?
5. Develop a case study of a retailer that uses rewards
or loyalty programs to influence demand and price
elasticity of demand for their products. How do
these programs influence both current and future
demand?
6. Suppose Mr. Masaki operates a newspaper stand
in Japan. He sells The Japan Times, an English-
language newspaper published in Japan, at the
same price as all other newspaper stands do.
a. What is the price elasticity of demand for The
Japan Times sold by Mr. Masaki?
b. If we look at the whole market for The Japan
Times, will the price elasticity of demand
remain the same as your answer in question (a)?
7. Find examples in the current business news media
of how eBay and other online sellers obtain infor-
mation about the price elasticity of demand by
making unannounced temporary adjustments to
their prices and fee structures.
8. Suppose you are the manager of The Vila Gale
Fortaleza, a five star hotel in Brazil, and expect-
ing a huge number of tourists during 2014 FIFA
World Cup held from 12 June to 13 July 2014. In
each of the following cases, what pricing deci-
sion should you make in order to make higher
total revenue?
a. Price elasticity of demand is known to be –2.
b. Price elasticity of demand is known to be –0.8.
c. Price elasticity of demand is known to be –1.
48Scott McCartney, “Airlines Try Business-Fare Cuts, Find They Don’t Lose Revenue,” Wall Street Journal,
November 22, 2002.
M03_FARN0095_03_GE_C03.INDD 115 13/08/14 1:41 PM
In this chapter, we explore how both managers and economists use marketing and other consumer data to analyze the factors influencing demand for different products. Many firms, particularly large corpora-tions, hire economists who employ sophisticated statistical or economet-
ric techniques to estimate demand functions or to forecast future demand.
Most managers, however, will not be involved with forecasting and under-
taking the complex statistical analyses performed by business economists.
Managers are more likely to work with marketing and consumer research
departments to profile and understand a company’s customers and to try to
anticipate changes in consumer behavior.
We begin this chapter with a case that focuses on the use of new technolo-
gies to understand consumer purchasing behavior and to target advertising
precisely to different groups of consumers. We then illustrate and evaluate
the techniques managers and marketing departments typically use to obtain
this information. Next we look briefly at how economists use the economet-
ric technique of multiple regression analysis to empirically estimate demand
functions. The goal of this discussion is not to train you as business econ-
omists who produce these statistical analyses, but to help you become bet-
ter consumers of this type of work and to see its usefulness for managerial
decision making. We end the chapter by illustrating the interrelationships
between the marketing/consumer research approach to analyzing consumer
demand, which managers favor, and the formal econometric approach used
by economists.
4
Techniques for Understanding
Consumer Demand and
Behavior
116
M04_FARN0095_03_GE_C04.INDD 116 11/08/14 5:27 PM
Although we focus in this chapter on traditional methods used
by marketers and economists to understand and analyze con-
sumer behavior, companies are increasingly using data from
consumers’ television-viewing behavior and credit card pur-
chases to better understand purchasing decisions and to target
advertising messages based on these decisions.1 Data-gathering
firms match information on consumers’ TV-viewing behavior
with other personal data to help advertisers buy ads targeted to
shows watched by different groups of consumers. In addition,
cable companies are developing new systems designed to show
highly targeted ads based on much more detailed consumer
characteristics.
Traditionally, advertisers bought commercials on shows
that were popular with different age and socioeconomic
groups. New data from cable TV boxes combined with other
household data now allow advertisers to target audience seg-
ments more narrowly, such as Chicago residents shopping for
plane tickets to Los Angeles. Companies such as Simulmedia
obtain information from cable company boxes on when chan-
nels are changed. They then bundle these data into different
groups of viewers, such as “wild n’ crazies” (young males) and
“hecklers” (stand-up comedy), which are then used for adver-
tising campaigns.
Cable companies such as Cablevision Systems Corp. are
also developing systems that can show different commercials
at the same time to different households viewing the same
program. This technology was first used by the U.S. Army to
target varying recruitment ads to different groups of viewers.
The technology anonymously matches names and addresses
of Cablevision’s subscribers with data provided by advertisers.
However, all of these approaches face potential backlash from
consumers or regulatory groups over concerns regarding the
invasion of privacy and the use of personal information.
Visa, Inc. and MasterCard, Inc. use similar approaches to sell
their consumer-purchasing data for use in online advertising.2
These companies own some of the world’s largest databases of
sales transactions, which include 43 billion annual credit and
debit card transactions for Visa and 23 billion for MasterCard.
Given the hundreds of companies tracking consumers’ online
behavior, information, such as gender, age, location, income,
and interests, is now a commodity. Future strategies focus on
merging data about consumer’s online behavior with activities
in real marketplaces. One example would be to target a weight-
loss ad to a person who had just made a purchase at a fast-food
chain and then follow up to see if the individual bought the
advertised products. MasterCard proposed such a strategy in
2011, but then tabled the idea given concerns and regulations
on the use of consumer data by financial-services companies.
All of these strategies incorporate inherent conflicts between
technologies that allow the collection and sale of new, detailed
personal information on consumer behavior for profit-making
activities and concerns over the invasion of privacy.
Retailer Target also faced this dilemma as it developed
statistical methods to be able to identify women who were
pregnant before other retailers knew the baby was on the
way.3 Target marketers learned from neuroscience research
that it is difficult to change consumer buying habits once
they are ingrained. However, there are brief periods, such as
around the birth of a child, when old routines fall apart and
buying habits are in flux. Using data from its baby shower
registry, Target researchers analyzed purchases of 25 prod-
ucts that allowed them to develop a “pregnancy prediction”
score and to estimate a woman’s due date. This allowed the
company to send coupons timed to specific stages of a preg-
nancy with the expectation that these women would purchase
other items as well as baby products and become regular
Target customers. Because this approach could easily be con-
sidered an invasion of women’s privacy, Target began mixing
coupons for baby products with advertisements for all types
of products that pregnant women would never buy. They
found that women would use the coupons as long as the baby
ads looked random.
Case for Analysis
The Use of New Technology to Understand and Impact Consumer Behavior
117
1This discussion is based on Jessica E. Vascellaro, “TV’s Next
Wave: Tuning in to You,” Wall Street Journal (Online),
March 7, 2011.
2This discussion is based on Emily Steel, “Using Credit Cards
to Target Web Ads,” Wall Street Journal (Online), October 25,
2011.
3Charles Duhigg, “How Companies Learn Your Secrets,” The
New York Times (Online), February 16, 2012.
M04_FARN0095_03_GE_C04.INDD 117 11/08/14 5:27 PM
118 PArT 1 Microeconomic Analysis
Understanding Consumer Demand and
Behavior: Marketing Approaches
The opening case illustrates current and proposed uses of new technology to gather
additional data about consumer behavior and to profit from the sale of these data.
These new approaches build on methods that marketing departments have tradi-
tionally used to analyze consumer behavior such as
1. Expert opinion
2. Consumer surveys
3. Test marketing and price experiments
4. Analyses of census and other historical data
5. Unconventional methods
Much of the discussion of these nonstatistical approaches to learning about demand
and consumer behavior is found in the marketing literature.4 We briefly summarize
this literature—which is usually covered more extensively in marketing courses—
and then relate these approaches to statistical or econometric demand estimation,
the approach used most often by economists.
Analyzing demand and consumer behavior involves the study of what people say,
what they do, or what they have done.5 Surveys of consumers, panels of experts,
or the sales force working in the field can provide information on how people say
they behave. Test marketing and price experiments focus on what people actually
do in a market situation. Analyses of census and other historical data and statisti-
cal or econometric demand estimation are based on data showing how consumers
behaved in the past. These studies then use that behavior as the basis for predicting
future demand. Some of these techniques are implemented in an uncontrolled mar-
keting environment, whereas others are used in a controlled research framework.
Let’s start by looking at the role of expert opinion.
Expert Opinion
Sales personnel or other experts, such as dealers, distributors, suppliers, market-
ing consultants, and members of trade associations, may be interviewed for their
expert opinion on consumer behavior. At least 10 experts from different func-
tions and hierarchical levels in the organization should be involved with making an
expert judgment on a particular product. For example, large appliance companies
and automobile producers often survey their dealers for estimates of short-term
demand for their products. This approach is especially useful in multiproduct situ-
ations, where other strategies may be prohibitively expensive.
The inherent biases of this approach are obvious, as sales personnel and oth-
ers closely related to the industry may have strong incentives to overstate con-
sumer interest in a product. These individuals may also have a limited view of the
entire set of factors influencing product demand, particularly factors related to the
Expert opinion
An approach to analyzing consumer
behavior that relies on developing
a consensus of opinion among sales
personnel, dealers, distributors,
marketing consultants, and trade
association members.
4Kent B. Monroe, Pricing: Making Profitable Decisions, 2nd ed. (New York: McGraw-Hill, 1990); Robert J.
Dolan and Hermann Simon, Power Pricing: How Managing Price Transforms the Bottom Line (New York:
Free Press, 1996); Financial Times, Mastering Marketing: The Complete MBA Companion in Marketing
(London: Pearson Education, 1999); Philip Kotler, Marketing Management: The Millennium Edition
(Upper Saddle River, NJ: Prentice Hall, 2000); Vithala R. Rao (ed.), Handbook of Pricing Research in
Marketing (Cheltenham, UK: Edward Elgar, 2009); Thomas T. Nagle, John E. Hogan, and Joseph Zale, The
Strategy and Tactics of Pricing, 5th ed. (Upper Saddle River, NJ: Prentice-Hall, 2011).
5The following discussion of the marketing approaches used to understand consumer behavior is based
largely on Kotler, Marketing Management: The Millennium Edition; Dolan and Simon, Power Pricing:
How Managing Price Transforms the Bottom Line; Rao (ed.), Handbook of Pricing Research in
Marketing; and Nagle, Hogan, and Zale, The Strategy and Tactics of Pricing.
M04_FARN0095_03_GE_C04.INDD 118 11/08/14 5:27 PM
ChAPTEr 4 Techniques for Understanding Consumer Demand and Behavior 119
overall level of activity in the economy. Therefore, this approach works better in
business-to-business markets, where there are fewer customers and where experts
are likely to know the markets well.
Consumer Surveys
Consumer surveys include both direct surveys of consumer reactions to prices and
price changes, and conjoint analyses of product characteristics and prices.
In direct consumer surveys, consumers are asked how they would respond
to certain prices, price changes, or price differentials. Questions may include the
following:
•    At what price would you definitely purchase this product?
•    How much are you willing to pay for the product?
•    How likely are you to purchase this product at a price of $XX?
•    What price difference would cause you to switch from Product X to Product Y?6
These surveys are easily understood and less costly to implement than other
approaches to analyzing consumer behavior. Surveys have the greatest value when
there are a relatively small number of buyers who have well-defined preferences
that they are willing to disclose in a survey format. Surveys are most useful for new
products, industrial products, and consumer durables that have a long life, as well
as for products whose purchase requires advanced planning. In these surveys, mar-
ket researchers may also collect information on consumer personal finances and
consumer expectations about the economy.
Limitations to this approach include the issue of whether consumer responses to
the questions reflect their actual behavior in the marketplace. This problem is par-
ticularly important regarding reactions to changes in prices. Can consumers know
and accurately respond to questions on how they would behave when facing dif-
ferent prices for various products? Surveys also tend to focus on the issue of price
in isolation from other factors that influence behavior. There may be response
biases with this approach because interviewees may be reluctant to admit that they
will not pay a certain price or that they would rather purchase a cheaper product.
Surveys also typically ask for a person’s response at a time when they are not actu-
ally making the purchase, so they may not give much thought to their answer.
Consumer surveys are not always successful in obtaining accurate information.
In one survey conducted by a major hotel chain that covered all aspects of the
hotel’s operations, including the prices, guests were asked what price was consid-
ered too high, as well as what the highest acceptable price was.7 The results of this
survey indicated that the hotel chain’s prices in various cities were about as high
as business guests would pay. Managers realized there was a bias in the survey
because respondents were also asked what price they were currently paying for
the hotel rooms. Respondents were unwilling to tell the hotel management that
they would have paid more than the rates they were currently being charged. Thus,
the survey biased the results on willingness to pay toward the current hotel rates.
A more sophisticated form of consumer survey is conjoint analysis, which has
been used in the pricing and design of products ranging from computer hardware
and software to hotels, clothing, automobiles, and information services. This tech-
nique is used in a more controlled research environment compared with inter-
views, which are more unstructured. In this approach, a consumer is faced with an
Direct consumer surveys
An approach to analyzing consumer
behavior that relies on directly
asking consumers questions about
their response to prices, price
changes, or price differentials.
Conjoint analysis
An approach to analyzing consumer
behavior that asks consumers to
rank and choose among different
product attributes, including price,
to reveal their valuation of these
characteristics.
6An application of a similar approach and an extensive discussion of the pricing of services are found
in Stowe Shoemaker and Anna S. Mattila, “Pricing in Services,” in Handbook of Pricing Research in
Marketing, ed. Vithala R. Rao (Cheltenham, UK: Edward Elgar, 2009), 535–56.
7This example is drawn from Monroe, Pricing: Making Profitable Decisions, 107–8.
M04_FARN0095_03_GE_C04.INDD 119 11/08/14 5:27 PM
120 PArT 1 Microeconomic Analysis
array of products that have different attributes and prices and is asked to rank and
choose among them. The analysis allows the marketer to determine the relative
importance of each attribute to the consumer. Conjoint analysis does not directly
measure stated purchase intentions, but it focuses on the preferences that underlie
those intentions. This technique allows marketers to identify consumer segments
for which consumers have a different willingness to pay and varying price elastici-
ties. It also gives researchers the ability to check to see if consumers’ responses are
at least consistent. Computer interviewing has become a standard procedure for
conjoint analysis.8
The advantage of conjoint analysis is that it presents the consumer with a real-
istic set of choices among both product characteristics and prices. For example,
in the case of a new automobile, managers might develop an analysis that focuses
on attributes such as brand, engine power, fuel consumption, environmental per-
formance, and price. Different levels of each of the attributes are presented to the
consumer. Comparisons are set up where the consumer has to make a trade-off
between different characteristics. Thus, his or her choices reveal information about
consumer preferences for the product characteristics.9 Note that conjoint analysis
employs an approach to consumer behavior that is similar to the economic indiffer-
ence curve model described in the appendix to Chapter 3 of this text.
Test Marketing and Price Experiments
Test marketing and price experiments are particularly important for analyzing
consumer reaction to new products. Test marketing allows companies to study
how consumers handle, use, and repurchase a product, and it provides information
on the potential size of the market. In sales-wave research, consumers who are
initially offered the product at no cost are then reoffered the product at different
prices to determine their responses. Simulated test marketing involves selecting
shoppers who are questioned about brand familiarity, invited to screen commer-
cials about well-known and new products, and then given money to purchase both
the new and the existing products. Full-scale test marketing usually occurs over
a period of a few months to a year in a number of cities and is accompanied by
a complete advertising and promotion campaign. Marketers must determine the
number and types of cities for the testing and the type of information to be col-
lected. Information on consumer behavior in the test cities is gathered from store
audits, consumer panels, and buyer surveys.
Price experiments to determine the effect of changes in prices may be con-
ducted in test market cities or in a laboratory setting. Direct mail catalogs can
also be used for these experiments, as prices can be varied in the catalogs shipped
to different regions of the country without a high level of consumer awareness.
Although testing in a laboratory situation helps to control for other factors influenc-
ing consumer behavior, the disadvantage of this approach is that it is not a natural
shopping environment, so consumers may behave differently in the experimental
environment. Doing an experiment in an actual test market may be more realistic,
but it raises problems about controlling the influence on consumer behavior of
variables other than price.10
Test marketing
An approach to analyzing consumer
behavior that involves analyzing
consumer response to products in
real or simulated markets.
Price experiments
An approach to analyzing consumer
behavior in which consumer
reaction to different prices is
analyzed in a laboratory situation or
a test market environment.
8For an illustration of the use of conjoint analysis by a small sporting goods manufacturer, see Box 12-4, “A
Conjoint Study: Power Powder Ski,” in Nagle, Hogan, and Zale, The Strategy and Tactics of Pricing, 5th ed.,
291–93.
9More details of this approach to analyzing consumer behavior are presented in Kotler, Marketing
Management: The Millennium Edition, 339–40; Dolan and Simon, Power Pricing: How Managing Price
Transforms the Bottom Line, 55–69; and Nagle, Hogan, and Zale, The Strategy and Tactics of Pricing,
269–304.
10Additional discussion of these approaches is provided by Kamel Jedidi and Sharan Jagpal, “Willingness
to Pay: Measurement and Managerial Implications,” in Handbook of Pricing Research in Marketing, ed.
Vithala R. Rao (Cheltenham, UK: Edward Elgar, 2009), 37–60.
M04_FARN0095_03_GE_C04.INDD 120 11/08/14 5:27 PM
ChAPTEr 4 Techniques for Understanding Consumer Demand and Behavior 121
Analysis of Census and Other Historical Data
The most recent U.S. census is always a vital marketing tool, given the develop-
ment of targeted marketing, which defines various market segments or groups
of buyers for particular products based on the demographic, psychological, and
behavioral characteristics of these individuals. For example, Sodexho Marriott
Services, a provider of food services to universities and other institutions, analyzed
census data to develop menu programs specifically designed for students on par-
ticular campuses. Starbucks Corporation has used complex software algorithms to
analyze both census and historical sales data in order to obtain a positive or nega-
tive response to every address considered as a potential store site.11
Companies such as Claritas of San Diego have provided consulting services on
how to use census data to develop marketing plans and analyze consumer behav-
ior. Using census data on age, race, and median income and other survey lifestyle
information, such as magazine and sports preferences, Claritas developed 62 clus-
ters or consumer types for targeted marketing. Hyundai Corporation used these
data and buyer profiles to determine which of its models will appeal to consumers
in different parts of a community and to plan the locations of new dealerships.
Hyundai also used cluster data to send test-drive offers to certain neighborhoods,
instead of entire cities, and reported that it cut its cost per vehicle sold in half as a
result of this targeting strategy.12
Many companies have used census and survey data to develop new marketing
strategies focused on older customers, a population segment that has often been
overlooked in the past.13 A study by AARP, the advocacy group for people over 50,
found that, for many products, the majority of people over 45 were not loyal to a
single brand. Companies have begun to realize that this segment of the population
is increasing in size, may have more discretionary income than younger groups,
and may respond to appropriate advertising.
More companies are also using retail store scanner data that provide almost
immediate information on the sales of different products. Scanner data are much
less costly to obtain than panel data from research companies that collect data
from panels of a few thousand households. Scanner data have become the major
source of information on the price sensitivity of consumer-packaged goods.14
Unconventional Methods
In May 2001, Procter & Gamble announced plans to send video crews with cam-
eras into 80 households around the world to record the daily routines of the occu-
pants.15 The company anticipated that this approach would yield better and more
useful data than the consumer research methods discussed above because con-
sumer behavior would be directly observed in a household setting rather than in an
experimental environment. This approach would also avoid the response bias that
can be present in a consumer survey. More recently, Procter & Gamble began part-
nering with Google Inc. to develop strategies appealing to online customers. The
company invited “mommy bloggers”—women who run popular Web sites about
child rearing—to tour the baby division to increase awareness and obtain feedback
on Procter & Gamble products.16
Targeted marketing
Selling that centers on defining
different market segments or
groups of buyers for particular
products based on the
demographic, psychological, and
behavioral characteristics of the
individuals.
11Amy Merrick, “New Population Data Will Help Marketers Pitch Their Products,” Wall Street Journal,
February 14, 2001.
12Ibid.
13Kelly Greene, “Marketing Surprise: Older Consumers Buy Stuff, Too,” Wall Street Journal, April 6, 2004.
15Emily Nelson, “P & G Plans to Visit People’s Homes to Record (Almost) All Their Habits,” Wall Street
Journal, May 17, 2001.
16Ellen Byron, “A New Odd Couple: Google, P&G Swap Workers to Spur Innovation,” Wall Street Journal,
November 19, 2008.
14Nagle, Hogan, and Zale, The Strategy and Tactics of Pricing, 271–76.
M04_FARN0095_03_GE_C04.INDD 121 11/08/14 5:27 PM
122 PArT 1 Microeconomic Analysis
To try to halt a 30-year decline in the sales of white bread, Sara Lee Corp. used
taste testers to try to determine the optimum amount of whole grains that could
be introduced into white bread to provide additional health benefits without
discouraging users of traditional white bread who preferred that taste and con-
sistency to darker and grainier whole grain breads.17 The result, a bread called
Soft & Smooth with 70 calories and 2 grams of fiber per slice, became one of the
best-selling brands in 2006 and encouraged other companies to produce similar
breads.
Retailers have found that Western-style supermarkets with clean, wide aisles
and well-stocked shelves do not work well in India, particularly for lower-middle-
class shoppers who are more comfortable in tiny, cramped stores.18 One retailer,
Pantaloon Retail (India) Ltd., spent $50,000 in a store to replace long, wide
aisles with narrow, crooked ones to make the store messier, noisier, and more
cramped. Products were clustered on low shelves and in bins because customers
were used to shopping from stalls and finding products such as wheat, rice, and
lentils in open containers that they could handle and inspect. Pantaloon Retail
(India) Ltd. eventually became India’s largest retailer in 2007 with annual sales of
$875 million.
The case that opened this chapter illustrated how companies are now using new
technologies to obtain data from cable television subscribers and credit card users
to gather ever more detailed information on consumer buying habits and to link
online and real-world behavior.
Evaluating the Methods
All research on consumer behavior involves extrapolating from a sample of data
to a larger population. When designing surveys and forming focus groups for
interviews, marketers must be careful that participating individuals are represen-
tative of the larger population. For example, Procter & Gamble does 40 percent
of its early exploratory studies of new products in its hometown of Cincinnati.
Given the number of P&G employees and retirees in the area, the company needs
to make certain that its focus group members do not have an undue positive
bias toward the products or are not influenced by family connections with the
company.19
As mentioned earlier, responses given in an experimental format may not reflect
actual consumer behavior. Although economists have recently used more experi-
mental techniques, for many years these researchers preferred to rely on mar-
ket data that showed how consumers actually behaved, not how they said they
would behave. Surveys and focus groups must also be designed to determine the
independent effect of each of the demand function variables on product sales or
quantity demanded. Finding simple correlations between market variables and
product sales does not mean that these variables have the same effect with other
variables held constant. Nagle et al. have argued that managerial judgment should
always be used in combination with any of these measurement techniques. For
example, if managers know that 80 percent of the company’s consumers for a par-
ticular product are women who are employed full time, that information should
be used to determine how any in-home survey or shopping center experiment is
undertaken.20
17Steven Gray, “How Sara Lee Spun White, Grain into Gold,” Wall Street Journal, April 25, 2006.
18Eric Bellman, “In India, a Retailer Finds Key to Success Is Clutter,” Wall Street Journal (Online), August
8, 2007.
19Emily Nelson, “P & G Keeps Focus Groupies of Cincinnati Busy as Guinea Pigs in Product Studies,” Wall
Street Journal, January 24, 2002.
20Nagle, Hogan, and Joseph, The Strategy and Tactics of Pricing, 294–300.
M04_FARN0095_03_GE_C04.INDD 122 11/08/14 5:27 PM
ChAPTEr 4 Techniques for Understanding Consumer Demand and Behavior 123
Consumer Demand and Behavior:
Economic Approaches
As we mentioned in the introduction to this chapter, many companies hire business
economists to develop quantitative estimates of the relationships among the vari-
ables influencing the demand for their products. Results of a survey of 538 compa-
nies employing 4 to over 380,000 employees published in the mid-1990s indicated
that 37.4 percent of the companies had economics departments.21
Economists typically use the statistical technique of multiple regression anal-
ysis to estimate the effect of each relevant independent variable on the quantity
demanded of a product, while statistically holding constant the effects of all other
independent variables. This approach involves the analysis of historical data to
develop relationships among the variables and to predict how changes in these
variables will affect consumer demand.
In the physical sciences, many of these types of relationships can be tested experi-
mentally in the laboratory. However, experiments in the social and policy sciences
are often very expensive, time-consuming, and complex to perform. Although the use
of experimental approaches has been increasing in different areas of economics,22
most product demand research still relies on statistical or econometric techniques,
such as multiple regression analysis, to examine the relationship between two vari-
ables, while statistically holding constant the effects of all other variables.
In the remainder of this section, we present an introduction to the use of mul-
tiple regression analysis and references for further study of the topic. We begin by
focusing on a case involving one dependent and one independent variable, which
we illustrate with a Microsoft Excel spreadsheet. We then move to an Excel case
involving two independent variables to show how additional variables can mod-
ify the results of an analysis. Although both Excel examples are too simplistic for
real-world market analysis, they illustrate the basic principles of the econometric
approach to demand estimation. We next present a discussion of how regression
analysis has been used to examine the factors influencing the demand for automo-
biles. In the last section of the chapter, we discuss the relationship between the
consumer research data that managers and marketers use and the statistical analy-
sis of consumer behavior that economists undertake.23
Multiple regression
analysis
A statistical technique used to
estimate the relationship between
a dependent variable and an
independent variable, holding
constant the effects of all other
independent variables.
Managerial rule of Thumb
Marketing Methods for Analyzing Consumer Behavior
When using expert opinion, consumer surveys, test marketing, and price experiments to analyze con-
sumer behavior, managers must consider the following points:
1. Whether the participating groups are truly representative of the larger population?
2. Whether the answers given in these formats represent actual market behavior?
3. How to isolate the effects of different variables that influence demand? ■
21John J. Casson, “The Role of Business Economists in Business Planning,” Business Economics 31
(July 1996): 45–50. Information about current jobs for business economists can be found at the National
Association for Business Economics Career Center, http://nabe.com/careers/index.html.
22John A. List, “Why Economists Should Conduct Field Experiments and 14 Tips for Pulling One Off,”
Journal of Economic Perspectives 25 (Summer 2011): 3–16.
23Economists also use models to forecast the future values of economic variables based on trends in these
values over time. We do not include these techniques in this book.
M04_FARN0095_03_GE_C04.INDD 123 11/08/14 5:27 PM
124 PArT 1 Microeconomic Analysis
Relationship Between One Dependent and One
Independent Variable: Simple Regression Analysis
Let’s begin with a very simple hypothetical example of a demand function. Suppose
that a manager has a sample of data on price and quantity demanded for oranges,
shown in Figure 4.1 and in the bottom part of Table 4.1 [Actual Q (lbs.), Actual
P (cents/lb.)].24 These data could be either cross-sectional data or time-series
data. If the data are cross-sectional data, they represent the behavior of differ-
ent individuals facing different prices for oranges at a specific point in time. If the
data are time-series data, they represent a set of observations on the same obser-
vational unit at a number of points in time, usually measured annually, quarterly,
or monthly. Many recent studies use panel data sets, which are based on the same
cross-sectional data observed at several points in time.25
If we want to estimate the relationship between quantity demanded and price, we
can first just examine the data points in Figure 4.1 and Table 4.1. These data points
show what appears to be a negative relationship between the variables—that is,
as price decreases, quantity demanded increases, or as price increases, quantity
demanded decreases.
Quantitative Measure Most managers need more information about this re-
lationship than can be inferred just by examining the raw data in Figure 4.1 and
Table 4.1. Managers want a quantitative measure of the size of this relationship
that shows how much quantity demanded will change as price either increases or
decreases. One quantitative measure would be to draw the line that best reflects
the relationship shown by the data points in Figure 4.1 and Table 4.1. We would
like to draw a straight line indicating a linear relationship between the variables
because a linear relationship is the easiest case to analyze. However, we can see
in Figure 4.1 that all the data points will not fall on a single straight line. For exam-
ple, at a price of 70 cents per pound, four individuals demand different quantities.
Thus, there is variation in the data, which means that some data points will deviate
from any line fitted to the data. We want to find the line that “best fits” the data.
As with any straight line, a linear demand relationship can be expressed in an
equation, as shown in Equation 4.1.
4.1 Q = a − bP
where
Q = quantity demanded
a = vertical intercept
b = slope of the line = ∆Q/∆P
P = price
Cross-sectional data
Data collected on a sample
of individuals with different
characteristics at a specific point
in time.
Time-series data
Data collected on the same
observational unit at a number of
points in time.
Panel data
Cross-sectional data observed at
several points in time.
24This example is drawn from Jan Kmenta, Elements of Econometrics (New York: Macmillan, 1971). See
that text for a complete derivation and discussion of the statistical procedures and results.
25William H. Greene, Econometric Analysis, 7th ed. (Upper Saddle River, NJ: Prentice Hall, 2012).
50 100
Quantity (lbs.)
P
ri
ce
(
ce
nt
s/
lb
.)
1500
0
50
100
150FIgUrE 4.1
hypothetical Demand for
Oranges
This figure plots the sample data of
the demand for oranges showing
price (cents per lb.) and quantity
demanded (lbs.).
M04_FARN0095_03_GE_C04.INDD 124 11/08/14 5:27 PM
ChAPTEr 4 Techniques for Understanding Consumer Demand and Behavior 125
TABlE 4.1 Simple regression Analysis results
rEgrESSION STATISTICS
Multiple R 0.943
R square 0.889
Adjusted R square 0.878
Standard error 8.360
Observations 12.000
ANAlYSIS OF VArIANCE (ANOVA)
Degrees of
Freedom
Sum of
Squares
Mean
Square

F-statistic
Significance of
F-statistic
Regression 1.000 5601.111 5601.111 80.143 0.000
Residual 10.000 698.889 69.889
Total 11.000 6300.000

Coefficients

Standard Error

t-statistic

P-value
lower
95 percent
Upper
95 percent
Intercept 210.444 12.571 16.741 0.00000001 182.435 238.454
Price –1.578 0.176 –8.952 0.00000434 –1.970 –1.185
rESIDUAl OUTPUT

Observation

Predicted Q

residuals
Actual Q
(lbs.)
Actual P
(cents/lb.)
1 52.667 2.333 55 100
2 68.444 1.556 70 90
3 84.222 5.778 90 80
4 100.000 0.000 100 70
5 100.000 –10.000 90 70
6 100.000 5.000 105 70
7 100.000 –20.000 80 70
8 107.889 2.111 110 65
9 115.778 9.222 125 60
10 115.778 –0.778 115 60
11 123.667 6.333 130 55
12 131.556 –1.556 130 50
The vertical intercept, a, represents the quantity demanded of the product
as a result of other variables that influence behavior that are not analyzed in
Equation 4.1. For example, the quantity demanded may be influenced by an indi-
vidual’s income or by the size of the firm’s advertising budget. The slope parameter,
b, shows the change in quantity demanded that results from a unit change in price.
We have assumed that b is a negative number, given the usual inverse relationship
between price and quantity demanded. Once we know the parameters, a and b,
we know the specific relationship between price and quantity demanded shown in
Table 4.1, and we know how quantity demanded changes as price changes.
M04_FARN0095_03_GE_C04.INDD 125 11/08/14 5:27 PM
126 PArT 1 Microeconomic Analysis
Simple Regression Analysis The relationship between price and quantity
demanded can be estimated in this hypothetical example using simple regres-
sion analysis, as there is only one independent variable (P) and one dependent
variable (Q). Regression analysis, as noted above, is a statistical technique that
provides an equation for the line that “best fits” the data. “Best fit” means minimiz-
ing the sum of the squared deviations of the sample data points from their mean
or average value.26 Most of the actual data points will not lie on the estimated
regression line due to the variation in consumer behavior and the influence of
variables not included in Equation 4.1. However, the estimated line captures the
relationship between the variables expressed in the sample data.
To estimate Equation 4.1, a manager needs to collect data on price and quantity
demanded for a sample of individuals, as represented by the data points in Figure
4.1 and Table 4.1. The manager can then use any standard statistical software pack-
age to estimate the regression parameters, coefficients a and b in Equation 4.1, for
that sample of data.27 The computer program estimates the parameters of the equa-
tion and provides various summary statistics.
The results of such an estimation process, using the Excel regression feature, are
shown in the middle rows of Table 4.1. The estimated value of the intercept term
is 210.444, while the estimated value of the price coefficient is –1.578. The demand
relationship is shown in Equation 4.2:
4.2 Q = 210.444 − 1.578P
The price coefficient, –1.578, shows that the quantity demanded of oranges
decreases by 1.578 pounds for every one cent increase in price. However, both
economists and managers are usually more interested in the price elasticity of
demand than in the absolute changes in quantity and price. Because this is a lin-
ear demand curve, the price elasticity varies along the demand curve. Equation 4.3
shows the calculation of price elasticity at the average price (70 cents per pound)
and average quantity demanded at that price (100 pounds), based on Equation 4.2.
4.3 eP =
(∆Q)(P)
(∆P)(Q)
=
(-1.578)(70)
(100)
= − 1.105
Equation 4.3 shows that the demand for oranges is slightly price elastic using the
average values of the data in this sample. The percentage change in the quantity
demanded of oranges is slightly greater than the percentage change in price.
Significance of the Coefficients and Goodness of Fit There are numer-
ous questions involving how well the regression line fits the sample data points in
any regression analysis. Two important issues are:
1. Hypothesis testing for the significance of the estimated coefficients
2. The goodness of fit of the entire estimating equation
Because the estimated coefficients in Equation 4.2 are derived from a sample
of data, there is always a chance that the sizes of the estimated coefficients are
dependent on that particular sample of data and might differ if another sample
Simple regression analysis
A form of regression analysis
that analyzes the relationship
between one dependent and one
independent variable.
26The technical details of this process can be found in any standard econometrics textbook. See, for exam-
ple, Robert S. Pindyck and Daniel L. Rubinfeld, Econometric Models and Economic Forecasts, 4th ed. (New
York: McGraw-Hill, 1998); Jeffrey M. Wooldridge, Econometric Analysis of Cross Section and Panel Data,
2nd ed. (Cambridge, MA: The MIT Press, 2010), and the books by Greene and Kmenta noted above.
27Standard statistical packages include SAS, SPSS, and STATA. Spreadsheet software packages such as
Excel also include regression analysis.
M04_FARN0095_03_GE_C04.INDD 126 11/08/14 5:27 PM
ChAPTEr 4 Techniques for Understanding Consumer Demand and Behavior 127
was used. The coefficients might also not be different from zero in the larger popu-
lation. This issue would be of particular concern with the small data sample in
Table 4.1. Table 4.1 includes the predicted value of quantity demanded and the
residual, or the difference between the actual and predicted values, for each obser-
vation. Figure 4.2 plots the predicted quantity demanded at each price with the
actual quantity demanded. Although the actual and predicted values appear to be
relatively similar, we need a quantitative measure of how well the data fit the esti-
mated equation.
Regression analysis packages provide an estimate of the standard error of each
estimated coefficient, a measure of how much the coefficient would vary in regres-
sions based on different samples. A small standard error means that the coefficient
would not vary substantially among these regressions. In Table 4.1, the standard
error of the price coefficient is 0.176, while that of the constant term is 12.571.
Managers can use a t-test, based on the ratio of the size of the coefficient to its
standard error, to test for the significance of the coefficients of the independent
variables in a regression analysis. The t-test is used to test the hypothesis that a
coefficient is significantly different from zero (i.e., whether there is a high probabil-
ity that, in repeated drawings of samples from the population, the coefficient will
be a number different than zero or H1: B ≠ 0) versus the hypothesis that the coef-
ficient is equal to zero (H0: B = 0). This result is typically indicated by a t-statistic
greater than 2.0, which means that a manager can be 95 percent certain that the
coefficient is not zero in the larger population.
Large values of the t-statistic show statistically significant results because the
standard error is small relative to the size of the estimated coefficient. In Table 4.1,
the t-statistic for the price coefficient is –8.952, while that for the constant term
is 16.741. Because both of these numbers are greater than 2 in absolute value, a
manager can be at least 95 percent certain that the estimated coefficients are sta-
tistically significant and that the data support the hypothesis H1: B ≠ 0. The Excel
program shows the actual degree of significance associated with the t-statistics.
There is a 434 in 100,000,000 chance that the price coefficient is not statistically
significant, while there is a 1 in 100,000,000 chance that the constant term is not
statistically significant.
The Excel program also calculates confidence intervals around the estimated
coefficients. These statistics show the range of values in which we can be confi-
dent that the true coefficient actually lies with a given degree of probability, usually
95 percent. Given the results in Table 4.1, a manager can be 95 percent confident
that the true value of the price coefficient lies between –1.185 and –1.970 and that
the constant term lies between 182.435 and 238.454.
The goodness of fit of the entire estimating equation to the data set is shown by
the coefficient of determination (R2). The value of this coefficient ranges from
0 to 1, with the size of the coefficient indicating the fraction of the variation in the
dependent variable that is explained statistically by the variables included in the
estimating equation. In Table 4.1, the variation in quantity demanded is due partly
to the variation in price (the regression effect) and partly to the effect of a random
Standard error
A measure of the precision of an
estimated regression analysis
coefficient that shows how much
the coefficient would vary in
regressions from different samples.
t-test
A test based on the size of the
ratio of the estimated regression
coefficient to its standard error that
is used to determine the statistical
significance of the coefficient.
Confidence interval
The range of values in which we
can be confident that the true
coefficient actually lies with a given
degree of probability, usually
95 percent.
Coefficient of
determination (R2)
A measure of how the overall
estimating equation fits the data,
which shows the fraction of the
variation in the dependent variable
that is explained statistically by the
variables included in the equation.
50 100
Price
Q
ua
nt
it
y
150
Q
Predicted Q
0
0
50
100
150 FIgUrE 4.2
Simple regression Analysis
Actual Versus Predicted results
This figure plots the actual and
predicted quantity demanded relative
to price in the simple regression
analysis of quantity and price.
M04_FARN0095_03_GE_C04.INDD 127 11/08/14 5:27 PM
128 PArT 1 Microeconomic Analysis
disturbance (the residual or error effect). The coefficient of determination tests
how well the overall model fits the data by decomposing the overall variation into
the variation resulting from each of these effects.
The coefficient of determination is defined as the ratio of the sum of squared
errors from the regression effect to the total sum of squared errors (regression plus
residual effect). In Table 4.1, this ratio is 5,601/6,300 = 0.889 (shown as the value
of the R2 statistic at the top of the table). There is no absolute cutoff point for the
value of the R2 statistic. The values of the coefficient of determination are typically
higher for time-series data than for cross-sectional data, as many variables move
together over time, which can explain the variation in the dependent variable. R2
statistics for time-series analyses can exceed 0.9 in value, while those for cross-
sectional studies are often in the range of 0.3–0.4.
When more variables are added to a regression equation, the R2 statistic can
never decrease in size. Thus, one method for obtaining a higher value of this statis-
tic is to keep adding independent variables to the estimating equation. Because this
procedure could give misleading results, managers can also use the adjusted R2
statistic, which is defined in Equation 4.4.
4.4 R−2 = 1 − (1 − R2)
(n − 1)
(n − k)
where
R
−2= adjusted R2
R2 = coefficient of determination
n = number of observations
k = number of estimated coefficients
The number of observations (n) minus the number of estimated coefficients (k)
is called the degrees of freedom in the estimating equation. You cannot have
more estimated coefficients than observations in an equation. The estimated equa-
tion in Table 4.1 has 10 degrees of freedom because there are 12 observations and
2 estimated coefficients. The adjusted R2 statistic is typically lower than the coef-
ficient of determination because it adjusts for the number of degrees of freedom in
the estimating equation, and the statistic could actually be negative. In Table 4.1,
the value of the adjusted R2 statistic is 0.878 compared to 0.889 for the coefficient
of determination. There is only a small difference between the two statistics, given
the 10 degrees of freedom in the equation.
An alternative measure of goodness of fit is the F-statistic, which is the ratio of
the sum of squared errors from the regression effect to the sum of squared errors
from the residual effect, or the variation explained by the equation relative to the
variation not explained.28 A larger F-statistic means that more variation in the data
is explained by the variables in the equation. The value of the F-statistic in Table 4.1
is 80.143, which is well beyond the 95 percent probability level.
The F-statistic can be used to test the significance of all coefficients jointly in
equations that have multiple independent variables. It is similar in concept to the
t-statistic for testing the significance of individual regression coefficients. It is
possible that the t-statistics might indicate that the individual coefficients are not
statistically significant, while the F-statistic is statistically significant. This result
could occur if the independent variables in the equation are highly correlated with
each other. Their individual influences on the dependent variable may be weak,
while the joint effect is much stronger.
Adjusted R2 statistic
The coefficient of determination
adjusted for the number of degrees
of freedom in the estimating
equation.
Degrees of freedom
The number of observations (n)
minus the number of estimated
coefficients (k) in a regression
equation.
F-statistic
An alternative measure of goodness
of fit of an estimating equation that
can be used to test for the joint
influence of all the independent
variables in the equation.
28These terms are adjusted for their degrees of freedom.
M04_FARN0095_03_GE_C04.INDD 128 11/08/14 5:27 PM
ChAPTEr 4 Techniques for Understanding Consumer Demand and Behavior 129
Relationship Between One Dependent and Multiple
Independent Variables: Multiple Regression Analysis
We now extend the Excel example of simple regression analysis showing the rela-
tionship between price and quantity demanded of oranges in Table 4.1 to a multiple
regression analysis, which adds advertising expenditure to the estimating equation
in Table 4.2. Although we know that many other variables should also influence the
TABlE 4.2 Multiple regression Analysis results
rEgrESSION STATISTICS
Multiple R 0.980
R square 0.961
Adjusted R square 0.952
Standard error 5.255
Observations 12.000
ANOVA
Degrees of
Freedom
Sum of
Squares
Mean
Square

F-statistic
Significance of
F-statistic
Regression 2.000 6051.510 3025.755 109.589 0.000000481
Residual 9.000 248.490 27.610
Total 11.000 6300.000

Coefficients
Standard
Error

t-statistic

P-value
lower
95 percent
Upper
95 percent
Intercept 116.157 24.646 4.713 0.001 60.404 171.909
Price –1.308 0.129 –10.110 0.000 –1.601 –1.015
Advertising 11.246 2.784 4.039 0.003 4.947 17.545
rESIDUAl OUTPUT

Observation

Predicted Q

residuals
Quantity
(lbs.)
Price
(cents/lb.)
Advertising
Expenditure ($)
1 47.222 7.778 55 100 5.50
2 69.297 0.703 70 90 6.30
3 92.497 –2.497 90 80 7.20
4 103.327 –3.327 100 70 7.00
5 95.455 –5.455 90 70 6.30
6 107.263 –2.263 105 70 7.35
7 87.583 –7.583 80 70 5.60
8 111.553 –1.553 110 65 7.15
9 122.029 2.971 125 60 7.50
10 115.281 –0.281 115 60 6.90
11 124.632 5.368 130 55 7.15
12 123.861 6.139 130 50 6.50
Average 100 70 6.70
M04_FARN0095_03_GE_C04.INDD 129 11/08/14 5:27 PM
130 PArT 1 Microeconomic Analysis
demand for oranges, we can illustrate multiple regression analysis by simply add-
ing one more variable to the equation.
We use multiple regression analysis to estimate the demand function in Equation 4.5.
4.5 Q = a − bP + cADV
where
Q = quantity demanded
a = constant term
b = coefficient of price variable = ∆Q/∆P, all else held constant
P = price
c = coefficient of advertising variable = ∆Q/∆ADV, all else held
constant
ADV = advertising expenditure
As with the simple regression analysis example, we are estimating a linear rela-
tionship between the dependent variable, quantity demanded, and the two indepen-
dent variables—price and advertising expenditure. The constant term, a, shows the
effect on quantity demanded of other variables not included in the equation. The
coefficients, b and c, show the effect on quantity demanded of a unit change in each
of the independent variables. Each coefficient shows this effect while statistically
holding constant the effect of the other variable. Thus, using multiple regression
analysis to estimate demand relationships from behavioral data solves the “all else
held constant” problem by statistically holding constant the effects of the other
variables included in the estimating equation.
The demand relationship estimated in Table 4.2 is shown in Equation 4.6, using
the variables defined in Equation 4.5.
4.6 Q = 116.157 − 1.308P + 11.246ADV
We can see that the coefficient of the price variable in Equation 4.6 is different
from that in Equation 4.2 even though we use the same data in both equations.
The difference arises from the fact that Equation 4.6 includes advertising expen-
diture, so that this equation shows the effect of price on quantity demanded,
holding constant the level of advertising. Because no other variables are held
constant in Equation 4.2, the price variable coefficient in that equation may pick
up the effects of other variables not included in the equation that also influ-
ence the quantity demanded of oranges. This result is likely to occur if there
are other excluded variables that are highly correlated with price. Thus, it is
important to have a well-specified estimating equation based on the relevant
economic theory.
The coefficients of the price and advertising variables in Equation 4.6 show
the change in quantity demanded resulting from a unit change in each of these
variables, all else held constant. As with the simple regression analysis example,
these coefficients can be used to calculate the relevant elasticities. Using the aver-
age values of price, quantity demanded, and advertising expenditure, we calcu-
late the price elasticity of demand in Equation 4.7 and the advertising elasticity in
Equation 4.8.
4.7 eP =
(∆Q)(P)
(∆P)(Q)
=
(−1.308)(70)
(100)
= −0.9156
4.8 eADV =
(∆Q)(ADV)
(∆ADV)(Q)
=
(11.246)(6.70)
(100)
= 0.7535
M04_FARN0095_03_GE_C04.INDD 130 11/08/14 5:27 PM
ChAPTEr 4 Techniques for Understanding Consumer Demand and Behavior 131
The price elasticity calculated in Equation 4.7 is smaller than that calculated in
Equation 4.3. In fact, the estimated elasticity coefficient in Equation 4.7—derived
from Equation 4.6, with the level of advertising expenditure held constant—
indicates that demand is price inelastic, while the coefficient in Equation 4.3—
derived from Equation 4.2, which does not include the level of advertising
expenditure—indicates elastic demand. It appears that the price variable in
Equation 4.2 is picking up some of the effect of advertising on demand because
the latter variable is not included in that equation. Managers must realize that
econometric results can vary with the specification of the demand equation.
All relevant variables need to be included in these equations to derive the most
accurate empirical results.
Table 4.2 also presents the summary statistics for the multiple regression analy-
sis. We can see that the standard errors of the two independent variables and
the constant term are all small relative to the size of the estimated coefficients,
so that the t-statistics are larger than 2 in absolute value. The two independent
variables and the constant term are statistically significant well beyond the 95
percent level of confidence. Table 4.2 also shows the confidence intervals for all
of the terms.
Figure 4.3 shows the actual and predicted values of quantity demanded relative to
price, while Figure 4.4 shows the same values relative to advertising expenditure.
Regarding the goodness of fit measures, the value of the coefficient of determina-
tion (R2) is 0.961, while that of the adjusted R2 statistic is 0.952. Although the latter
is smaller than the former as expected, both statistics increased in value from the
simple regression analysis in Table 4.1, indicating the greater explanatory power of
the multiple regression model. The F-statistic is also highly significant in Table 4.2.
Other Functional Forms
The linear demand functions, estimated in Equations 4.2 and 4.6, imply both that
there is some maximum price that drives consumers’ quantity demanded of the
product back to zero and that there is some maximum quantity that people demand
at a zero price. As we have seen, the price elasticity of demand also changes at dif-
ferent prices along a linear demand curve. These characteristics of a linear demand
function may not always adequately represent the behavior of different groups of
individuals or the demand for various products, particularly at the end points of the
demand curve.
Q
Predicted Q
50 100
Price
Q
ua
nt
it
y
1500
0
50
100
150 FIgUrE 4.3
Multiple regression Analysis,
Fit of Price Variable
This figure plots the actual and
predicted quantity demanded
relative to price in the multiple
regression analysis of quantity, price,
and advertising expenditure.
Q
Predicted Q
2.00 4.00
Advertising
Q
ua
nt
it
y
8.000.00
0
100
200
6.00
FIgUrE 4.4
Multiple regression Analysis,
Fit of Advertising Variable
This figure plots the actual and
predicted quantity demanded
relative to advertising expenditure
in the multiple regression analysis
of quantity, price, and advertising
expenditure.
M04_FARN0095_03_GE_C04.INDD 131 11/08/14 5:27 PM
132 PArT 1 Microeconomic Analysis
It is often hypothesized that a multiplicative nonlinear demand function of the
form shown in Equation 4.9 (where the variables are defined as in Equation 4.5)
better represents individuals’ behavior:
4.9 QX = (a)(PXb)(ADV c)
This function, illustrated in general in Figure 4.5, is called a log-linear demand
function because it can be transformed into a linear function by taking the loga-
rithms of all the variables in the equation. This function is also called a constant-
elasticity demand function because the elasticities are constant for all values of the
demand variables and are represented by the exponents, b and c, in Equation 4.9.29
Thus, the price and advertising elasticities can be read directly from the statistical
results if this type of function is used in the estimation process. No further calcula-
tions are needed to determine the elasticities. This function may also better repre-
sent consumer behavior in certain cases because it implies that as price increases,
quantity demanded decreases, but does not go to zero.
Demand Estimation Issues
Demand functions estimated for actual products are obviously much more complex
than the simple examples presented in Equations 4.2 and 4.6. Managers could not use
the results of such simple equations for decision making. However, these examples
provide a starting point for understanding the more complex analyses discussed
below. The estimation process and the choice of functional form in real-world
demand equations are based on the issues presented in these simple examples.
The variables included in a multiple regression analysis may be influenced by
data availability, as well as by the underlying economic theory. Data for demand
estimation are often drawn from large-scale surveys undertaken by the federal
government, universities, nonprofit groups, industry or trade associations, and
company consumer research departments. In many cases, analysts would like to
include certain variables, but a consistent set of observations for all individuals in
the analysis may not be available. Some data sources may have better information
on economic variables, while others may have more data on personal characteris-
tics of the individuals included in the analysis. Analysts may also have to use other
variables as proxies for the variables of greatest interest.
29Using calculus, the price elasticity can be calculated as follows for a simple constant elasticity demand
function:
eP = (dQX/dPX)(PX/QX)
QX = (a)(PXb)
eP = [(a)(b)PXb − 1][PX/QX] = [(b)(a)PXb]/[(a)(PXb )] = b
Similar calculations follow for other terms in such a function.
Demand
x
x
x
x
x
xx
x
x
x
x
0 Q
PFIgUrE 4.5
log-linear Demand Curve
A log-linear demand curve has a
constant price elasticity everywhere
along the curve.
M04_FARN0095_03_GE_C04.INDD 132 11/08/14 5:27 PM
ChAPTEr 4 Techniques for Understanding Consumer Demand and Behavior 133
Every multiple regression analysis is influenced by the sample of data—time-
series, cross-sectional, or panel—that is used. The analyst wants to estimate behav-
ioral relationships that can be generalized beyond the sample of observations
included in the analysis. Yet large-scale data collection can be very expensive and
time-consuming. Thus, the analyst must be concerned that the estimated relation-
ships may hold only for the sample of data analyzed, and not the larger population.
As we discussed above, the analyst engages in hypothesis testing to determine how
much confidence can be placed in the results of a particular analysis and whether
these outcomes can be generalized to a larger population.
Case Study of Statistical Estimation
of Automobile Demand
We now discuss issues arising with the use of multiple regression analysis to esti-
mate the demand for automobiles and the associated elasticities.30 This discussion
is drawn from a research study that clearly illustrates many of the methodological
problems just presented.
Automobile demand studies have used both cross-sectional and time-series anal-
ysis with aggregate and disaggregated data. Studies have been undertaken for the
entire market, market segments (domestic vs. foreign), and particular brands of
automobiles. Thus, given the differences in data sets, the functional forms of the
estimating equations, and the variables included, we would expect to find a range
of elasticity estimates in these studies. Aggregate time-series studies generally esti-
mate market automobile price elasticities to be less than 1 in absolute value and
income elasticities to be greater than +2.00, indicating a lack of sensitivity to price
for automobiles as a commodity, but a strong sensitivity in the demand for automo-
biles to changes in income. The disaggregated cross-sectional studies found price
elasticities for particular vehicle types ranging from –0.51 to –6.13. The large elas-
ticities at the upper end of the range should not be surprising, given the degree of
substitution between different brands of cars.
Price elasticities have been found to be smaller for subcompact and compact
vehicles compared to larger models and for two-vehicle households compared to
one-vehicle households. The cross-section literature also found income elasticities
greater than +1.00 and, in some cases, greater than +5.00. A 1985 study estimated
an income elasticity of +1.96 for a Chevy Chevette and +7.49 for a Mercedes 280S,
indicating a substantial consumer response for these models to changes in income.
Automobile demand by households owning one vehicle tended to be less sensitive
to changes in income than by two-vehicle households.
Managerial rule of Thumb
Using Multiple regression Analysis
In using multiple regression analysis to estimate consumer demand, a manager must decide which
variables to include in the analysis. Various types of statistical problems can arise if relevant variables
are excluded from the analysis or if irrelevant variables are included. The choice of variables is derived
from economic theory, real-world experience, which is the problem under consideration, and com-
mon sense. ■
30This discussion is based on Patrick S. McCarthy, “Market Price and Income Elasticities of New Vehicle
Demands,” Review of Economics and Statistics 78 (August 1996): 543–47.
M04_FARN0095_03_GE_C04.INDD 133 11/08/14 5:27 PM
134 PArT 1 Microeconomic Analysis
Automobile demand studies typically include price, income, credit availabil-
ity, and automobile stocks as independent variables. One issue that has been
debated in the literature is whether to include variables measuring automobile
quality, which are typically derived from Consumer Reports and surveys by J.D.
Power and Associates, a major marketing research firm. Thus, complex research
studies often use data from sources that both managers and consumers read.
Excluding quality variables from demand estimation studies could create econo-
metric problems. The estimated price elasticity of demand coefficient would be
biased downward if a model is estimated without including quality variables and
if quality is positively associated with price and demand. Statistical and econo-
metric problems can also arise if the independent variables in the model are
highly correlated with each other. It may be difficult to separate out the effect
of price from the other variables in this case. These issues again illustrate the
problem of estimating the relationship between price and quantity demanded,
“all else held constant.”
In a 1996 study, Patrick S. McCarthy31 estimated automobile demand based
on data from the J.D. Power and Associates 1989 New Car Buyer Competitive
Dynamics Survey of 33,284 households. This survey contained information on the
vehicle purchased, household socioeconomic and demographic characteristics,
and various activities associated with purchasing the vehicle. McCarthy’s sample
of 1,564 households, which was approximately 5 percent of the usable survey
records, was randomly drawn from the larger survey to enable generalizations
to be made to the larger population. The author supplemented the data from the
J.D. Power survey with data on price, warranty, exterior and interior size, fuel
economy, reliability, and safety from the 1989 Automotive News Market Data
Book, Consumer Reports, and the 1989 Car Book. He obtained gasoline prices
from the Oil and Gas Journal and population estimates from the U.S. Bureau of
the Census.
Table 4.3 shows the independent variables, estimated coefficients, and summary
statistics for McCarthy’s study. The independent variables included measures of
automobile costs (price and operating cost per mile), physical characteristics and
vehicle style (horsepower, length, government crash test results, and vehicle type),
quality (results of a consumer satisfaction index), manufacturer, consumer search
activities (the number of first and second visits to different dealers and whether
the consumer repurchased the same brand as previously), and household socio-
economic data. The study used multinomial logit analysis, a special form of regres-
sion analysis in which the dependent variable is a discrete variable (to purchase or
not purchase a specific vehicle) rather than a continuous variable (the number of
vehicles purchased).
McCarthy was satisfied with the precision of the estimating model in terms of
the estimated signs of the variables, the t-statistics, and the coefficient of deter-
mination. Most of the signs of the variables were estimated as predicted, and the
t-statistics indicated that the variables were statistically significant. The value of
the coefficient of determination (0.26) was low, but comparable to those of other
cross-sectional studies.32
The results in Table 4.3 show that both higher vehicle prices, relative to annual
income, and higher vehicle operating costs lowered the quantity demanded of
automobiles (the negative sign on those coefficients). Thus, the estimated demand
curve is downward sloping. Vehicle safety, net horsepower, and overall vehicle
31Ibid.
32The coefficient of determination and other measures of goodness of fit are slightly different in this model
than in the standard linear regression model, given the discrete nature of the dependent variable.
M04_FARN0095_03_GE_C04.INDD 134 11/08/14 5:27 PM
ChAPTEr 4 Techniques for Understanding Consumer Demand and Behavior 135
TABlE 4.3 The Demand for Automobiles
INDEPENDENT VArIABlES COEFFICIENT (T-STATISTIC)
Cost-related attributes Vehicle price/annual income –2.452 (–9.1)
Operating cost per mile (cents) –0.4498 (–5.8)
Metropolitan population if >50,000 0.0000173 (1.4)
Vehicle style and physical attributes Crash test variable 0.2409 (3.0)
Net horsepower 0.00949 (6.0)
Overall length (inches) 0.0166 (5.4)
SUV, van, pickup truck 1.445 (4.8)
Sports car segment –1.277 (–4.7)
Luxury segment—domestic –0.4944 (–3.7)
Perceived quality Consumer satisfaction index 0.0085 (3.5)
Vehicle search costs 1st dealer visit—domestic 3.034 (11.1)
1st dealer visit—European 4.274 (15.2)
1st dealer visit—Asian 3.726 (11.6)
Subsequent dealer visits—Domestic 0.3136 (5.6)
Subsequent dealer visits—European 0.7290 (5.7)
Subsequent dealer visits—Asian 0.3337 (5.9)
Repurchase same brand 2.320 (2.0)
Socioeconomic variables Resident of Pacific Coast –1.269 (–5.0)
Age > 45 years old 0.9511 (4.8)
Manufacturing brand variables Chrysler 1.007 (4.7)
Ford, General Motors 1.721 (8.5)
Honda, Nissan, Toyota 1.267 (9.1)
Mazda 1.005 (5.5)
Summary statistics R2 0.26
Number of observations 1564
Source: Patrick S. McCarthy, “Market Price and Income Elasticities of New Vehicle Demands,” Review of Economics and Statistics 78 (August 1996): 543–47. © 1996
by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.
length all had a significantly positive effect on automobile demand (as evidenced
by the positive coefficients for these variables). Increased values of these vari-
ables would shift the demand curve to the right. Consumers in this sample exhib-
ited a greater demand for vans, SUVs, and pickup trucks relative to automobiles
and station wagons and a smaller demand for sports cars and domestic cars in the
luxury segment.
We can also see that demand was positively related to increases in perceived
quality (the positive coefficient on the consumer satisfaction index variable) and
that search costs influenced vehicle demands. McCarthy argued that the positive
coefficients on the dealer-visits variables indicated that the information benefits
from an additional visit more than offset the additional search costs, but that the
additional benefits declined with subsequent visits. The variable showing whether
M04_FARN0095_03_GE_C04.INDD 135 11/08/14 5:27 PM
136 PArT 1 Microeconomic Analysis
a consumer repurchased the same vehicle brand also had a positive coefficient,
indicating a positive effect on demand. Because repurchasing the same brand low-
ers search and transactions costs, this variable had the expected positive sign.
These results indicate that consumers react not just to the monetary price of an
automobile, but also to the full purchasing costs, including the costs of obtaining
information and searching for the vehicle. The study also determined that younger
consumers and those residing on the Pacific Coast had smaller demands for domes-
tic vehicles than other age and geographic groups.
Table 4.4 shows McCarthy’s estimated price and income elasticities for both the
entire market and the domestic, European, and Asian segments. The estimated
demand for automobiles in this study was generally price inelastic, although the
elasticity estimate for European models was slightly greater than 1 in absolute
value. The cross-price elasticities were estimated to be positive numbers, indicat-
ing substitute goods, as economic theory suggests. Sales of European and Asian
automobiles responded more to changes in the prices of substitute brands than
did sales of U.S. automobiles. All income elasticities were found to be greater than
+1.00, indicating substantial sensitivity to income changes.
The McCarthy study is particularly useful for managers because it is short and
well written, and it focuses on the issues relevant to managerial decision making.
Although academic consumer demand studies do not always meet these criteria,
they can be useful starting points for managers. While the results of academic
research studies may sometimes be too general for managerial decision making,
they can be suggestive of strategies that managers should pursue. For example,
McCarthy found that vehicle characteristics, quality, and consumer search vari-
ables were important influences on automobile demand.
One other problem with academic studies is the time lag often involved in their
publication. The McCarthy article was published in 1996, using data from 1989. This
type of lag is typical for academic research because articles are peer reviewed and
revised several times.33 While this time lag may limit the usefulness of academic
research for managerial decision making, it does not make these studies worth-
less. The peer review process increases the reliability of the research results. In
many cases, these results are available online or in the form of working papers long
before they are officially published. Studies of past market behavior can also give
managers insights into future trends.
TABlE 4.4 Automobile Demand Elasticities
OwN-PrICE
ElASTICITY
CrOSS-PrICE
ElASTICITY
INCOME
ElASTICITY
Entire market –0.87 0.82 1.70
Market segment
Domestic –0.78 0.28 1.62
European –1.09 0.76 1.93
Asian –0.81 0.61 1.65
Source: Patrick S. McCarthy, “Market Price and Income Elasticities of New Vehicle Demands,” Review of Economics
and Statistics 78 (August 1996): 543–47. © 1996 by the President and Fellows of Harvard College and the Massachusetts
Institute of Technology.
33The use of electronic communication has considerably reduced the time lags in the academic review process.
M04_FARN0095_03_GE_C04.INDD 136 11/08/14 5:27 PM
ChAPTEr 4 Techniques for Understanding Consumer Demand and Behavior 137
Relationships Between Consumer Market Data
and Econometric Demand Studies
In the 1998 book Studies in Consumer Demand: Econometric Methods Applied
to Market Data, Jeffrey A. Dubin illustrated the relationships between consumer
market data, which managers typically use, and formal econometric demand stud-
ies based on these data.35 In many cases, researchers analyze market data to obtain
insights on what variables to include in their econometric models of demand.
Although Dubin employed advanced econometric methods to estimate the demand
for various products, we focus on two selected cases—Carnation Coffee-mate and
Carnation Evaporated Milk—to illustrate the relationships between consumer mar-
ket data and econometric models of demand. We also discuss a recent study esti-
mating the demand for cheese in the United States.
Case Study I: Carnation Coffee-mate
To estimate the value of intangible assets, such as brand names, Dubin used
Carnation Coffee-mate as one of his examples. He began with consumer surveys
drawn from Carnation’s marketing files and interviews with key individuals who
were marketing the products during the 1980s. Dubin used a Carnation Consumer
Research Department survey to help define the market for the product. According
Managerial rule of Thumb
Using Empirical Consumer Demand Studies
Empirical consumer demand studies are important to managers because they show the types of data
available for analyzing the demand for different products. Many data sources, such as industry and
consumer surveys that researchers discover, may not have been widely publicized. Demand studies
also discuss previous analyses, and they indicate how researchers conceptualize the problem of esti-
mating the demand for a particular product. ■
34Gerard J. Tellis, “The Price Elasticity of Selective Demand: A Meta-Analysis of Econometric Models of
Sales,” Journal of Marketing Research 25 (November 1988): 331–41; Raj Sethuraman and Gerard J. Tellis,
“An Analysis of the Tradeoff Between Advertising and Price Discounting,” Journal of Marketing Research
28 (May 1991): 160–74.
35Jeffrey A. Dubin, Studies in Consumer Demand: Econometric Methods Applied to Market Data (Boston:
Kluwer Academic Publishers, 1998).
Many of these issues regarding the estimation of price and advertising elastici-
ties have been raised in the marketing literature.34 Elasticity studies, particularly
of specific product brands, may produce biased estimates if they do not include
variables measuring product quality, the distribution of the product, advertising
expenditures, other promotion activities such as coupons and rebates, and lagged
prices and sales. Thus, one of the major problems in statistical demand studies is to
correctly specify the model being estimated and to locate data on all the variables
that should be included in the model.
M04_FARN0095_03_GE_C04.INDD 137 11/08/14 5:27 PM
138 PArT 1 Microeconomic Analysis
to this survey, 37 percent of all cups of coffee were whitened, with milk used in half
of these cases and a nondairy powdered creamer in another 20 percent. Coffee-
mate was the best-selling nondairy powdered creamer, with Cremora by Borden a
major competitor. In addition to milk, other substitutes for Coffee-mate included
cream, evaporated milk, and powdered milk.
Trends in coffee consumption also affected the demand for whiteners, given
the complementary relationship between the two goods. A beverage industry
survey showed that there had been a long-term decline in the per-capita daily
consumption of coffee in the United States between 1962 and 1985. Although
the average number of cups consumed by adults and the proportion of the
population drinking coffee declined, these trends were offset by increases in
the total U.S. population, so that the total amount of coffee consumed actually
increased.
Marketing studies showed that coffee consumption differed by season, region,
gender, and age. Socioeconomic status also played a role in consumption.
A Carnation marketing study showed that Coffee-mate consumption was highest
among coffee drinkers who had incomes under $10,000, had no more than a high
school education, were employed in blue-collar occupations, lived in smaller
cities or rural areas, and were African Americans. Studies also indicated that
Coffee-mate had higher brand loyalty than its competitors and that Coffee-mate
users were less likely to use coupons to purchase the product.
Carnation Coffee-mate Demand Model Variables and Elasticities
Dubin based his demand model for Coffee-mate on these survey results. He used a
constant elasticity model, as illustrated in Equation 4.9 and Figure 4.5, so that his
estimated coefficients were the various elasticities of demand. Dubin included the
following variables in his analysis:
•    The price of Coffee-mate
•    The prices of substitute goods
•    Variables accounting for trends over time and seasonality effects
•    Real income (adjusted for price changes) per capita
•    Frequency of coffee consumption
•    The total volume of all commodity sales in the region
•    Real advertising expenditure of branded creamers
•    Retail support measures, including in-aisle displays, in-ad coupons, and special
pricing
He focused on the 16-ounce size of Coffee-mate, which had the highest sales
volume in the product line and was marketed primarily in the retail distribution
channel.
Dubin estimated a price elasticity coefficient of –2.01 for the 16-ounce Coffee-
mate, as well as positive cross-price elasticities with its competitor brands. In
addition to the role of price, he found seasonal effects on the demand for Coffee-
mate (increased consumption in February and March), resulting from the pat-
terns of coffee consumption. Although the real income and coffee consumption
variables were not significantly different from zero, the all-commodity sales vari-
able was highly significant, indicating the impact of activity in larger markets on
the demand for nondairy creamer. Of the retail support variables, only in-ad cou-
pons and special prices showed positive effects on the demand for Coffee-mate.
Displays of the product within the stores did not have an impact on consumer
M04_FARN0095_03_GE_C04.INDD 138 11/08/14 5:27 PM
ChAPTEr 4 Techniques for Understanding Consumer Demand and Behavior 139
demand, according to the study results. Increased advertising for Coffee-mate
also did not increase the demand for the product. However, increased advertising
for Cremora actually increased the demand for Coffee-mate. Dubin attributed this
result to the increased consumer awareness of creamers from advertising even if
that advertising was not directed specifically to Coffee-mate. Thus, the study of
Coffee-mate both confirmed the predictions of economic theory about price and
cross-price elasticities and tested for the influence of other variables suggested by
consumer research.
Case Study II: Carnation Evaporated Milk
Dubin also examined the market for evaporated milk, focusing on the leading
brand, which was produced by Carnation as well. Marketing studies had shown that
there were two distinct market segments—those individuals who used evaporated
milk in coffee and everyday foods, such as soups, potatoes, and sauces, and those
who used it for holidays and seasonal foods. These groups had different purchasing
patterns, brand preferences, and demographic characteristics. A 1987 Carnation
marketing study found that while only 13 percent of all evaporated milk consum-
ers made five or more purchases per year, they represented 61 percent of the total
category sales volume. In addition, 60 percent of evaporated milk consumers made
only one purchase per year (representing 15 percent of sales volume). Compared to
its competitors, Carnation sales were more concentrated among light users of evap-
orated milk. Much consumer research also indicated that Hispanics and African
Americans tended to be heavy users of evaporated milk for both everyday and holi-
day foods unique to their cultures. This demographic trend created a geographic
pattern of demand, with increased consumption in the South and Southwest, where
many members of these groups live. Evaporated milk consumption was found to be
greater in the fall and winter months, given the use of this product in coffee, baked
goods, and soups, products more likely to be consumed during those months.
Consumer research also indicated that younger, less affluent, and less educated
households were more likely to purchase store label or generic evaporated milk
than brand names.
Carnation Evaporated Milk Demand Model Variables and Elasticities
Given this background, Dubin estimated the demand for Carnation evaporated
milk as a function of the following variables:
•    The price of the Carnation product
•    The price of substitute goods
•    Variables accounting for trends over time, seasonality effects, and regional
differences
•    Real income level
•    The percent of the population that is Hispanic
•    Real advertising expenditures
•    Retail support measures, including in-aisle displays, local advertising, and special
pricing
Dubin found that Carnation evaporated milk had a price elasticity of –2.03, while
its competitors had smaller elasticities of –0.88 (PET Evaporated Milk) and –1.22 (all
other brands). He estimated the expected positive cross-price elasticities between
these products. Dubin found a positive effect of retail support, particularly through
M04_FARN0095_03_GE_C04.INDD 139 11/08/14 5:27 PM
140 PArT 1 Microeconomic Analysis
displays and local advertising. He found that Carnation’s advertising increased the
overall demand for evaporated milk, while PET’s advertising was not significant in
influencing demand. Although real income was positively related to the demand for
evaporated milk, the Hispanic population variable was not significant in the analy-
sis. The latter result was surprising, given the emphasis placed on this subgroup in
the consumer marketing studies. Either this variable was not important by itself,
or it was correlated with other variables, and the statistical analysis was unable to
determine its independent effect.
Case Study III: The Demand for Cheese
in the United States
The demand for cheese in the United States has been influenced by numerous fac-
tors including an increased availability of cheese varieties, greater use of cheese by
fast food and pizza restaurants, increased use of cheese by food manufacturers and
individuals cooking at home, increased consumption of ethnic foods using large
amounts of cheese, and changes in consumer demographics.36 Cheese is also now
sold in many forms including consumer-sized cuts, bagged shredded cheese, and
processed cheese slices. Cheese production and sales have become key compo-
nents of the U.S. dairy industry.
Davis et al. used 2006 Nielsen Homescan data containing demographic and food
purchase information for a nationwide panel of representative households to esti-
mate the demand for cheese in its different forms and the corresponding elas-
ticities. Households in the panel had a handheld device to scan at home all food
purchased at any retail outlet. Each purchase record contained data on product
characteristics, quantity purchased, price paid with and without promotions, date
of purchase, store, and brand information. Researchers also obtained information
on the size and composition of each household, household income, and the ori-
gin, age, race, gender, education, and occupation of household members. Davis
et al. based their econometric analysis on a Tobit demand system estimator that
accounted for the fact that some households had zero purchases of cheese during
the relevant time period.
Davis et al. found that all of their estimated own-price and cross-price elastici-
ties were statistically significant and that all own-price elasticities were negative
as expected. The largest own-price elasticities were for cottage cheese (–2.49),
grated cheese (–2.07), and shredded cheese (–3.74), indicating sizeable changes
in the quantity demanded of these cheeses in response to a 1 percent change in
their prices. The cross-price elasticities for all cheese forms were positive, indi-
cating that they were substitute goods. Consumers were likely to switch to other
cheese forms if the price of their initial choice of cheese increased beyond a certain
threshold. Demographic factors were not as important in the analysis as the price
and income variables. However, household size and whether a female was the head
of a household did impact the demand for cheese. Results of studies such as this
can assist cheese manufacturers and marketers in making decisions on production
changes and marketing strategies.
36This discussion is based on Christopher G. Davis, Donald Blayney, Diansheng Dong, Steven T. Yen, and
Rachel J. Johnson, “Will Changing Demographics Affect U.S. Cheese Demand?” Journal of Agricultural and
Applied Economics 43 (May 2011): 259–73.
M04_FARN0095_03_GE_C04.INDD 140 11/08/14 5:27 PM
ChAPTEr 4 Techniques for Understanding Consumer Demand and Behavior 141
Managerial rule of Thumb
Using Consumer Market Data
Business economists and researchers use the consumer market data familiar to managers and market-
ers to estimate statistical/econometric models of demand and consumer behavior. These demand
studies, in turn, can assist managers in developing competitive strategies by indicating the impor-
tance of the characteristics influencing the demand for different products and by showing what trade-
offs consumers may be willing to make among those characteristics. ■
Summary
In this chapter, we illustrated two major approaches to gathering information
about consumer behavior and demand for different products: (1) marketing and
consumer research methods that include surveys, experiments, and test market-
ing; and (2) statistical and econometric approaches to formally estimating demand
relationships. Managers tend to favor the former approach, while economists in
business and academia use the latter.37 Most of the data for econometric analy-
ses, however, are derived from consumer research studies. We have suggested that
managers be familiar with both approaches because each provides useful informa-
tion on consumer behavior.
Managers need to realize that marketing analysis builds on the fundamental eco-
nomic concepts of demand and elasticity. Marketers take these basic economic
concepts and develop analyses of brand differentiation, market segmentation,
and new product pricing. While some of the formal statistical approaches used by
economists to estimate demand relationships may appear abstract and academic to
managers and marketers, these approaches may do a better job of determining the
effects of different variables on demand, while holding all else constant. This infor-
mation is useful to both academic researchers attempting to improve the methods
of demand estimation and managers needing to make decisions about advertising
spending or how to counter the strategic moves of a competitor.
Key Terms
adjusted R2 statistic, p. 128
coefficient of determination (R2), p. 127
confidence interval, p. 127
conjoint analysis, p. 119
cross-sectional data, p. 124
degrees of freedom, p. 128
direct consumer surveys, p. 119
expert opinion, p. 118
F-statistic, p. 128
multiple regression analysis, p. 123
panel data, p. 124
price experiments, p. 120
simple regression analysis, p. 126
standard error, p. 127
targeted marketing, p. 121
test marketing, p. 120
time-series data, p. 124
t-test, p. 127
37Skouras et al. have suggested that these differences have arisen because improving business performance
has been the primary purpose of marketing, while economics has been more focused on improving the
organization of society. These researchers characterized marketing as a discipline concerned with business
practice, while economics is a social science discipline. See Thanos Skouras, George J. Avlonitis, and Kostis
A. Indounas, “Economics and Marketing on Pricing: How and Why Do They Differ?” Journal of Product &
Brand Management 14 (2005): 362–74.
M04_FARN0095_03_GE_C04.INDD 141 11/08/14 5:27 PM
142 PArT 1 Microeconomic Analysis
50 100
Quantity (lbs. per capita)
P
ri
ce
($
/h
un
dr
ed
w
ei
gh
t)
2000
0
2
4
6
8
150
FIgUrE 4.E1
Demand for Potatoes, 1989–1998
Source: Daniel B. Suits, “Agriculture,” in The Structure of American
Industry, ed. Walter Adams and James Brock, 10th ed. (Upper Saddle
River, NJ: Prentice-Hall, 2001).
40Mike Esterl, “How to Build Buzz for Bud: More Alcohol, Lime-a-Rita,” Wall Street Journal (Online), March
29, 2012.
38Joseph B. White, “Auto Makers Look Past Baby Boomers,” Wall Street Journal (Online), January 16, 2013.
39Emily Glazer, “The Eyes Have It: Marketers Now Track Shoppers’ Retinas,” Wall Street Journal (Online),
July 12, 2012.
Exercises
Technical Questions
1. In each of the following examples, describe how
the information given about consumer demand
helped managers develop the appropriate strate-
gies to increase profitability and how this informa-
tion was obtained:
a. Auto industry executives have begun to focus at-
tention on their 20-, 30-, and 40-year-old custom-
ers, known as Generations X and Y, and away
from the baby boomer generation. Recognizing
that baby boomers are at least 60 years old,
managers realize that their future depends on
adapting to the tastes of younger generations.
The auto industry is now offering more smart-
phone-driven multimedia systems and is con-
sidering increased use of autonomous driving
capability. Luxury car producers are developing
less-expensive models, and companies such as
Toyota are redesigning their cars to be more
compact, efficient, and sporty.38
b. Companies such as Procter & Gamble Co.,
Unilever PLC, and Kimberly Clark Corp. are
now using retina-tracking cameras to test con-
sumer responses to new products. Kimberly
Clark wanted to know which designs on its Viva
paper towels were noticed in the first 10 sec-
onds a customer looked at a shelf. This is the
period when shoppers typically place items
in their carts. Research has shown that what
people want to do and what they say they want
to do are often quite different. Companies are
making increased use of this technology and
three-dimensional computer simulations of
product designs due to the lower costs of this
technology.39
c. Anheuser-Busch InBev NV and other beer pro-
ducers are trying to win back customers who
have switched to smaller brewers or to liquor.
Anheuser launched Bud Light Platinum with a 6
percent alcohol content compared with 4.2 per-
cent in Bud Light. Platinum is sweeter and sold
in a cobalt blue bottle designed to be more pop-
ular in bars where much of this beer is sold. The
company experiments with three new beers
each day in its research brewery. Anheuser is
also developing new products such as Bud
Light Lime-a-Rita, which tastes like a margarita,
and a tea-and-lemonade drink and a cider, each
containing 4 percent alcohol. The company is
using its Clydesdale horses to bring more free
beer samples to festivals and fairs.40
2. The following figure plots the average farm prices
of potatoes in the United States for the years 1989
to 1998 versus the annual per capita consumption.
Each point represents the price and quantity data
for a given year. Explain whether simply drawing
the line that approximates the data points would
give the demand curve for potatoes.
3. Explain why the estimated price variable co-
efficient in simple regression analysis is
greater than that in multiple regression anal-
ysis. What is its effect on the price elasticity
of demand?
M04_FARN0095_03_GE_C04.INDD 142 11/08/14 5:27 PM
ChAPTEr 4 Techniques for Understanding Consumer Demand and Behavior 143
Application Questions
1. Find recent evidence in the Wall Street Journal
and other business publications on how com-
panies are expanding the use of the techniques
described in the opening case to understand and
impact consumer behavior.
2. “All else held constant” is the major problem fac-
ing all methods of estimating the demand for busi-
ness products. Compare and contrast how the
marketing and economic approaches deal with
this problem.
3. Explain what types of biases arise in the different
approaches to understanding consumer demand
and behavior.
4. If you are doing an empirical analysis for a prod-
uct manufactured by your company, discuss what
independent variables should be included in the
demand estimation studies and how the esti-
mated coefficients could help you make marketing
decisions.
M04_FARN0095_03_GE_C04.INDD 143 11/08/14 5:27 PM
In this chapter, we analyze production and cost, the fundamental build-ing blocks on the supply side of the market. Just as consumer behavior forms the basis for demand curves, producer behavior lies behind the sup-ply curve. The prices of the inputs of production and the state of tech-
nology are two factors held constant when defining a market supply curve.
Production processes (or “production functions,” as economists call them)
and the corresponding cost functions, which show how costs vary with the
level of output produced, are also very important when we analyze the behav-
ior and strategy of individual firms and industries.
We begin this chapter with a case that discusses efficiency and costs in
the fast-food industry. Next we discuss short-run versus long-run production
and costs and present a model of a short-run production function. We also
examine economic data on the differences in productivity among firms and
industries. We then present a model of short-run cost functions and discuss
evidence on the shapes of these cost functions. We also distinguish between
the types of costs measured by accountants and the cost concepts used by
economists.
5 Production and Cost Analysis in the Short Run
144
M05_FARN0095_03_GE_C05.INDD 144 13/08/14 1:40 PM
The fast-food industry in the United States has typically used
drive-through windows to increase profitability. With 65 per-
cent of fast-food revenue derived from drive-through win-
dows, these windows have become the focal point for market
share competition among fast-food outlets such as Wendy’s,
McDonald’s, Burger King, Arby’s, and Taco Bell. Even chains
that did not use drive-through windows in the past, such as
Starbucks and Dunkin’ Donuts, have added them to their
stores.1
Production technology changes have included the use
of separate kitchens for the drive-through window, timers
to monitor the seconds it takes a customer to move from the
menu board to the pickup window, kitchen redesign to mini-
mize unnecessary movement, and scanners that send custom-
ers a monthly bill rather than having them pay at each visit.
Now, in an attempt to cut costs and increase speed even fur-
ther, McDonald’s franchises have tested remote order-taking.2
It takes an average of 10 seconds for a new car to pull up to
a drive-through menu after one car has moved forward. With
a remote call center, an order-taker can answer a call from a
different McDonald’s where another customer has already
pulled up. Thus, a call center worker in California may take
orders from Honolulu, Gulfport, Miss., and Gillette, Wyo. This
means that during peak periods, a worker can take up to 95
orders per hour. The trade-offs with this increased speed at
the drive-through window are employee dissatisfaction with
constant monitoring and the stress of the process, decreases in
accuracy in filling orders, and possible breakdowns in commu-
nication over long distances. However, this technology may be
expanded to allow stores, such as Home Depot, to equip carts
with speakers that customers could use to wirelessly contact a
call center for shopping assistance.
In Asia and other parts of the world where crowded cit-
ies and high real estate costs limit the construction of
drive-throughs, McDonald’s and KFC have added motorbike
delivery as part of their growth strategy.3 Fifteen hundred of
the 8,800 restaurants in McDonald’s Asia/Pacific, Middle
East, and Africa division offer delivery, while half of the new
restaurants KFC builds in China each year will offer deliv-
ery. The delivery option requires an area in the restaurant to
assemble orders that are placed in battery-powered induction
heating boxes. Along with cold items in insulated contain-
ers, all of the orders are placed on the back of yellow and red
McDonald’s branded motorbikes or electric scooters. Most
McDonald’s delivery orders are phoned in, but the company
has started offering Internet-based ordering in Singapore and
Turkey. The number of call centers may be reduced in the
future as online ordering increases. Neither McDonald’s nor
KFC plan to use this technology in the United States, where
McDonald’s derives two-thirds of its sales from drive-through
customers.
This case illustrates how firms can use production tech-
nology to influence their costs, revenues, and profits. Because
firms in more competitive markets may not have much abil-
ity to influence the prices of their products, they may depend
more on strategies to increase the number of customers and
lower the costs of production. These strategies may involve
changing the underlying production technology, lowering
the prices paid for the inputs used, and changing the scale of
operation.
To analyze these issues, we’ll first discuss the nature of
a firm’s production process and the types of decisions that
managers make regarding production. We’ll then show how a
firm’s costs of production are related to the underlying pro-
duction technology. Because the time frame affects a man-
ager’s decisions about production and cost, we distinguish
between the short run and the long run and discuss the impli-
cations of these time frames for managerial decision making.
This chapter focuses on short-run production and cost deci-
sions, while we analyze production and cost in the long run
later (Chapter 6).
Case for Analysis
Production and Cost Analysis in the Fast-Food Industry
145
1Jennifer Ordonez, “An Efficiency Drive: Fast-Food Lanes,
Equipped with Timers, Get Even Faster,” Wall Street Journal,
May 18, 2000.
2Matt Richtel, “The Long-Distance Journey of a Fast-Food
Order,” The New York Times (Online), April 11, 2006.
3Julie Jargon, “Asia Delivers for McDonald’s,” Wall Street
Journal (Online), December 13, 2011.
M05_FARN0095_03_GE_C05.INDD 145 13/08/14 1:40 PM
146 PArt 1 Microeconomic Analysis
Defining the Production Function
To analyze a firm’s production process, we first define a production function and
distinguish between fixed and variable inputs and the short run versus the long run.
The Production Function
A production function describes the relationship between a flow of inputs and
the resulting flow of outputs in a production process during a given period of time.
The production function describes the physical relationship between the inputs
or factors of production and the resulting outputs of the production process. It is
essentially an engineering concept, as it incorporates all of the technology or knowl-
edge involved with the production process. The production function illustrates how
inputs are combined to produce different levels of output and how different com-
binations of inputs may be used to produce any given level of output. It shows the
maximum amount of output that can be produced with different combinations of
inputs. This concept rules out any situations where inputs are redundant or wasted
in production. The production function forms the basis for the economic decisions
facing a firm regarding the choice of inputs and the level of outputs to produce.4
A production function can be expressed with the notation in Equation 5.1:
5.1 Q ∙ f(L, K, Mc)
where
Q = quantity of output
L = quantity of labor input
K = quantity of capital input
M = quantity of materials input
As with demand relationships, Equation 5.1 is read “quantity of output is a function
of the inputs listed inside the parentheses.” The ellipsis in Equation 5.1 indicates that
more inputs may be involved with a given production function. There may also be
different types of labor and capital inputs, which we could denote by LA, LB, LC and
KA, KB, and KC, respectively. Note that in a production function, capital (K) refers to
physical capital, such as machines and buildings, not financial capital. The monetary
or cost side of the production process (i.e., the financial capital needed to pay for
workers and machines) is reflected in the functions that show how costs of produc-
tion vary with different levels of output, which we’ll derive later in the chapter.
A production function is defined in a very general sense and can apply to large-
scale production processes, such as the fast-food outlets in this chapter’s opening
case analysis, or to small firms comprising only a few employees. The production
function can also be applied to different sectors of the economy, including both
goods and services. In this chapter, we use very simple production functions to
illustrate the underlying theoretical concepts, while the examples focus on more
complex, real-world production processes.
Fixed Inputs Versus Variable Inputs
Managers use both fixed inputs and variable inputs in a production function. A fixed
input is one whose quantity a manager cannot change during a given time period,
while a variable input is one whose quantity a manager can change during a given
Production function
The relationship between a flow
of inputs and the resulting flow of
outputs in a production process
during a given period of time.
Fixed input
An input whose quantity a manager
cannot change during a given
period of time.
Variable input
An input whose quantity a manager
can change during a given period
of time.
4The production function incorporates engineering knowledge about production technology and how
inputs can be combined to produce the firm’s output. Managers must make economic decisions about what
combination of inputs and level of output are best for the firm.
M05_FARN0095_03_GE_C05.INDD 146 13/08/14 1:40 PM
ChAPter 5 Production and Cost Analysis in the Short Run 147
time period. A factory, a given amount of office space, and a plot of land are fixed
inputs in a production function. Automobiles or CD players can be produced in the
factory, accounting services can be undertaken in the office, and crops can be grown
on the land. However, once a manager decides on the size of the factory, the amount
of office space, or the acreage of land, it is difficult, if not impossible, to change
these inputs in a relatively short time period. The amount of automobiles, CD play-
ers, accounting services, or crops produced is a function of the manager’s use of the
variable inputs in combination with these fixed inputs. Automobile workers, steel
and plastic, accountants, farm workers, seed, and fertilizer are all variable inputs
in these production processes. The amount of output produced varies as managers
make decisions regarding the quantities of these variable inputs to use, while hold-
ing constant the underlying size of the factory, office space, or plot of land.
Short-Run Versus Long-Run Production Functions
Two dimensions of time are used to describe production functions: the short run and
the long run. These categories do not refer to specific calendar periods of time, such
as a month or a year; they are defined in terms of the use of fixed and variable inputs.
A short-run production function involves the use of at least one fixed input. At
any given point in time, managers operate in the short run because there is always
at least one fixed input in the production process. Managers and administrators
decide to produce beer in a brewery of a given size or educate students in a school
with a certain number of square feet. The size of the factory or school is fixed in the
short run either because the managers have entered into a contractual obligation,
such as a rental agreement, or because it would be extremely costly to change the
amount of that input during the time period.
In a long-run production function, all inputs are variable. There are no fixed
inputs because the quantity of all inputs can be changed. In the long run, managers
can choose to produce cars in larger automobile plants, and administrators can
construct new schools and abandon existing buildings. Farmers can increase or
decrease their acreage in another planting season, depending on this year’s crop
conditions and forecasts for the future. Thus, the calendar lengths of the short run
and the long run depend on the particular production process, contractual agree-
ments, and the time needed for input adjustment.
Short-run production
function
A production process that uses at
least one fixed input.
Long-run production
function
A production process in which all
inputs are variable.
Managerial rule of thumb
Short-run Production and Long-run Planning
Managers always operate in the short run, but they must also have a long-run planning horizon.
Managers need to be aware that the current amount of fixed inputs, such as the size of a factory or
amount of office space, may not be appropriate as market conditions change. Thus, there are more
economic decisions for managers in the long run because all inputs can be changed in that time frame
and inputs can be substituted for each other. ■
Productivity and the Fast-Food Industry
The fast-food case that opened this chapter gave a good illustration of the differ-
ences between short- and long-run production functions. With a given technol-
ogy and fixed inputs, as employees at the drive-through windows work faster to
reduce turnaround time for a drive-through customer, the quality of the service
begins to decline, and worker frustration and dissatisfaction increase. This situa-
tion represents the increased use of variable inputs relative to the fixed inputs in
the short run. The management response to these problems has been to implement
M05_FARN0095_03_GE_C05.INDD 147 13/08/14 1:40 PM
148 PArt 1 Microeconomic Analysis
new technologies for the production process: placing an intercom at the end of the
drive-through line to correct mistakes in orders; finding better ways for employees
to perform multiple tasks in terms of kitchen arrangement; and, most recently, out-
sourcing the drive-through calls to remote call centers or offering delivery in certain
countries. This situation represents the long run, in which all inputs can be changed.
Model of a Short-Run Production Function
In this section, we discuss the basic economic principles inherent in a short-run
production function, illustrated in the fast-food example. To do so, we need to
define three measures of productivity, or the relationship between inputs and out-
put: total product, average product, and marginal product. We then examine how
each measure changes as the level of the variable input changes.
Total Product
Total product is the total quantity of output produced with given quantities of
fixed and variable inputs.5 To illustrate this concept, we use a very simple produc-
tion function with one fixed input, capital (K ), and one variable input, labor (L).
This production function is illustrated in Equation 5.2.
5.2 TP or Q ∙ f(L, K )
where
TP or Q = total product or total quantity of output produced
L = quantity of labor input (variable)
K = quantity of capital input (fixed)
Equation 5.2 presents the simplest type of short-run production function. It has
only two inputs: one fixed (K )and one variable (L). The bar over the K denotes the
fixed input. In this production function, the amount of output (Q) or total product
(TP) is directly related to the amount of the variable input (L), while holding con-
stant the level of the fixed input (K ) and the technology embodied in the produc-
tion function.
Average Product and Marginal Product
To analyze the production process, we need to define two other productivity mea-
sures: average product and marginal product. The average product is the amount
of output per unit of variable input, and the marginal product is the additional
output produced with an additional unit of variable input. These relationships are
shown in Equations 5.3 and 5.4.
5.3 AP ∙ TP/L or Q/L
where
AP = average product of labor
5.4 MP ∙ ∆TP/∆L ∙ ∆Q/∆L
where
MP = marginal product of labor
total product
The total quantity of output
produced with given quantities of
fixed and variable inputs.
Average product
The amount of output per unit of
variable input.
Marginal product
The additional output produced
with an additional unit of variable
input.
5This variable is sometimes called total physical product to emphasize the fact that the production function
shows the physical relationship between inputs and outputs. We use total product for simplicity.
M05_FARN0095_03_GE_C05.INDD 148 13/08/14 1:40 PM
ChAPter 5 Production and Cost Analysis in the Short Run 149
Table 5.1 presents a numerical example of a simple production function based
on the underlying equations shown in the table. Marginal product in Table 5.1 can
be calculated either for discrete changes in labor input (Column 5) or for infini-
tesimal changes in labor input using the specific marginal product equation in the
table (Column 6). Column 5 shows the marginal product between units of input
(Column 2), whereas Column 6 shows the marginal product calculated precisely
at a given unit of input. Column 6 gives the exact mathematical relationships dis-
cussed below.
Relationships Among Total, Average,
and Marginal Product
Let’s examine how the total, average, and marginal product change as we increase
the amount of the variable input, labor, in this short-run production function,
holding constant the amount of capital and the level of technology. We can see
in Table 5.1 that the total product or total amount of output (Column 3) increases
rapidly up to 4.5 units of labor. This result means that the marginal product, or the
additional output produced with an additional unit of labor (Column 6), is increas-
ing over this range of production. Between 4.5 and 10 units of labor, the total prod-
uct (Column 3) is increasing, but the rate of increase, or the marginal product, is
becoming smaller (Columns 5 and 6). Total product reaches its maximum amount
of 217 units when 10 units of labor are used, but total product decreases if 11 units
of labor are employed. The marginal product of labor is 5 as labor is increased from
9 to 10 units and –6 as labor is increased from 10 to 11 units (Column 5). Therefore,
the marginal product is zero when the total product is precisely at its maximum
value of 217 units (Column 6).
tAbLe 5.1 A Simple Production Functiona
QUANtItY OF
CAPItAL (K)
(1)
QUANtItY OF
LAbOr (L)
(2)
tOtAL
PrODUCt (TP)
(3)
AVerAGe
PrODUCt (AP)
(4)
MArGINAL PrODUCt
(MP) (ΔTP/ΔL)
(5)
MArGINAL PrODUCt
(MP) (dTP/dL)
(6)
10 1 14 14.0 14 18
10 2 35 17.5 21 24
10 3 62 20.7 27 28
10 4 91 22.8 29 30
10 4.5 106 23.6 30 30.25
10 5 121 24.2 30 30
10 6 150 25.0 29 28
10 6.75 170 25.1875 26.67 25.1875
10 7 175 25.0 25 24
10 8 197 24.6 22 18
10 9 212 23.6 15 10
10 10 217 21.7 5 0
10 11 211 19.2 –6 –12
aIn this example, the underlying equations showing total, average, and marginal products as a function of the amount of labor, L (with the level of capital assumed
constant), are
TP = 10L ∙ 4.5L2 ∙ 0.3333L3
AP = 10 ∙ 4.5L ∙ 0.3333L2
MP = dTP/dL = 10 ∙ 9L ∙ 1.0L2
M05_FARN0095_03_GE_C05.INDD 149 13/08/14 1:40 PM
150 PArt 1 Microeconomic Analysis
The average product of labor, or output per unit of input (Column 4), also
increases in value as more units of labor are employed. It reaches a maximum
value with 6.75 units of labor and then decreases as more labor is used in the
production process. As you can see in Table 5.1, when the marginal product of
labor is greater than the average product of labor (up to 6.75 units of labor), the
average product value increases from 14 to 25.1875 units of output per input.
When more units of labor are employed, the marginal product becomes less than
the average product, and the average product decreases in value. Therefore, the
marginal product must equal the average product when the average product is at
its maximum value.6
Figures 5.1a and 5.1b show the typical shapes for graphs of the total, average, and
marginal product curves. These graphs illustrate the relationships in Table 5.1, but
are drawn more generally to move beyond this specific numerical example. Labor
input is measured on the horizontal axis of both Figures 5.1a and 5.1b, with differ-
ent quantities shown as L1, L2, and L3. The total product is measured on the vertical
axis of Figure 5.1a, while the average and marginal products are measured on the
vertical axis of Figure 5.1b. The variables are measured on separate graphs because
the sizes of the numbers are quite different, as was shown in Table 5.1.
TP
L2L1 L3
TP
L0
(a) Total product (TP): Short-run production function.
AP,
MP
L2L1 L3
AP
MP
L0
(b) Average product (AP) and marginal product (MP):
Short-run production function.
FIGUre 5.1
the Short-run Production
Function
The short-run production function
illustrates the law of diminishing
returns where the marginal product,
or the additional output produced
with an additional unit of variable
input, eventually decreases.
6The maximum point of average product in Table 5.1 occurs at 6.75 units of labor, where both the average
and the marginal products have the value of 25.1875 units of output per input. This relationship holds for
any average and marginal variables. Suppose your average grade on two exams is 80. Your third exam is
your marginal grade. If you receive a 90 on the third exam, your average grade increases to 83.3. However, if
you receive a grade of 60 on your third exam, your average drops to 73.3. If the marginal variable is greater
than the average variable, the average variable increases. If the marginal variable is less than the average
variable, the average variable decreases.
M05_FARN0095_03_GE_C05.INDD 150 13/08/14 1:40 PM
ChAPter 5 Production and Cost Analysis in the Short Run 151
As in Table 5.1, Figure 5.1a shows the total product (or level of output) first
increasing very rapidly up to labor input level L1 and then increasing at a slower
rate as more labor input is added. The total product curve becomes flatter and flat-
ter until it reaches a maximum output level at labor input level L3. If more labor is
added beyond level L3, the total amount of output, or the total product, decreases.
This total product curve implies that the marginal product of labor first increases
rapidly, then decreases in size, and eventually becomes zero or even negative in
value, as illustrated in Figure 5.1b.
We can also see in Figure 5.1b the typical relationship between the marginal
product and average product curves. Between zero and L2 units of labor, the mar-
ginal product curve lies above the average product curve, which causes the average
product curve to increase. Beyond L2 units of input, the marginal product curve
lies below the average product curve, which causes the average product curve to
decrease. Therefore, the marginal product curve must intersect the average prod-
uct curve at the maximum point of the average product curve. Table 5.2 summa-
rizes these relationships.
Economic Explanation of the Short-Run
Production Function
Why do the graphs of total, average, and marginal products in Figures 5.1a and
5.1b typically have these shapes? To answer this question, we need to focus on the
marginal product curve. In Figure 5.1b, the marginal product curve increases up
to labor input level L1. We call this the region of increasing marginal returns.
Once we have employed L1 units of labor, the marginal product of labor begins
to decline and keeps decreasing until it becomes zero, when L3 units of labor are
utilized. This portion of the marginal product curve illustrates what is known as the
law of diminishing marginal returns (or the law of the diminishing marginal
product). All short-run marginal product curves will eventually have a downward
sloping portion and exhibit this law. Beyond L3 units of labor, the marginal product
of labor is negative. This is the region of negative marginal returns.
The law of diminishing marginal returns occurs because the capital input and the
state of technology are held constant when defining a short-run production func-
tion. As more units of labor input are added to the fixed capital input, the marginal
product may increase at first (zero to L1 units of labor in Figure 5.1b), but the curve
will eventually decline and possibly reach zero or negative values (beyond L3 units
of labor in Figure 5.1b). The additional output generated by the additional units
of the variable input (the marginal product) must decrease at some point because
there are too many units of the variable input combined with the fixed input. (For
Increasing marginal
returns
The results in that region of the
marginal product curve where the
curve is positive and increasing, so
that total product increases at an
increasing rate.
Law of diminishing
marginal returns or law of
the diminishing marginal
product
The phenomenon illustrated
by that region of the marginal
product curve where the curve is
positive, but decreasing, so that
total product is increasing at a
decreasing rate.
Negative marginal returns
The results in that region of the
marginal product curve where the
curve is negative and decreasing, so
that total product is decreasing.
tAbLe 5.2 relationships Among total Product (TP), Average Product (AP),
and Marginal Product (MP) in Figures 5.1a and 5.1b
INPUt rANGe eFFeCt ON tOtAL AND/Or AVerAGe PrODUCt eFFeCt ON MArGINAL PrODUCt
Input values: zero to L1 TP increases at increasing rate MP is positive and increasing
Input values: L1 to L3 TP increases at decreasing rate MP is positive and decreasing
Input values: beyond L3 TP decreases MP is negative and decreasing
Input values: L3 TP is at a maximum MP equals zero
Input values: zero to L2 AP increases MP is greater than AP
Input values: beyond L2 AP decreases MP is less than AP
Input values: L2 AP is at a maximum MP equals AP
M05_FARN0095_03_GE_C05.INDD 151 13/08/14 1:40 PM
152 PArt 1 Microeconomic Analysis
example, there are too many automobile workers in the factory, too many accoun-
tants in the office space, or too many farmhands on the plot of land.) The pro-
duction process becomes constrained by the amount of the fixed input, so that
additional units of the variable input become redundant.
Although a firm is constrained by its scale of production (the amount of its fixed
inputs) and by the state of technology embodied in the production function, the
entire set of curves in Figures 5.1a and 5.1b can shift if the firm either changes the
scale of production or adopts new technology. As we saw in the fast-food exam-
ple, this was the managerial response to diminishing returns in the drive-through
window.
Real-World Firm and Industry
Productivity Issues
The model of a short-run production function is very important for the develop-
ment of the theory of cost and profit maximization and for the analysis of firms
in different market environments. Before proceeding with short-run cost theory,
we’ll discuss several other examples of productivity differences among firms and
industries.
Other Examples of Diminishing Returns
The poultry industry has always faced the problem that chickens, unlike pigs and
cattle, cannot be herded.7 Chickens raised for meat are allowed to roam freely
inside huge chicken houses, so that poultry farmers have traditionally had to rely on
human catchers to run around inside the barns grabbing chickens by hand. Adding
increased amounts of catchers to a chicken house would easily result in diminish-
ing returns. Human catchers are typically expected to grab as many as 1,000 birds
an hour. As with the drive-through fast-food windows, output quality deteriorates
as birds are injured through the speed of the process. Bruised chickens cannot be
sold at grocery meat counters.
After years of failure, manufacturers finally produced machines capable of catch-
ing and caging chickens, up to 150 birds per minute. A five-man crew with this
mechanical harvester can do the work of eight men alone, with chicken injuries
reduced by as much as 50 percent. This technological change would shift the pre-
vious set of marginal and average product curves upward, representing increased
productivity.
Online retailers, which face increased demand during the holidays, have to decide
whether it is more efficient to hire additional workers to fill the orders or to change
technology by using robots.8 Amazon.com has hired more workers who walk 18 to
20 miles per day down aisles lined with shelves to load carts with orders and bring
them back to packing stations. The company holds weekly brainstorming sessions
to prevent diminishing returns and increase productivity. Crate & Barrel employs
robots who carry shelves with the company’s products to workers who, without
walking around the building, pick the items they need to fill orders. The choice
between these approaches is influenced by the costs of hiring additional workers
versus the costs of the robots, which may not be used in nonpeak seasons.
As hospitals treat increasing numbers of patients, concerns have arisen about
how to reduce the number of medical errors and improve patient safety. Errors
7Scott Kilman, “Poultry in Motion: With Invention, Chicken Catching Goes High-Tech,” Wall Street Journal,
June 4, 2003.
8Geoffrey A. Fowler, “Holiday Help: People vs. Robots,” Wall Street Journal (Online), December 19, 2010.
M05_FARN0095_03_GE_C05.INDD 152 13/08/14 1:40 PM
ChAPter 5 Production and Cost Analysis in the Short Run 153
and accidents are examples of diminishing returns in this production process.
Although procedures related to human error, such as encouraging nurses to wash
their hands more often and improving physicians’ handwriting on prescriptions,
have been instituted, changes in the nature of the capital inputs, the hospital build-
ings themselves, are now being undertaken. Technological innovations include the
design of identical rooms so that doctors and nurses can find equipment easily,
placing nurses’ stations so that all patients are visible, and using filters and ultravi-
olet devices to trap and kill germs and improve the hospital airflow. These changes
have reduced infection rates, injuries from falls, and medication errors, thus low-
ering patient length of stay, which, in turn, frees up beds and allows hospitals to
serve more patients.9
Hospital emergency rooms have typically encountered diminishing returns
where patients have to wait so long that they leave without treatment. The number
of emergency departments has decreased by one-third over the past two decades
while the number of patients seeking care has increased by 40 percent and, as a
result, long waits have become common. Many patients, who leave without being
seen, need medical care and/or hospitalization. Hospitals have attempted to offset
diminishing returns by streamlining the method of triage from 10 minutes to three
to five minutes, using less-costly nurse practitioners and physicians assistants, and
reserving beds for only the sickest patients.10
In the luxury handbag industry, Louis Vuitton incurred diminishing returns and
production shortages from the organization of its production process. Traditionally,
each factory had approximately 250 employees with each worker specializing in
one skill such as cutting leather and canvas, gluing and sewing, or making pockets.
Specialists worked on only one batch of bags at a time, while half-completed purses
waited on carts until someone wheeled them to the next section of the assembly
line. Using techniques from the Japanese auto industry, Vuitton has now organized
groups of 6 to 12 workers arranged in clusters of U-shaped workstations containing
sewing machines on one side and assembly tables on the other. Workers pass their
work around the cluster and are able to make more types of bags because each
worker is less specialized.11
Productivity and the Agriculture Industry
New production methods for agricultural crops have led to large increases in pro-
ductivity in this sector over time. A significant example is an experiment in China
that resulted in a doubling of rice crop yields without the use of expensive chemi-
cal fungicides.12 Instead of continuing the practice of planting a single type of
rice, farmers planted a mixture of two different types of rice. This change greatly
reduced the incidence of rice blast, the major disease of this crop, and, in turn,
increased productivity and allowed farmers to abandon expensive chemical treat-
ments of their crops.
Concerns still exist about diminishing returns in rice production as the increased
demand for this food staple has caused farmers to increase fertilizer and water
use, exhausting the soil and draining the water table. Many rice farmers have also
begun planting two crops a year, which places further demands on the soil. Recent
9Gautam Naik, “To Reduce Errors, Hospitals Prescribe Innovative Designs,” Wall Street Journal, May 8,
2006.
10Laura Landro, “ERs Move to Speed Care: Not Everyone Needs a Bed,” Wall Street Journal (Online),
August 2, 2011.
11Christina Passariello, “Louis Vuitton Tries Modern Methods on Factory Lines,” Wall Street Journal,
October 9, 2006.
12Carol Kaesuk Yoon, “Simple Method Found to Vastly Increase Crop Yields,” New York Times, August 22,
2000.
M05_FARN0095_03_GE_C05.INDD 153 13/08/14 1:40 PM
154 PArt 1 Microeconomic Analysis
technological innovations that have attempted to offset these diminishing returns
include developing seed varieties that can withstand droughts or floods, planting
rice in dry soil rather than flooded paddies, altering the way rice plants perform
photosynthesis, and developing hybrid varieties than can increase yields by as
much as 20 percent.13
Productivity and the Automobile Industry
The automobile industry is an obvious example of an industry in which huge pro-
ductivity increases have occurred over time, beginning with Henry Ford’s use of
the assembly line at the beginning of the twentieth century. However, Japan’s use
of improved production techniques in the 1970s and 1980s created major problems
for the U.S. auto industry. The number of vehicles per worker had ranged between
8 and 15 for both domestic and foreign producers in 1960. Although productivity
for General Motors, Ford, and Chrysler remained in that range in 1983, the number
of vehicles per worker increased to 42 for Nissan and 58 for Honda in that year.14
The Japanese productivity advantage in the early 1980s did not result primarily
from differences in technology or labor.15 Approximately two-thirds of the cost
advantage resulted from changes in management focusing on inventory systems,
relations with suppliers, and plant layout. Japanese production was organized
around a lean and coordinated system, with inventories delivered from nearby sup-
pliers every few hours. Workers could stop the assembly line as soon as problems
arose, which improved quality and eliminated the need for repair stations. The
organization of the Japanese workforce with far fewer job classifications also gave
Japanese plants greater flexibility and less downtime than U.S. plants.
In response to these productivity differences, the U.S. automobile industry has
initiated drastic productivity and management changes over the past 20 years,
including redesigned production operations, reorganized management procedures,
and the closing of outdated plants. Between 1979 and 1998, assembly productivity
increased 45 percent at Chrysler and 38 percent at General Motors and Ford. The
relative disadvantage in productivity of the Big Three U.S. automakers in terms of
total hours per vehicle produced relative to the Japanese decreased from 45 per-
cent in 1995 to 14 percent in 2005. Some of the Big Three auto plants are among
the most productive automotive facilities in North America. However, although the
U.S. automakers are adapting their assembly lines to produce different models on
the same line, they still lag behind the Japanese in this area.
A recent technological innovation is the use of the Internet to link companies
with their auto parts suppliers to facilitate bidding on and executing contracts.16
Traditionally, the supply process involved periodic contracts with thousands of
suppliers for a variety of parts, components, and general supplies. Bids were evalu-
ated through phone calls and exchange of paper. Ford and GM introduced online
supply exchanges in 1999 for price quotes, bidding, and monitoring the physical
movement of supplies. Separate exchanges have been replaced by Covisint, which
13Patrick Barta, “Feeding Billions, a Grain at a Time,” Wall Street Journal, July 28, 2007.
14Michael A. Cusumano, The Japanese Automobile Industry (Cambridge, MA: Harvard University Press,
1985), 187–88.
15The discussion of productivity in the automobile industry is based on John E. Kwoka Jr., “Automobiles:
Overtaking an Oligopoly,” in Industry Studies, ed. Larry L. Duetsch, 2nd ed. (Armonk, NY: Sharpe, 1998),
3–27; James W. Brock, “Automobiles,” in The Structure of American Industry, eds. Walter Adams and
James W. Brock, 10th ed. (Upper Saddle River, NJ: Prentice Hall, 2001), 114–36; and James W. Brock, “The
Automobile Industry,” in The Structure of American Industry, ed. James W. Brock, 12th ed. (Upper Saddle
River, NJ: Prentice Hall, 2009), 155–82.
16This discussion is based on: John E. Kwoka, Jr., “Automobiles: The Old Economy Collides with the New,”
Review of Industrial Organization 19 (2001): 55–69; and John Larkin, “Global Collaboration Easier, More
Productive and Safer,” Automotive Industries 186 (2) (Second Quarter 2007): 60–61.
M05_FARN0095_03_GE_C05.INDD 154 13/08/14 1:40 PM
ChAPter 5 Production and Cost Analysis in the Short Run 155
served DaimlerChrysler, Renault S.A., and Nissan Motor Co., in addition to Ford and
GM. In early 2007, General Motors announced that it was using Covisint to connect
and integrate more than 18,000 production and nonproduction suppliers including
firms in Europe, Asia, and Latin America. Cost savings of up to 15 percent of annual
purchasing costs have been estimated from the use of this new technology.
Even with more advanced technologies, diminishing returns can still occur in the
auto production process. Toyota Motor Corporation’s goal of overtaking General
Motors as the world’s No. 1 auto maker and fast-paced expansion resulted in an
increasing number of quality problems in North America, Japan, and elsewhere
that threatened Toyota’s image. Toyota began to rely more heavily on computer-
aided design tools that shorten vehicle-development times by skipping steps such
as making physical prototypes to test components. Computer-aided engineering
tools also allow potential design flaws to slip through the production process.
To overcome these problems, Toyota began adding as much as three to six more
months to projects with a normal development lead time of two to three years.17
However, management pressure for rapid growth combined with the complex-
ity of the company’s products led to further quality problems in 2009 and 2010.
A slipping floor mat that was believed to entrap the accelerator pedal led to an
accident that killed a California state trooper and three of his passengers in August
2009. This malfunction of the floor mat led to a recall of 3.8 million vehicles in the
United States. Toyota also dealt with the controversy over whether its electronic
throttle-control system led to unintended acceleration. In 2010, Toyota created a
new quality control group of 1,000 engineers in Japan, and the company formed
rapid-response teams to address quality or safety problems around the world.18
Productivity Changes Across Industries
Productivity changes differ substantially across industries in the United States.
While productivity for the overall economy increased 0.45 percent per year from
1958 to 1996, annual growth ranges varied from 1.98 percent in electronic and elec-
tric equipment to –0.52 percent in government enterprises.19
Data released in 2000 showed accelerating labor productivity in a range of indus-
tries, including the service sector and durable goods manufacturing. Many of these
productivity gains can be attributed to the increased use of information technology
(IT) in these industries.20
More recent analyses indicate that information technology accounted for almost
80 percent of the increase in productivity growth in the late 1990s. Each genera-
tion of new computer equipment greatly outperformed prior generations. Given
large price declines for information technology investment, firms made massive
investments in IT equipment and software and substituted IT assets for other pro-
ductive inputs. The impact of IT on productivity growth declined in both a rela-
tive and absolute sense in the post-2000 period with IT investment accounting for
about one-third of the productivity growth in this period. This was still a substan-
tial impact, given that IT investment was less than 5 percent of aggregate output.21
17Norihiko Shirouzu, “Toyota May Delay New Models to Address Rising Quality Issues,” Wall Street Journal,
August 25, 2006.
18Mike Ramsey, “Corporate News: U.S. Ends Toyota Probe—Recall of Additional 2.17 Million Vehicles for
Floor-Mat Problems Closes Case,” Wall Street Journal (Online), February 25, 2011; Robert E. Cole, “What
Really Happened to Toyota?” Sloan Management Review 52 (4) (Summer 2011): 29–35.
19Dale W. Jorgenson and Kevin W. Stiroh, “U.S. Economic Growth at the Industry Level,” American
Economic Review 90 (2) (May 2000): 161–67.
20Martin Neil Bailey, “The New Economy: Post-Mortem or Second Wind?” Journal of Economic
Perspectives 16 (2) (Spring 2002): 3–22.
21Dale W. Jorgensen, Mun S. Ho, and Kevin J. Stiroh, “A Retrospective Look at the U.S. Productivity Growth
Resurgence,” Federal Reserve Bank of New York Staff Reports, no. 277, February 2007.
M05_FARN0095_03_GE_C05.INDD 155 13/08/14 1:40 PM
156 PArt 1 Microeconomic Analysis
A study of productivity growth over the entire period 1960 to 2007 found that the
leaders in innovation among the IT-using sectors were wholesale and retail trade.
Companies such as Wal-Mart and Cicso developed integrated supply chains around
the world that linked electronic cash registers at retail outlets with business-to-
business ordering systems. Two IT-producing sectors, semiconductors and comput-
ers, sustained very rapid growth throughout the period. Agriculture, broadcasting,
and telecommunication services also contributed strongly to productivity growth.22
Research has also shown that U.S. firms maintained a productivity advantage over
European firms because they were better able to exploit the use of IT.23
Model of Short-Run Cost Functions
We now analyze how a firm’s costs of production vary in the short run, where at
least one input of production is fixed. We first discuss the economic definition of
cost and then develop cost functions that show the relationship between the cost
of production and the level of output, all other factors held constant.
Measuring Opportunity Cost: Explicit Versus
Implicit Costs
Economists have a very specific way of defining the costs of production that man-
agers should, but do not always, consider. To correctly measure all the relevant
costs of production, managers need to make certain they are measuring the oppor-
tunity costs of the resources they are using. Opportunity costs reflect the cost of
using resources in one activity (production by one firm) in terms of the opportuni-
ties forgone in undertaking the next best alternative activity. In most cases, these
costs are explicit costs because they are paid to other individuals and are found
in a firm’s bookkeeping or accounting system. However, even these bookkeeping
costs may reflect an accounting definition rather than a true economic definition
of opportunity cost. In other cases, these costs are implicit costs. This means that
although they represent the opportunity cost of using a resource or input to pro-
duce a given product, they are not included in a firm’s accounting system and may
be difficult to measure.
In many cases, the prices that a firm actually pays for its inputs reflect the oppor-
tunity cost of using those inputs. For example, if the wages of construction work-
ers are determined by the forces of demand and supply and if all workers who
want to work are able to do so, the monetary or explicit cost paid to those workers
accurately reflects their opportunity cost or their value in the next best alternative.
If the workers are currently employed by Firm A, managers at Firm B must pay a
wage at least equal to that paid by Firm A if they want to hire the workers away
from Firm A. If a firm leases office space in a building or a farmer rents a plot of
land, the explicit rental payments to the owners of these inputs reflect the opportu-
nity cost of using those resources.
What happens if the firm already owns the building or the plot of land? In these
cases, there may not be any budgetary or accounting cost recorded. Does this zero
accounting cost mean that the opportunity cost of using those resources is also
zero? The answer to this question is usually no, because the firm could rent or lease
those resources to another producer. If Firm A could rent the office space it owns
Cost function
A mathematical or graphic
expression that shows the
relationship between the cost of
production and the level of output,
all other factors held constant.
Opportunity cost
The economic measure of cost
that reflects the use of resources in
one activity, such as a production
process by one firm, in terms
of the opportunities forgone
in undertaking the next best
alternative activity.
explicit cost
A cost that is reflected in a payment
to another individual, such as a
wage paid to a worker, that is
recorded in a firm’s bookkeeping or
accounting system.
Implicit cost
A cost that represents the value
of using a resource that is not
explicitly paid out and is often
difficult to measure because it is
typically not recorded in a firm’s
accounting system.
22Dale W. Jorgenson, Mun S. Ho, and Jon D. Samuels, “Information Technology and U.S. Productivity
Growth: Evidence from a Prototype Industry Production Account,” Journal of Productivity Analysis 36
(2011): 159–75.
23Nicholas Bloom, Raffaella Sadun, and John Van Reenen, “Americans Do It Better: US Multinationals and
the Productivity Miracle,” American Economic Review 102 (1) (2012): 167–201.
M05_FARN0095_03_GE_C05.INDD 156 13/08/14 1:40 PM
ChAPter 5 Production and Cost Analysis in the Short Run 157
to Firm B for $100,000 per year, then the opportunity cost to Firm A of using that
space in its own production is $100,000 per year. This is an implicit cost if it is not
actually included in the firm’s accounting system.
If managers do not recognize the concept of opportunity cost, they may have too
much investment tied up in the ownership of buildings, given the implicit rate of
return on these assets compared with the return on other uses of these resources.
For example, Reebok made the strategic decision to contract with other manufac-
turers around the world to produce its shoes rather than invest in plants and equip-
ment itself. Its managers estimated that there was a greater rate of return from
these activities than from investment in buildings.24
Another example of an implicit cost is the valuation of the owner’s or family
member’s time in a family-operated business. In such businesses, family members
may not explicitly be paid a salary, so the costs of their time may not be included
as a cost of production. However, this practice overstates the firm’s profitability. If
the owner or family member could earn $40,000 per year by working in some other
activity, that figure represents the opportunity cost of the individual’s time in the
family business, but this cost may be implicit and not be reflected in any existing
financial statement. It does reflect a real cost of using those resources in a produc-
tion process.
In certain cases, accounting costs may not accurately represent the true opportu-
nity cost of using the resource, given the distinction between historical and oppor-
tunity cost. Historical costs reflect what the firm paid for an input when it was
purchased. For machines and other capital equipment, this cost could have been
incurred many years in the past. Firms have their own accounting systems to write
off or depreciate this historical cost over the life of the capital equipment. In many
cases, these depreciation guidelines are influenced by Internal Revenue Service
regulations and other tax considerations. From an opportunity cost perspective,
the issue is what that capital equipment could earn in its next best alternative use
at the current time. This rate of return may bear little relationship to historical cost
or an annual depreciation figure.
Accounting Profit Measures Versus
Economic Profit Measures
The other important example of opportunity cost relates to the return on financial
capital invested in a firm. If investors can earn 10 percent in an alternative invest-
ment of similar risk, this 10 percent return is an implicit cost of production. A firm
must pay at least 10 percent on its invested capital to reflect the true opportunity
cost of this resource and to prevent investors from placing their money elsewhere.
A firm’s profit is defined as the difference between its total revenue from sales and
its total cost of production. Given the different approaches used by accountants
and economists, we now distinguish between accounting and economic profit.
Accounting profit measures typically focus only on the explicit costs of produc-
tion, whereas economic profit measures include both the explicit and the implicit
costs of production.
There are numerous problems involved in correctly calculating a firm’s eco-
nomic profit, many of which relate to the value of the capital costs of plant and
equipment.25 The appropriate capital cost measure is an annual rental fee or the
price of renting the capital per time period, not the cost of the machine when it
was purchased. The rental cost should be based on the replacement cost of the
historical cost
The amount of money a firm paid
for an input when it was purchased,
which for machines and capital
equipment could have occurred
many years in the past.
Profit
The difference between the total
revenue a firm receives from the
sale of its output and the total cost
of producing that output.
Accounting profit
The difference between total
revenue and total cost where cost
includes only the explicit costs of
production.
economic profit
The difference between total
revenue and total cost where cost
includes both the explicit and any
implicit costs of production.
24This example is drawn from Shlomo Maital, Executive Economics (New York: Free Press, 1994), 30.
25This discussion is based on Dennis W. Carlton and Jeffrey M. Perloff, Modern Industrial Organization,
4th ed. (Boston: Pearson Addison-Wesley, 2005), 247–53.
M05_FARN0095_03_GE_C05.INDD 157 13/08/14 1:40 PM
158 PArt 1 Microeconomic Analysis
equipment or the long-run cost of purchasing an asset of comparable quality.
This rental rate should be calculated after economic depreciation is deducted
on the equipment. Economic depreciation reflects the decline in economic value
of the equipment, not just an accounting measure, such as straight-line depre-
ciation. Advertising and research and development expenditures also create
problems for the calculation of economic profit because, as with capital equip-
ment, the benefits of these expenditures typically extend over a number of years.
Economic profit should also be calculated on an after-tax basis and adjusted for
different degrees of risk because investors generally dislike risk and must be
compensated for it.
This distinction between accounting and economic profit has played an
important role at the Coca-Cola Company.26 Coca-Cola had long followed a
strategy of obtaining its resources through equity financing—selling stock to
shareholders—rather than debt financing—borrowing from banks. Thus, the com-
pany had very low explicit interest payments on its books. Realizing that share-
holders could also invest elsewhere, former CEO Roberto Goizueta calculated
that the opportunity cost of the shareholders’ equity capital was a 16 percent rate
of return. He then learned that all Coke’s business activities except soft drinks
and juices returned only 8 to 10 percent per year. Coca-Cola was essentially bor-
rowing money from shareholders at 16 percent per year and paying them only an
8 percent return. These opportunity costs are difficult to detect because Coke’s
treasurer did not write an annual check for 16 percent of the company’s equity
capital. The cost was reflected in Coke’s capital stock growing less rapidly than it
could have grown.
Goizueta’s response to this management problem was to turn an implicit cost
into an explicit cost:27
His solution was first to sell off those businesses whose capital made a
lower return—i.e., less than 16 percent—than it cost, and second, introduce
a system of accounting in which every operating division of Coca-Cola knew
precisely its economic profit. What he meant by economic profit was sales
revenue minus operating costs, including an opportunity-cost charge for
capital. Those divisions earning a 16 percent return on their shareholder’s
capital were told that their economic profit was zero. And each division’s
operations were judged solely on the basis of the economic profit it earned.
The results of doing so at Coca-Cola were not slow in coming. “When you
start charging people for their capital,” Goizueta said, “all sorts of things
happen. All of a sudden, inventories get under control. You don’t have three
months’ concentrate sitting around for an emergency. Or you figure out that
you can save a lot of money by replacing stainless-steel containers with
cardboard and plastic.”
26This example is drawn from Maital, Executive Economics, 23–25.
27Ibid., 24–25.
Managerial rule of thumb
the Importance of Opportunity Costs
Measuring true opportunity costs can be difficult for managers because accountants are trained to
examine and measure costs explicitly paid out. Valuing implicit costs may seem like an imaginary exer-
cise to accountants. However, as in the Coca-Cola example, managers must recognize the importance
of these costs and devise strategies for turning implicit costs into explicit costs that can be used for
strategic decision making. ■
M05_FARN0095_03_GE_C05.INDD 158 13/08/14 1:40 PM
ChAPter 5 Production and Cost Analysis in the Short Run 159
Definition of Short-Run Cost Functions
A short-run cost function shows the relationship between output and cost for a
firm based on the underlying short-run production function we looked at earlier in
the chapter. Thus, the shapes of the marginal and average product curves in Figure
5.1b influence the shapes of the short-run cost curves, or how costs change as pro-
duction is increased or decreased. Given that the production function shows only
the technology of how inputs are combined to produce outputs, we must introduce
an additional piece of information, the prices of the inputs of production, to define
cost functions. To continue with the example presented in Table 5.1, Equation 5.2,
and Figures 5.1a and 5.1b, we define PL as the price per unit of labor (the variable
input) and PK as the price per unit of capital (the fixed input). The former can be
thought of as the wage rate per worker, while the latter can be considered the price
per square foot of office space or the price per acre of land.
We use this information on production and input prices to define the family of
short-run cost functions in Table 5.3. Even though we define some of the cost func-
tions in Table 5.3 in terms of the inputs of production (labor and capital), we show
numerical and graphical relationships between costs and the level of output (costs
as a function of output). The underlying production function gives us the relation-
ship between the level of labor input (L) and the resulting level of output (Q).
Fixed Costs Versus Variable Costs
Three categories of costs—total, average, and marginal, with further subdivisions
between fixed and variable costs—are shown in Table 5.3. Total fixed cost is the
cost of using the fixed input, K . It is defined as the price per unit of capital times
the quantity of capital (i.e., price per square foot of office space times the number
of square feet). Because the quantity of capital does not change, total fixed cost
remains constant regardless of the amount of output produced. Total variable
cost is defined as the price per unit of labor (or wage rate) times the quantity of
labor input. This cost does change when different levels of output are produced
because it reflects the use of the variable input. Total cost is the sum of total fixed
and total variable costs.
Each of the average costs listed in Table 5.3 is the respective total cost variable
divided by the amount of output produced. Average fixed cost is the total fixed
cost per unit of output, while average variable cost is the total variable cost per
unit of output. As you can see in Table 5.3, average total cost is defined as total
cost per unit of output, but it also equals average fixed cost plus average variable
cost. This equivalence results from the fact that TC = TFC + TVC. Dividing each
one of these terms by Q gives the relationship ATC = AFC + AVC.
Short-run cost function
A cost function for a short-run
production process in which
there is at least one fixed input of
production.
total fixed cost
The total cost of using the fixed
input, which remains constant
regardless of the amount of output
produced.
total variable cost
The total cost of using the variable
input, which increases as more
output is produced.
total cost
The sum of the total fixed cost plus
the total variable cost.
Average fixed cost
The total fixed cost per unit of
output.
Average variable cost
The total variable cost per unit of
output.
Average total cost
The total cost per unit of output,
which also equals average fixed
cost plus average variable cost.
tAbLe 5.3 Short-run Cost Functions (based on the production
function in equation 5.2 and input prices PL and PK)
COSt FUNCtION DeFINItION
Total fixed cost TFC ∙ (PK )(K)
Total variable cost TVC ∙ (PL) (L)
Total cost TC ∙ TFC ∙ TVC
Average fixed cost AFC ∙ TFC/Q
Average variable cost AVC ∙ TVC/Q
Average total cost ATC ∙ TC/Q ∙ AFC ∙ AVC
Marginal cost MC ∙ ∆TC/∆Q ∙ ∆TVC/∆Q
M05_FARN0095_03_GE_C05.INDD 159 13/08/14 1:40 PM
160 PArt 1 Microeconomic Analysis
Marginal cost is the additional cost of producing an additional unit of output. As
you can see in Table 5.3, MC = ΔTC/ΔQ = ΔTVC/ΔQ. This equivalence results from
the fact that marginal cost shows the changes in costs as output changes. Total
variable costs change as the level of output varies, but total fixed costs are con-
stant regardless of the level of output. Therefore, total fixed costs do not influence
the marginal costs of production, and the above definition holds.
Table 5.4 presents short-run cost functions that are based on the production
function from Table 5.1, a price per unit of capital of $50, and a price per unit of
labor of $100.
Relationships Among Total, Average, and Marginal Costs
The first three columns of Table 5.4 show the production function drawn from Table
5.1. Total fixed cost (Column 4) shows the total cost of using the fixed input, which
remains constant at $500 ($50 per unit times 10 units), regardless of the amount of
output produced. Total variable cost in Column 5 ($100 times the number of units
of labor used) increases as more output is produced. Total cost (Column 6) is the
sum of total fixed and total variable costs.
Average fixed cost (Column 7) decreases continuously as more output is pro-
duced. This relationship follows from the definition of average fixed cost, which
is total fixed cost per unit of output. Because total fixed cost is constant, average
fixed cost must decline as output increases and spreads the total fixed cost over a
larger number of units of output. Both average variable cost (Column 8) and aver-
age total cost (Column 9) first decrease and then increase. We can see that average
total cost always equals average fixed cost plus average variable cost. Marginal
cost (Column 10) also first decreases and then increases much more rapidly than
either average variable cost or average total cost.
Figures 5.2a and 5.2b show the typical shapes for graphs of the total, average,
and marginal cost curves. Although these graphs illustrate the relationships in
Table 5.4, they are drawn to present the general case of these functions.
In Figure 5.2a, total fixed costs (TFC) are represented by a horizontal line, as
these costs are constant regardless of the level of output produced. Note that these
fixed costs are incurred even at a zero level of output. If land is rented or office
Marginal cost
The additional cost of producing
an additional unit of output, which
equals the change in total cost or
the change in total variable cost as
output changes.
tAbLe 5.4 Short-run Cost Functions (based on the production function from table 5.1
and input prices PK = $50 and PL = $100)
K
(1)
L
(2)
TP = Q
(3)
TFC
(4)
TVC
(5)
TC
(6)
AFC
(7)
AVC
(8)
ATC
(9)
MC
(10)
10 0 0 $500 $0 $500
10 1 14 $500 $100 $600 $35.71 $7.14 $42.85 $7.14
10 2 35 $500 $200 $700 $14.29 $5.71 $20.00 $4.76
10 3 62 $500 $300 $800 $8.06 $4.84 $12.90 $3.70
10 4 91 $500 $400 $900 $5.49 $4.40 $9.89 $3.45
10 5 121 $500 $500 $1,000 $4.13 $4.13 $8.26 $3.33
10 6 150 $500 $600 $1,100 $3.33 $4.00 $7.33 $3.45
10 7 175 $500 $700 $1,200 $2.86 $4.00 $6.86 $4.00
10 8 197 $500 $800 $1,300 $2.54 $4.06 $6.60 $4.55
10 9 212 $500 $900 $1,400 $2.36 $4.25 $6.61 $6.67
10 10 217 $500 $1,000 $1,500 $2.30 $4.61 $6.91 $20.00
M05_FARN0095_03_GE_C05.INDD 160 13/08/14 1:40 PM
ChAPter 5 Production and Cost Analysis in the Short Run 161
space is leased, these costs must be covered even if no output is produced with
those fixed inputs. Total variable costs, on the other hand, are zero when no out-
put is produced because the variable input is used only when there is a positive
amount of output. Total variable costs are shown as increasing slowly at first and
then more rapidly as output increases. The total cost curve has the same general
shape as the total variable cost curve because the distance between the two curves
is total fixed cost, which is constant (TC = TFC + TVC ⇒ TFC = TC – TVC). The
total cost of producing zero units of output is represented by the distance 0A, or the
amount of the fixed costs. The total fixed cost is the vertical distance between the
total cost and total variable cost curves at any level of output.28
In Figure 5.2b, the average fixed cost curve is declining throughout the range of
production for the reasons discussed above. Both average variable cost and aver-
age total cost are drawn as U-shaped curves, showing that these average costs first
decrease, reach a minimum point, and then increase. Average total cost lies above
average variable cost at every unit of output, but the distance between the two
curves decreases as output increases, as that distance represents average fixed
cost, which is declining (ATC = AFC + AVC ⇒ AFC = ATC – AVC).
Marginal cost in Figure 5.2b is also a U-shaped curve, showing that marginal cost
first decreases, reaches a minimum level, and then increases very rapidly as output
increases. Why would a marginal cost curve typically have this shape? Look back
at Figure 5.1b, which shows the short-run production function that underlies these
cost functions. Note the range of diminishing returns or declining marginal product
in Figure 5.1b. If the additional output obtained from using an additional unit of
labor input is decreasing, then marginal cost, or the additional cost of producing
another unit of output, must be increasing. Thus, the explanation for the upward
sloping short-run marginal cost curve is the existence of diminishing returns in the
short-run production function.
Likewise, the shape of the average variable cost curve in Figure 5.2b is deter-
mined by the shape of the underlying average product curve in Figure 5.1b. When
average product increases, average variable cost decreases. If average product
decreases, average variable cost increases.
$
Q2Q1 Q3
TC TVC
TFC
Q0
(a) Total cost (TC ), total variable cost (TVC ), and
total fixed cost (TFC ) functions.
A
FIGUre 5.2
Short-run Cost Functions
The short-run total cost functions in Figure 5.2a are related to the average and marginal cost functions in Figure 5.2b.
$
Q2Q1 Q3
MC
AVC
ATC
AFC
Q0
(b) Marginal cost (MC ), average total cost (ATC ),
average variable cost (AVC ), and average fixed
cost (AFC ) functions.
28In Table 5.4 total variable cost and total cost may look as if they are increasing at a constant rate. When
these costs are plotted against the level of output, not the level of input, they exhibit the shapes of the
curves in Figure 5.2a.
M05_FARN0095_03_GE_C05.INDD 161 13/08/14 1:40 PM
162 PArt 1 Microeconomic Analysis
Also observe in Figure 5.2b that the marginal cost curve intersects the average
variable cost curve at its minimum point and the average total cost curve at its
minimum point. This is the same average–marginal relationship that we discussed
when describing the short-run production function earlier in the chapter. If mar-
ginal cost is less than average variable cost, as shown between zero and Q2 units of
output in Figure 5.2b, average variable cost is decreasing. Beyond Q2 units of out-
put, marginal cost is greater than average variable cost, so average variable cost is
increasing. Thus, the marginal cost curve must intersect the average variable cost
curve at its minimum point, or Q2 units of output.
The same relationships hold between marginal cost and average total cost.
Marginal cost is less than average total cost up to Q3 units of output. This causes
average total cost to decrease in this range. Beyond Q3 units of output, marginal
cost is greater than average total cost, so average total cost increases. Thus, the
marginal cost curve must intersect the average total cost curve at its minimum
point, or Q3 units of output. The only difference in this marginal–average relation-
ship between the production and cost functions is that the marginal cost curve
intersects the average cost curves at their minimum points, whereas the marginal
product curve intersects the average product curve at its maximum point. This
intersection occurs at either a maximum or a minimum point of the average curves.
Relationship Between Short-Run Production and Cost
The relationships we’ve described in this chapter show the influence of the under-
lying production technology on the costs of production. These relationships, based
on the production function defined in Equation 5.2 and graphed in Figures 5.1a
and 5.1b, are explored further in Table 5.5. This table shows that marginal cost
and marginal product are inversely related to each other, as are average variable
cost and average product. The derivation in the right column of Table 5.5 uses the
definitions of marginal cost and average variable cost to show the inverse relation-
ship between these costs and marginal product and average product, respectively.
These relationships are shown graphically in Figures 5.3a and 5.3b.
Figures 5.3a and 5.3b show the relationship between short-run production and
cost functions. In these figures, labor input level L1 is used to produce output level
Q1, while labor input L2 is used to produce output level Q2. The graphs clearly
show the inverse relationship between the product and cost variables. The mar-
ginal product of labor increases up to L1 input level, so the marginal cost of produc-
tion decreases up to Q1 units of output. The decreasing marginal product beyond L1
units of labor (diminishing returns) causes the marginal cost curve to rise beyond
Q1 units of output. The average product curve increases, reaches its maximum at
L2 units of input, and then decreases. This causes the average variable cost curve to
decrease, reach a minimum value at Q2 units of output, and then increase.
tAbLe 5.5 Short-run Production and Cost Functions
COSt/PrODUCtION reLAtIONShIP DerIVAtION
Relationship between marginal cost (MC)
and marginal product of labor (MPL)
MC ∙
∆TVC
∆Q
∙
(PL )(∆L)
∆Q
MC ∙
PL
(∆Q/∆L)
∙
PL
MPL
Relationship between average variable cost (AVC)
and average product of labor (APL)
AVC ∙
TVC
Q
∙
(PL )(L)
Q
AVC ∙
PL
(Q/L)
∙
PL
APL
M05_FARN0095_03_GE_C05.INDD 162 13/08/14 1:40 PM
ChAPter 5 Production and Cost Analysis in the Short Run 163
Other Short-Run Production and Cost Functions
We have argued that the underlying production function determines the shapes of
the short-run cost curves, and we have illustrated the standard case with a marginal
product curve that first increases and then decreases, resulting in decreasing and
then increasing marginal cost. These traditional-shaped curves result from dimin-
ishing returns in the production function as increased variable inputs are used rela-
tive to the amount of the fixed inputs.
Consider an alternative set of production and cost curves shown in Figures 5.4a,
5.4b, 5.4c, and 5.4d. Figure 5.4a shows a linear total product curve that results
in the constant marginal product curve in Figure 5.4b. This production function
MP,
AP
L2L1
AP
MP
L0
(a) Short-run production
$
Q2Q1
MC
AVC
Q0
(b) Short-run cost
FIGUre 5.3
the relationship between Short-
run Production and Cost
The shape of the short-run
production function in Figure 5.3a
determines the shape of the short-run
cost function in Figure 5.3b.
TP
TP
L0
(a) TP
MP,
AP
MP = AP
L0
(b) MP and AP
$
TFC
TVC
TC
Q0
(c) TC curves
$
MC = AVC
ATC
Q0
(d) MC and AC curves
FIGUre 5.4
Alternative Short-run Production
and Cost Functions
The underlying production function
influences the costs of production.
M05_FARN0095_03_GE_C05.INDD 163 13/08/14 1:40 PM
164 PArt 1 Microeconomic Analysis
exhibits constant, and not diminishing, returns to the variable input, labor. Because
the marginal product of labor is constant, the average product is also constant and
equal to the marginal product. Although diminishing returns will eventually set in
for this production process as the firm approaches the maximum capacity of its
fixed inputs, the production relationships shown in Figures 5.4a and 5.4b may be
valid over a wide range of input and output.
The implications of this production function for the costs of production are
shown in Figures 5.4c and 5.4d. If marginal product is constant over this range of
output, marginal cost must also be constant. There are no diminishing returns in
the production function that would cause the marginal cost of further production
to increase. Because marginal cost is constant, average variable cost is also con-
stant and equal to marginal cost. Average total cost decreases throughout because
it is being pulled down by the declining average fixed cost. Marginal cost must be
less than average total cost because average total cost is decreasing. Because mar-
ginal cost is constant, the total cost and total variable cost functions must be linear,
with the difference between the two curves equal to total fixed cost.
Managerial rule of thumb
Understanding Your Costs
Managers need to understand how their firm’s technology and prices paid for the inputs of produc-
tion affect the firm’s costs. They need to know the difference between costs that change with output
(variable costs) and those that are unrelated to the output level (fixed costs). They also need to under-
stand the difference between average cost (cost per unit of output) and marginal cost (the additional
cost of producing an additional unit of output). ■
Empirical Evidence on the Shapes of Short-Run
Cost Functions
Although we use the standard U-shaped cost curves (Figure 5.2b) for most of our
theoretical analysis, much empirical evidence on the behavior of costs for real-
world firms and industries indicates that total cost functions are linear and, there-
fore, marginal and average variable costs are constant for a wide range of output
(Figure 5.4d). There is even some evidence that firms may produce where marginal
cost is decreasing. Researchers have based their conclusions on both the econo-
metric estimation of cost functions and surveys of firms’ behavior.29
Econometric Estimation of Cost Functions
Much of the empirical estimation of cost functions was undertaken in the 1940s,
1950s, and 1960s.30 Joel Dean’s classic studies of a furniture factory, a leather belt
shop, and a hosiery mill all showed that a linear total cost function best fit the
29A focus on the average total cost of production may predate modern economic theory. Between 1800 and
1805, German sheet music publisher Gottfried Christoph Hartel calculated the average total cost of printing
sheet music using two different technologies: engraving and printing with movable type. His calculations
for engraved music implied a linear total cost function with a fixed setup cost of 900 Pfennigs per sheet and
a constant marginal cost of 5 Pfennigs per sheet. These calculations influenced his decision not to publish
Ludwig van Beethoven’s early works, for which sales volumes were uncertain and average total costs were
high, but to publish a number of the composer’s later works. See Frederic M. Scherer, “An Early Application
of the Average Total Cost Concept,” Journal of Economic Literature 39 (September 2001): 897–901.
30Jack Johnston, Statistical Cost Analysis (New York: McGraw-Hill, 1960); Joel Dean, Statistical Cost
Estimation (Bloomington, IN: Indiana University Press, 1976).
M05_FARN0095_03_GE_C05.INDD 164 13/08/14 1:40 PM
ChAPter 5 Production and Cost Analysis in the Short Run 165
data. These studies examined data sets where the plant, equipment, and technology
were relatively constant over the data period analyzed. Jack Johnston estimated
cost functions for British electric generating plants, road passenger transport, and
a multiproduct food processing firm. From both his own estimation work and a
comprehensive survey of existing studies, Johnston concluded that a constant mar-
ginal cost and declining average total cost best characterized the cost-output data
for a wide variety of firms.
More recent studies have used much more sophisticated econometric techniques
and have estimated cost structures in the context of larger decisions such as inven-
tory management. Analyzing the food, tobacco, apparel, chemical, petroleum, and
rubber industries from 1959 to 1984 and the automobile industry from 1966 to 1979,
one researcher found evidence for declining marginal costs of production.31 To
determine whether these results were related to the use of industry-level data, this
researcher reestimated cost equations for 10 divisions of the automobile industry
and still found evidence of declining marginal costs. Other researchers32 developed
elaborate models of firm pricing behavior that are consistent with a constant mar-
ginal cost of production.
Survey Results on Cost Functions
Although some early work used a survey or questionnaire approach to make infer-
ences about firms’ cost functions, most of the more recent research studies have
been econometric analyses. One notable exception is the survey by Alan Blinder
and his colleagues at Princeton University in the early 1990s.33 Blinder and his col-
leagues drew a sample of 333 firms in the private, unregulated, nonfarm, for-profit
sector of the economy, 200 of which participated in the survey.
The researchers asked officials in these companies a series of structured ques-
tions designed to test alternative theories about why firms do not change prices
regularly in response to changing economic conditions. Although the main goal of
the survey was to test hypotheses about price stickiness, the researchers included
a number of questions about the firms’ cost structures.
Officials in firms responding to the survey reported on average that 44 percent of
their costs were fixed and 56 percent were variable. If these results can be general-
ized to the entire economy, fixed costs appear to be more important to firms than
is shown in the standard cost curves of economic theory (see Figure 5.2b). Fixed
costs were less important in wholesale and retail trade (mean of 33 percent) and
construction and mining (mean of 29 percent) and more important in transporta-
tion, communications, and utilities (mean of 53 percent) and services (mean of 56
percent). The researchers found that many executives did not think in terms of
fixed versus variable costs. Eighteen executives, or 9 percent of the sample, did not
answer the question.
The researchers also had difficulty asking whether marginal cost varied with
production because many executives were not familiar with this concept. The
researchers had to frame the question in terms of the “variable costs of produc-
ing additional units.” The researchers often had to repeat, rephrase, or explain the
question to executives who did not understand the concept. Even with this effort,
10 interviewees were unable to provide an answer. The responses to this question
were quite surprising in light of standard economic theory.
31Valerie A. Ramey, “Nonconvex Costs and the Behavior of Inventories,” Journal of Political Economy 99
(1991): 306–34.
32Robert E. Hall, “Market Structure and Macroeconomic Fluctuations,” Brookings Papers on Economic
Activity 2 (1986): 285–322; Robert E. Hall, “The Relation Between Price and Marginal Cost in U.S. Industry,”
Journal of Political Economy 96 (1988): 921–47.
33Alan S. Blinder, Elie R. D. Canetti, David E. Lebow, and Jeremy B. Rudd, Asking About Prices: A New
Approach to Understanding Price Stickiness (New York: Sage, 1998).
M05_FARN0095_03_GE_C05.INDD 165 13/08/14 1:40 PM
166 PArt 1 Microeconomic Analysis
Forty-eight percent of the respondents indicated that their marginal costs were
constant, 41 percent said they were decreasing, and only 11 percent responded that
their marginal costs were increasing. Although some, if not many, respondents may
have confused marginal and average costs and may really have been reporting that
their average costs were decreasing, this survey response indicates that business
executives do not perceive the textbook U-shaped marginal cost curve to be rel-
evant in many situations.
Constant Versus Rising Marginal Cost Curves
Some of this discrepancy between textbook U-shaped cost curves and real-
world constant or declining marginal cost curves can be explained by the fact
that economic theory shows the range of possibilities for the cost relationships,
not what actually exists in different firms and industries. Econometric estima-
tion based on real-world data and surveys of executives may show constant or
declining marginal cost for the range of output that the firm is actually produc-
ing. Even if firms are currently producing with constant marginal cost, they will,
at some point, reach the capacity of their fixed inputs, which will cause marginal
cost to increase.
Another explanation for the discrepancy regarding the shapes of the cost curves
relates to the differences between agricultural and manufacturing production.34
The concept of diminishing returns and rising short-run marginal cost—with its
emphasis on the fixed, indivisible factors of production, such as land, and on the
variable, divisible factors, such as labor, which change in proportion to the use of
the fixed factors—was derived from agricultural settings. That producers expe-
rience diminishing returns is very plausible when adding additional amounts of
labor, capital equipment, seed, and fertilizer to a fixed amount of land. There is
no need to distinguish between the stock of the fixed input, land, and the flow of
services derived from it. The land provides services continuously and is not turned
off at night.
However, this model may be less appropriate in manufacturing and industrial set-
tings. Much research has indicated that inputs in these settings are likely to be used
in fixed proportions up to the capacity of the plant. Although the stock of a fixed
input is fixed, the flow of services from that stock may be varied and combined
with the services of a variable input in fixed proportions. The size of a machine
may be fixed, but the number of hours it is put in operation can be varied. Both
capital and labor services are variable in the short run and can be changed together
in fixed proportions, thus preventing diminishing returns and rising marginal costs
from occurring in many manufacturing operations.
In manufacturing assembly operations, the normal work period of the plant
is used to adjust the level of output in the short run. For example, automobile
assembly plants use a relatively fixed number of employees per shift and a preset
speed for the flow of materials and components through the line. Output can be
adjusted by changing the length of existing shifts or adding additional shifts in
the face of changing demand. Other assembly operations, such as a collection of
sewing machines in clothing manufacturing, are organized around workstations
rather than a rigid assembly line. Output is varied in these operations by chang-
ing the duration and intensity of the work period at the individual workstations.
34This discussion is based on Carol Corrado and John J. Mattey, “Capacity Utilization,” Journal of Economic
Perspectives 11 (1997): 151–67; Richard A. Miller, “Ten Cheaper Spades: Production Theory and Cost Curves
in the Short Run,” Journal of Economic Education 31 (Spring 2000): 119–30; and Richard A. Miller, “Firms’
Cost Functions: A Reconstruction,” Review of Industrial Organization 18 (2001): 183–200.
M05_FARN0095_03_GE_C05.INDD 166 13/08/14 1:40 PM
ChAPter 5 Production and Cost Analysis in the Short Run 167
In continuous processing operations, such as oil refineries, steel mills, cement
plants, and paper mills, plants operate nearly 24 hours per day, 7 days per week,
given the large shutdown and start-up costs. Output is typically varied by shutting
down part or all of the plant. In all of these cases, output is adjusted by increas-
ing or decreasing the amount of capital and labor services in constant proportion
so that diminishing returns do not occur and a constant marginal cost can be
maintained.
There may be areas other than manufacturing where this type of production tech-
nology is applicable. For example, even though the size of a restaurant is fixed,
managers may shut down part of the table space, given a lack of demand. Once
again, the services of the fixed input are varied even though the stock is constant.
These services can then be used in a fixed proportion with other variable inputs,
such as labor, to avoid the problem of rising marginal cost.
Implications for Managers
Costs play an important role in determining an effective competitive strategy,
particularly if a firm does not have much control over the price of its product.
The distinction between fixed and variable costs is important, as is the con-
cept of marginal cost. However, as noted in the survey by Blinder and his col-
leagues, many executives and managers are not familiar with these concepts.
Cost accounting systems often focus more on management, control, and Internal
Revenue Service considerations than on concepts useful for decision making.
It may also be more difficult for managers to cut costs when firms are profit-
able than when they are not because it may be less obvious that competitors are
catching up.35
Lack of knowledge about costs is not a recent phenomenon. Even though Henry
Ford pioneered the use of mass production and the assembly line as a cost-cutting
measure, he disliked bookkeepers and accountants. Shlomo Maital tells the follow-
ing story:
Once, walking into a room, Henry Ford asked an aide what the white-collar
workers in the room do. Told they were accountants, he ordered, “I want them
all fired. They’re not productive, they don’t do any real work.” The result was
chaos, as Arjay Miller (who later became president) discovered. Miller was
asked to obtain a monthly estimate of Ford company profits. Doing so required
estimates of revenues and costs. Sales projections were fairly straightforward.
But Miller was amazed to learn that the Ford Motor Co. estimated its costs by
dividing its bills into four piles (small, medium, large, extra-large), guessing at
the average sum of the bills in each pile, then measuring the height of each
pile and multiplying the height in inches by average bill size. The system was
not unlike that used 20 years earlier; when piles of bills were not quite so un-
wieldy, the understaffed accountants had weighed them.36
Maital also relates how Akio Morita, the founder of Sony Corp., made a better
strategic decision for his company based on his knowledge of the costs of pro-
duction. In 1955, Morita was trying to market a small, cheap, practical transistor
radio in the United States. Several buyers asked for price quotes on 5,000, 10,000,
30,000, 50,000, and 100,000 units. Because Sony’s current capacity was less than
1,000 radios per month, Morita knew that the entire production process would have
to be expanded to fill these large orders and that this would impact the costs of
35This insight is drawn from Maital, Executive Economics, 76.
36Ibid., 69.
M05_FARN0095_03_GE_C05.INDD 167 13/08/14 1:40 PM
168 PArt 1 Microeconomic Analysis
Summary
We have discussed and illustrated short-run production and cost in this chapter.
The discussion has focused on production functions where there is at least one
fixed input. These production functions all eventually incur diminishing returns
when increased units of the variable inputs are used relative to the amount of
the fixed inputs and the additional amount of output produced begins to decline.
Diminishing returns are fundamental to all short-run production processes.
We then illustrated the impact of the production function on the costs of produc-
tion. Diminishing returns in production cause short-run marginal cost to increase
for a producer. We saw how the U-shaped cost curves of economic theory show
the full range of outcomes in a production process, but that real-world cost curves
may have different shapes. Marginal cost may be constant over a wide range of
output as managers take steps to prevent diminishing returns from occurring
immediately. We also discussed the concept of opportunity cost, which measures
the value of any resource in terms of its next best alternative use. Economists use
this concept when discussing cost, and managers should use it for correct decision
making. The latter do not always do so, given the problems in correctly measuring
opportunity costs.
Later we examine long-run production and cost, where all inputs in a production
function are variable. This discussion focuses on input substitution and the shape
of the long-run average cost curve. All of these issues are fundamental to the dis-
cussion of pricing and other competitive strategies.
Key Terms
accounting profit, p. 157
average fixed cost, p. 159
average product, p. 148
average total cost, p. 159
averaged variable cost, p. 159
cost function, p. 156
economic profit, p. 157
explicit cost, p. 156
fixed input, p. 146
historical cost, p. 157
implicit cost, p. 156
increasing marginal returns, p. 151
law of diminishing marginal returns or
law of the diminishing marginal
product, p. 151
long-run production function, p. 147
marginal cost, p. 160
marginal product, p. 148
negative marginal returns, p. 151
opportunity cost, p. 156
production function, p. 146
profit, p. 157
short-run cost function, p. 159
short-run production
function, p. 147
total cost, p. 159
total fixed cost, p. 159
total product, p. 148
total variable cost, p. 159
variable input, p. 146
37Ibid., 66–68.
production. Morita essentially drew the economist’s U-shaped average cost curve
showing that he would charge the regular price for 5,000 units and a discount for
10,000 units, but successively higher prices for 30,000, 50,000, and 100,000 units.
These higher prices reflected increased short-run average and marginal costs of
production.37
M05_FARN0095_03_GE_C05.INDD 168 13/08/14 1:40 PM
ChAPter 5 Production and Cost Analysis in the Short Run 169
Exercises
Technical Questions
1. The following table shows data for a simple pro-
duction function.

Capital (K)

Labor (L)
total
Product (TP)
Average
Product (AP)
Marginal
Product (MP)
10 0 0 — —
10 1 5
10 2 15
10 3 30
10 4 50
10 5 75
10 6 85
10 7 90
10 8 92
10 9 92
10 10 90
a. From the information in the table, calculate
marginal and average products.
b. Graph the three functions (put total product on
one graph and marginal and average products
on another).
c. For what range of output does this function
have diminishing marginal returns?
d. At what output is average product maximized?
2. The following table shows data for a simple pro-
duction function.

Capital (K)

Labor (L)
total
Product (TP)
Average
Product (AP)
Marginal
Product (MP)
10 0 — —
10 1 25
10 2 75
10 3 120
10 4 83
10 5 54
10 6 35
10 7 22
10 8 10
10 9 4
10 10 1
a. From the information in the table, calculate
total and average products.
b. Graph the three functions (put total product on
one graph and marginal and average products
on another).
c. For what range of output does this function
have diminishing marginal returns?
d. At what output is average product maximized?
3. Jim is considering quitting his job and using his
savings to start a small business. He expects that
his costs will consist of a lease on the building,
inventory, wages for two workers, electricity, and
insurance.
a. Identify which costs are explicit and which are
opportunity (implicit) costs.
b. Identify which costs are fixed and which are
variable.
4. Suppose Marcus is operating a bookstore, and he
made zero economic profit last year.
a. What was Marcus’s accounting profit likely to
be?
b. If the implicit costs had increased, what would
be the effect on Marcus’s economic and ac-
counting profits?
5. The following table shows data for the simple
production function used in Question 1. Capital
costs this firm $20 per unit, and labor costs $10 per
worker.
K L TP TFC TVC TC AFC AVC ATC MC
10 0 0
10 1 5
10 2 15
10 3 30
10 4 50
10 5 75
10 6 85
10 7 90
10 8 92
a. From the information in the table, calculate
total fixed cost (TFC), total variable cost (TVC),
total cost (TC), average fixed cost (AFC),
average variable cost (AVC), average total cost
(ATC), and marginal cost (MC).
b. Graph your results, putting TFC, TVC, and TC
on one graph and AFC, AVC, ATC, and MC on
another.
c. At what point is average total cost mini-
mized? At what point is average variable cost
minimized?
M05_FARN0095_03_GE_C05.INDD 169 13/08/14 1:40 PM
170 PArt 1 Microeconomic Analysis
6. The following table shows data for the simple
production function used in Question 2. Capital
costs this firm $50 per unit, and labor costs $20 per
worker.
K L MP TFC TVC TC AFC AVC ATC MC
10 0 — — —
10 1 25
10 2 75
10 3 120
10 4 83
10 5 54
10 6 35
10 7 22
10 8 10
10 9 4
10 10 1
a. From the information in the table, calculate
total fixed cost (TFC), total variable cost
(TVC), total cost (TC), average fixed cost
(AFC), average variable cost (AVC), average
total cost (ATC), and marginal cost (MC).
(Note that in this case, you are starting from
MP, not TP, and, thus, you should calcu-
late TP first if you didn’t already do that in
Question 2.)
b. Graph your results, putting TFC, TVC, and TC
on one graph and AVC, ATC, and MC on another.
c. At what point is average total cost mini-
mized? At what point is average variable cost
minimized?
7. Consider the shape of the production and cost
functions for two different firms.
a. For Firm 1, workers have constant marginal
product. That is, each worker produces exactly
the same amount as the previous worker. Use
this information to graph the approximate shape
of the firm’s short-run product and cost curves.
b. For Firm 2, workers have diminishing marginal
returns everywhere. That is, each worker always
produces less than the previous worker. Use this
information to graph the approximate shape of
the firm’s short-run product and cost curves.
8. Does an increase in rent lead to the same effect on
a firm’s average fixed cost (AFC), average variable
cost (AVC), average total cost (ATC), and marginal
cost (MC) as an increase in wage rates does?
9. Suppose that a firm’s only variable input is labor.
When 50 workers are used, the average product of
labor is 50, and the marginal product of the 50th
worker is 75. The wage rate is $80, and the total
cost of the fixed input is $500.
a. What is average variable cost? Show your
calculations.
b. What is marginal cost? Show your calculations.
c. What is average total cost? Show your
calculations.
d. Is each of the following statements true or
false? Explain your answer.
1. Marginal cost is increasing.
2. Average variable cost is increasing.
3. Average total cost is decreasing.
Application Questions
1. In the fast-food industry case that opened this
chapter, describe how diminishing returns set in
for the production process and how management
responded to this situation.
2. In order to promote animal welfare, Taiwan’s
government has set a minimum standard for free-
range eggs. If the egg producers provide each hen
an indoor space of no less than 8 square meters,
they will receive certification that distinguishes
their animal-friendly eggs from other battery-cage
eggs.38
Suppose you are an egg producer currently pro-
ducing battery-cage eggs in Taiwan and decide to
start producing free-range eggs next year. In order
to provide more space for the hens, you will have
to expand your farm.
a. What will be the effect of expanding the farm
on your total, average and marginal costs?
b. Does your answer in question (a) give you
enough information to decide whether to pro-
duce free-range eggs next year?
38“Taiwan sets standards for humane production of eggs,” WantChinaTimes, February 11, 2014.
M05_FARN0095_03_GE_C05.INDD 170 13/08/14 1:40 PM
ChAPter 5 Production and Cost Analysis in the Short Run 171
3. The following discussion describes a new inven-
tory system used by J. C. Penney39:
In an industry where the goal is rapid turnaround
of merchandise, J.C. Penney stores now hold
almost no extra inventory of house-brand shirts.
Less than a decade ago, Penney would have stored
thousands of them in warehouses across the U.S.,
tying up capital and slowly going out of style.
The entire program is designed and operated by
TAL Apparel Ltd., a closely held Hong Kong shirt
maker. TAL collects point-of-sale data for Penney’s
shirts directly from its stores in North America for
analysis through a computer model it designed.
The Hong Kong company then decides how many
shirts to make, and in what styles, colors, and
sizes. The manufacturer sends the shirts directly
to each Penney store, bypassing the retailer’s
warehouses and corporate decision makers.
a. Discuss how this case illustrates the concept of
the opportunity cost of capital.
b. How does this innovation also help in demand
management?
4. Explain why a change in a firm’s total fixed cost of
production will shift its average total cost curve,
but not its marginal cost curve.
5. Is it true that in a short-run production process,
the marginal cost curve eventually slopes upward
because firms have to pay workers a higher wage
rate as they produce more output? Explain your
answer.
39Gabriel Kahn, “Made to Measure: Invisible Supplier Has Penney’s Shirts All Buttoned Up,” Wall Street
Journal, September 11, 2003.
M05_FARN0095_03_GE_C05.INDD 171 13/08/14 1:40 PM
172
6 Production and Cost Analysis in the Long Run
In this chapter, we examine production and cost issues in the long run, where all inputs in a production process are variable. In doing so, we’ll build on the short-run production and cost issues you have learned. As you’ll learn in this chapter, a manager faces more decisions in the long
run because it is possible to change the combination of all inputs used in the
production process.
We begin this chapter with a case, the iPhone in China, which focuses on
the long-run decisions made by Apple Inc., and on similar decisions by other
manufacturing companies. This case is an example of a long-run production
function, in which all inputs can be varied and possibly substituted for each
other. We discuss both the feasibility of input substitution in technological
terms and the possible incentives for input substitution in various sectors of
the economy. We present an intuitive analysis of these issues in the chapter
and include the formal model of long-run production, the isoquant model, in
the chapter appendix.
We then define and examine long-run cost functions, focusing on a firm’s
long-run average cost. We show how this concept is derived in economic
theory, and we then provide numerous illustrations of the shapes of long-run
average cost curves for different firms and industries. We end the chapter by
discussing implications of a firm’s long-run average cost for a manager’s com-
petitive strategy.
M06_FARN0095_03_GE_C06.INDD 172 13/08/14 1:39 PM
173
Case for Analysis
The iPhone in China
At a dinner in February 2011, Steve Jobs of Apple Inc. report-
edly responded to President Barack Obama’s question about
what it would take to make iPhones in the United States and
bring those jobs home by saying that these jobs were not coming
back.1 Apple had made the long-run decision that its iPhones
and iPads would be made overseas, given cheaper labor, large-
scale foreign factories, and the flexibility, diligence, and indus-
trial skills of foreign workers. Labor cost is actually a small
component of total costs for most high-technology companies.
More important is the expense of buying parts and managing
supply chains that involve hundreds of companies.
These issues were illustrated in 2007 when Jobs demanded
an iPhone glass screen that could not be scratched about a
month before the phone was to appear in stores. Although
Apple had already selected an American company, Corning
Inc., to manufacture large panes of strengthened glass, the
problem was how to cut the glass to fit an iPhone screen.
A Chinese company, subsidized by the government, received
the contract. iPhones are now being assembled in a complex
known at Foxconn City that has 230,000 employees, many
working six days a week, often for 12-hour days, and living in
company barracks. Although Apple managers had estimated
that it would take nine months to find 8,700 industrial engi-
neers in the United States to oversee the assembly lines needed
to produce iPhones, it reportedly took 15 days to find those
engineers in China. Even though iPhone software and market-
ing campaigns were created in the United States, Apple made
the decision to locate its manufacturing in China because it
believed there were not enough U.S. workers with the needed
skills or factories with sufficient resources.
The quest by Apple for long-run efficiency and cost cutting
has not come without controversy.2 It has been alleged that
Chinese workers assembling iPhones and iPads work exces-
sive overtime, live in crowded dorms, and may have to stand
so long that their legs swell. There have been charges that these
suppliers did not properly dispose of hazardous waste, falsified
records, and disregarded workers’ health. Workers have been
killed by explosions in the plants and injured by the chemi-
cals used to clean iPhone screens. Apple developed a code of
conduct specifying that working conditions in its supply chain
are safe, workers are treated with respect, and manufacturing
processes will not harm the environment. The company’s own
audits found consistent violations of this code of conduct. In
March 2012, the Fair Labor Association found, in an audit of
35,500 workers at three Foxconn facilities, at least 50 legal or
code violations or policy gaps, including violation of 60-hour
workweeks and other health- and safety-related problems,
such as the lack of systems for protecting workers from exces-
sive heat.3
Other manufacturing companies face long-run decisions
similar to those of Apple Inc. The combination of inputs that
firms use, as well as location decisions, depends upon the type of
products manufactured and the costs of all the inputs of produc-
tion. Standard Motor Products in North Carolina, which makes
and distributes replacement auto parts, competes with Chinese
firms.4 Fuel injectors are made in the United States because they
require current technology, strong quality assurance, and highly
skilled workers. They are also likely to be made in small batches
for different makes and models of cars. Many of Standard’s cus-
tomers, who see the company as a distributor rather than a man-
ufacturer, expect the company to be able to deliver its products
anywhere in the United States within 48 hours. These decisions
led the company to increase its fuel injector production in the
United States. Company managers continually re-evaluate the
decision about whether to outsource, but argue that they would
need to save at least 40 percent of U.S. costs to do so.
Several Canadian manufacturers and other companies have
moved their facilities to the United States, citing more compet-
itive wages, lower energy costs, and increased productivity.5
In 2012, wages and benefits at a Caterpillar rail-equipment
plant in Illinois were less than half of those at the company’s
locomotive-assembly plant in Ontario. Navistar International
Corp. sought more flexible work rules and lower wage costs
when it closed its plant in Ontario and relocated production
to Ohio. Sweden’s Electrolux AB planned to close its Quebec
plant and manufacture its ovens, ranges, and cooktops in
Memphis, Tennessee. Bridgestone Corp. had built industrial
radial tires only in Japan, but decided in 2011 to build another
plant in South Carolina to get increased productivity and to
minimize transportation costs for many customers.
1This discussion is based on Charles Duhigg and Keith Bradsher,
“How the U.S. Lost Out on iPhone Work,” The New York Times
(Online), January 21, 2012.
2This discussion is based on Charles Duhigg and David Barboza,
“In China, Human Costs are Built into an iPad,” The New York
Times (Online), January 25, 2012.
3Jessica E. Vascellaro, “Audit Faults Apple Supplier,” Wall Street
Journal (Online), March 30, 2012.
4Adam Davidson, “Making It in America,” The Atlantic
(Online), January/February 2012.
5James R. Hagerty and Kate Linebaugh, “In U.S., a Cheaper
Labor Pool,” Wall Street Journal (Online), January 6, 2012.
M06_FARN0095_03_GE_C06.INDD 173 13/08/14 1:39 PM
174 PArT 1 Microeconomic Analysis
Model of a Long-Run Production Function
This case study illustrates long-run production functions, where all inputs in
the production process are variable and inputs may be substituted for each other.
The case also shows that the long run is a planning horizon. Managers at Apple Inc.
and the other companies discussed in the case considered new technologies and
changes in all of the inputs of production in their decisions.
A simplified long-run production function is presented in Equation 6.1:
6.1 Q = f (L, K)
where
Q = quantity of output
L = quantity of labor input (variable)
K = quantity of capital input (variable)
Unlike a short-run production function, both inputs in this production function
can be varied. Thus, the amount of output that can be produced is related to the
amount of both capital and labor used. In this section, we’ll discuss how changes
in the scale of production impact costs in the long run. But first let’s look at the
concept of input substitution, another important issue that arises when more than
one input is variable.
Input Substitution
Suppose that a firm has already decided that it wants to produce quantity Q1 in the
production function in Equation 6.1. With this production function, firms have still
another economic choice to make. Because both inputs are variable, the firm must
decide what combination of inputs to use in producing output level Q1. The firm
might use either a labor-intensive or a capital-intensive method of production. With
a labor-intensive method of production, managers use large amounts of labor
relative to other inputs to produce the firm’s product. However, it might also be
possible to use a production method that relies on large quantities of capital equip-
ment and smaller amounts of labor; this is called a capital-intensive method of
production. The number of methods that can be used depends on the degree of
input substitution, or the feasibility of substituting one input for another in the
production process.
A manager’s choice of inputs will be influenced by:
•    The technology of the production process
•    The prices of the inputs of production
•    The set of incentives facing the given producer6
Long-run production
function
A production function showing
the relationship between a flow
of inputs and the resulting flow
of output, where all inputs are
variable.
Labor-intensive method of
production
A production process that uses
large amounts of labor relative to
the other inputs to produce the
firm’s output.
Capital-intensive method
of production
A production process that uses
large amounts of capital equipment
relative to the other inputs to
produce the firm’s output.
Input substitution
The degree to which a firm can
substitute one input for another in
a production process.
6The formal rule to minimize the cost of using two variable inputs, labor (L) and capital (K), to produce a
given level of output in a production process is to use quantities of each input such that (MPL/PL) = (MPK/PK),
where MP is the marginal product showing the additional output generated by an additional unit of each input,
PL is the price per unit of labor, and PK is the price per unit of capital. The intuition of this rule is shown as
follows. Assume that there is diminishing marginal productivity (diminishing returns) for both inputs and that
the above ratio is 10/1 for labor and 5/1 for capital. If 1 more unit of labor and 1 less unit of capital are used
in the production process, there is a gain of 10 units of output and a loss of 5 units, so it makes sense to real-
locate the inputs. However, as more labor is used, its marginal product decreases, while the marginal product
of capital increases as less of this input is used. Thus, eventually the ratios will equalize—say, at 8/1. No further
reallocation of inputs will increase output for a given input cost or reduce cost for a given level of output. This
rule, which is formally derived in Appendix 6A, shows that managers minimize costs by considering both the
technology of the production process, which influences productivity, and the prices of the inputs.
M06_FARN0095_03_GE_C06.INDD 174 13/08/14 1:39 PM
ChAPTer 6 Production and Cost Analysis in the Long Run 175
The Technology of the Production Process Production functions vary
widely in the technological feasibility of input substitution. The development of the
assembly line in the automobile industry is one of the best examples of changes in
production technology and the substitution of capital for labor.7 Before Henry Ford
introduced the assembly line, it took 728 hours to assemble an automobile from a
pile of parts located in one place. Initially, Ford installed a system in which a winch
moved the auto-body frame 250 feet along the factory floor and workers picked up
parts spaced along that distance and fitted them to the car. Longer assembly lines,
more specialized workers, and automatic conveyer belts eventually resulted in tre-
mendous reductions in the time necessary to make one automobile.
The fast-food industry is another example of a production process built around a
capital-intensive assembly line in each franchise that includes conveyer belts and
ovens resembling commercial laundry presses. High-technology capital-intensive
production methods have also been developed to supply the inputs to these fran-
chises. The Lamb Weston plant in American Falls, Idaho, one of the biggest french
fry factories in the world, was founded in 1950 by F. Gilbert Lamb, inventor of the
Lamb Water Gun Knife, a device that uses a high-pressure hose to shoot potatoes
at a speed of 117 feet per second through a grid of sharpened steel blades to create
perfect french fries.8
In 2007, airlines and airports considered adopting radio-frequency ID (RFID) bag-
gage tags to replace existing bar code–printed tags. Industry studies had shown
that the RFIDs, which transmit a bag’s identifying number in a manner similar to a
toll-road pass, could reduce lost luggage by 20 percent. The system was estimated
to be 99 percent accurate in reading baggage tags, a significant improvement over
the 80 to 90 percent accuracy of optical scanners reading bar-coded tags. U.S. air-
lines spent approximately $400 million on lost luggage in 2006 to reimburse passen-
gers and deliver late bags to hotels and homes. This decision to change technology
had been limited in the past by the cost of the RFIDs of approximately $1.00 per tag
compared with 4 cents for a bar code–printed tag. However, the price of the RFID
tags had begun to decrease to as low as 15 cents per tag.9
In the railroad industry, managers have begun to consider the use of plastic rail-
road ties made from old tires, grocery bags, milk cartons, and Styrofoam coffee
cups as a replacement for wooden ties, which are vulnerable to rot, fungus, and
termite infestation. Quality is a crucial issue in this decision because railroad ties,
which are spaced 18 or 24 inches apart, must be stiff enough to support heavy-laden
freight trains but flexible enough to bounce back from their tremendous impact.
Plastic tie manufacturers claim their ties can last for at least 50 years, but these ties
typically cost twice as much as wood ties.10
Mining companies are now using high-tech equipment to lower their costs of
production by using fewer workers and by providing them increased protection.11
Rio Tinto connected its Australian mines to satellite links so that workers could
remotely drive drilling rigs, load cargo, and use robots to plant explosives. Robots
can drill 1 million holes in the ground in a year, eliminating thousands of man-hours
of work. BHP Billiton Ltd. and Caterpillar Inc. are designing driverless trucks, and
Rio Tinto is studying how to use a train that does not need human operators for
loading and delivery.
7This discussion is drawn from Shlomo Maital, Executive Economics (New York: Free Press, 1994), 94–95.
8For an extensive description of this production process, see Eric Schlosser, Fast Food Nation (Boston:
Houghton Mifflin, 2001), 130–31.
9Scott McCartney, “A New Way to Prevent Lost Luggage,” Wall Street Journal, February 27, 2007.
10Daniel Machalaba, “New Recyclables Market Emerges: Plastic Railroad Ties,” Wall Street Journal,
October 19, 2004.
11Robert Guy Matthews, “Miner Digs for Ore in the Outback with Remote-Controlled Robots,” Wall Street
Journal (Online), March 1, 2010.
M06_FARN0095_03_GE_C06.INDD 175 13/08/14 1:39 PM
176 PArT 1 Microeconomic Analysis
Changing technology affects all types of production. When AOL Inc. wanted to
determine if it was getting the best use of its video library, it needed to measure
which of its thousands of Web pages published daily contained videos. The com-
pany could have designed video-detecting software or hired temporary workers
for the task. However, it decided to use crowdsourcing—breaking a project into
small components and farming those tasks out to the general public by posting
the requests on a Web site. The project was operating within a week and took only
several months to complete. AOL estimated that the project cost as much as two
temporary workers hired for the same time. Analysts have estimated that crowd-
sourced labor can cost companies half as much as typical outsourcing. Some
crowdsourced tasks take only a few seconds and pay a few cents per task, while
more complex writing or transcription tasks may pay $10 or $20 per job.12
Other production processes may not be as conducive to substitution between
inputs, particularly if they involve a series of complex processes and a highly trained
labor force such as found in the pipe organ industry. At the Schantz Organ Co., the
largest maker of pipe organs in the United States, each worker takes an average of
30 to 40 hours of hand labor to bend specially made sheets of soft metal into the
61 pipes comprising one of the shorter “ranks” or rows of pipes with the seam of
each pipe hand-dabbed with solder. The number of ranks can range from 3 to 150 or
more. It takes four to five years for a worker to become a good pipe maker, while the
“voicers” who tune the pipes spend up to seven years as apprentices. Although some
new technologies, such as computer-controlled routers, have been adopted, labor
costs represent 57 percent of the $8 million sales revenue at Schantz. The average
labor cost relative to shipments for all U.S. manufacturing is 17 percent, while it is
as low as 2 percent in some highly automated sectors such as soybean processing.13
Economists have traditionally argued that input substitution may be less feasible
in the provision of services, particularly in the public sector, than in the produc-
tion of goods, a factor that has become increasingly important as the U.S. economy
has become more service oriented.14 In some service areas, this argument is being
questioned as more input substitution occurs than might be expected. For example,
hospitals are using an increasing number of robots to haul food trays, linens, trash,
medical records, and medications from one area to another in the facility. Robots
are also used to connect doctors with patients through videoconferencing and to
enable doctors to obtain clinical information from remote monitors in real time.15
The process of syndicating corporate loans among banks has undergone rapid
technological change. Until the late 1990s, syndicating a large corporate loan meant
that a bank had to distribute an offering document, often totaling 200 pages, to 50
to 100 banks using overnight mail, fax machines, and hordes of messengers. That
process, now largely handled through banking Web sites, may reduce the time to
close a deal by 25 percent.16
The U.S. Postal Service has developed increasingly sophisticated equipment to
detect illegible handwriting on envelopes. Computers have learned to interpret scrawls
and squiggles, with some machines processing 36,000 letters per hour. However, the
Postal Service still employs hundreds of workers at their Remote Encoding Centers
to sit in silence day and night interpreting incomprehensible writing.17
Input substitution is occurring even in the fine arts. In November 2003, the Opera
Company of Brooklyn announced that it would stage The Marriage of Figaro
12Rachel Emma Silverman, “Big Firms Try Crowdsourcing,” Wall Street Journal (Online), January 17, 2012.
13Timothy Aeppel, “Few Hands, Many Hours,” Wall Street Journal, October 27, 2006.
14William Baumol, “Macroeconomics of Unbalanced Growth: The Anatomy of the Urban Crisis,” American
Economic Review 62 (June 1967): 415–26.
16Steve Lohr, “Computer Age Gains Respect of Economists,” New York Times, April 14, 1999.
15Timothy Hay, “The Robots Are Coming to Hospitals,” Wall Street Journal (Online), March 15, 2012.
17Barry Newman, “Poor Penmanship Spells Job Security for Post Office’s Scribble Specialists,” Wall Street
Journal (Online), November 3, 2011.
M06_FARN0095_03_GE_C06.INDD 176 13/08/14 1:39 PM
ChAPTer 6 Production and Cost Analysis in the Long Run 177
with only 12 musicians and a technician overseeing a computer program that
would play all the other parts.18 The conductor, Paul Henry Smith, has developed
the Fauxharmonic Orchestra, a computer program composed of over a million
recorded notes played by top musicians. The latest software lets users choose from
a large library of digitally stored sounds, adjust for texture and nuance, and assem-
ble them into a complete symphony. A conductor’s jacket, a cyclist’s jersey embed-
ded with a dozen sensors, has been developed to map conductors’ movements and
physiology and translate them to control a piece of music.19
Empirical studies have found that labor productivity growth in the service indus-
tries has proceeded at about the economy-wide rate since 1995. These increases are
broad-based and not just found among a small number of large industries. Much of
this growth is related to the increased use of information technology.20
The Prices of the Inputs of Production As mentioned above in the case of
the adoption of radio-frequency ID baggage tags and plastic railroad ties, the prices
of the inputs of production also influence the degree of input substitution. To mini-
mize their costs of production, firms want to substitute cheaper inputs for more
expensive ones. How much substitution can occur in the face of high input prices
depends on the technology of the production process and institutional factors.
As the movement toward electricity deregulation intensified in southern California
and other parts of the country in the late 1990s and at the turn of the century, elec-
tricity prices fluctuated and in some cases increased dramatically as market forces
swept into the formerly regulated industry. Companies responded to increased
electricity prices through both input substitution and implementation of innova-
tive contracts with their service providers. Because Intel Corporation used huge
amounts of electricity to keep its automated, temperature- and humidity- sensitive
semiconductor-fabrication operations running 24 hours a day, the company could
not enter into interruptible supply contracts with electricity generators that would
provide lower prices, but a nonconstant supply. Instead, Intel negotiated voluntary
consumption restrictions through reduced lighting and air-conditioning levels, and
it designed factory equipment that was less energy-intensive.21
Increased costs of gas, oil, and electricity have continued to influence managerial
decisions about input use. Arla Foods, a farmer-owned cooperative based in Denmark
and the world’s fifth-largest dairy producer by revenue, cut energy use in response
to Denmark’s high taxes on energy consumption. Arla undertook about a dozen proj-
ects to save energy including changing the water chiller, replacing the absorption
dryer in the cheese-aging room, and repairing leaks in compressed-air pipes.22
Law firms have begun using software that enables the discovery process—
providing documents relevant to a lawsuit—to be done electronically rather than
by using higher-priced lawyers and paralegals. Linguistic discovery technologies
use specific search words to find and sort relevant documents, while other tech-
nologies use a sociological approach that mimics human deduction. This software
mines documents for the activities and interactions of people and seeks to visualize
chains of events. Some software can recognize the sentiment in an e-mail message
or detect subtle differences in the style of a message.23
18Jon E. Hilsenrath, “Behind Surging Productivity: The Service Sector Delivers,” Wall Street Journal,
November 7, 2003.
19Jacob Hale Russell and John Jurgensen, “Fugue for Man & Machine,” Wall Street Journal, May 5, 2007.
20Jack E. Triplett and Barry P. Bosworth, “’Baumol’s Disease Has Been Cured: IT and Multifactor
Productivity in U.S. Services Industries,” in The New Economy and Beyond: Past, Present, and Future, ed.,
Dennis W. Jansen (Northampton, MA: Edward Elgar, 2006), 34–71.
21Jonathan Friedland, “Volatile Electricity Market Forces Firms to Find Ways to Cut Energy Expenses,” Wall
Street Journal, August 14, 2000.
22Leila Abboud and John Biers, “Business Goes on an Energy Diet,” Wall Street Journal, August 27, 2007.
23John Markoff, “Armies of Expensive Lawyers, Replaced by Cheaper Software,” The New York Times
(Online), March 4, 2011.
M06_FARN0095_03_GE_C06.INDD 177 13/08/14 1:39 PM
178 PArT 1 Microeconomic Analysis
Managers have attempted to replace workers with machinery even in the fresh
fruit industry where many products have traditionally been picked by hand to main-
tain quality. One company in the Florida citrus industry turned to canopy shakers
to harvest half of the 40.5 million pounds of oranges grown annually from its 10,000
acres in southwestern Florida. In less than 15 minutes, these machines can shake
loose 36,000 pounds of oranges from 100 trees, catch the fruit, and drop it into a
storage car, a job that would have taken four pickers an entire day.24
Input substitution may also result from the scarcity of a particular input. In March
2012, an explosion at a German plant depleted supplies of a resin used to make auto-
mobile fuel and brake lines. Evonik Industries AG was the only integrated maker
of the resin, nylon-12, which is a precise blend of chemicals that can resist reacting
with gasoline and brake fluids. Concerned that the shortage would shut down auto
assembly plants, chemical companies such as DuPont began searching for replace-
ment materials. Evonik discussed with its customers whether another biologically
based polymer might work as a replacement for some applications of nylon-12.25
The ability to change all inputs in the face of changing input prices has affected the
development of numerous industries over time.26 Supermarkets have become the
dominant form of grocery store in the United States. Because these stores are a land-
intensive form of organization—given their size and the need for parking lots around
them—their development depends on the availability of large accessible plots of
land at relatively low prices. In Germany, where less land is available and the popula-
tion is more concentrated in central cities, small supermarkets or minimarkets have
increased productivity by making bulk purchases at the firm level and by providing
only a small variety of goods.
The ability to manufacture goods with cheaper labor abroad led to the decline of
U.S. manufacturing from 20 percent of GDP in 1980 to 12 percent in 2006. However,
even some types of manufacturing are better done closer to the customer. These
include appliances and electronic equipment that are high-end, locally customized,
delicate, very large, or have manufacturing processes that involve almost no labor,
such as medical testing or automated electric-component, chemical, or metal-
fabricating plants.27 In 2011 Otis Elevator Co. moved production from its factory
in Nogales, Mexico, to a new plant in South Carolina. Otis moved production to
Mexico in 1998 to minimize costs, but logistics costs increased substantially since
that time. The new plant will be closer to many of the company’s customers who
are on the East Coast of the United States. Savings will also be derived from hav-
ing all of the company’s white-collar workers associated with elevator design and
production located at the new factory. The plant will also use more automation to
reduce the need for production workers.28
The Incentives Facing a Given Producer The third factor influencing
input substitution is the set of incentives facing a given producer. Firms will sub-
stitute cheaper inputs of production for more expensive ones if they face major
incentives to minimize their costs of production.
The Role of Competitive Environments Input substitution will occur most
often in a competitive market environment where firms are trying to maximize
their profits or are operating under extreme conditions.
24Eduardo Porter, “In Florida Groves, Cheap Labor Means Machines,” New York Times, March 22, 2004.
25Jeff Bennett and Jan Hromadko, “Nylon-12 Haunts Car Makers,” Wall Street Journal (Online), April 17,
2012; Jeff Bennett, “Search Begins for New Resin,” Wall Street Journal (Online), April 18, 2012.
26This discussion is based on Martin Neil Bailey and Robert M. Solow, “International Productivity
Comparisons Built from the Firm Level,” Journal of Economic Perspectives 15 (Summer 2001): 151–72.
27Mark Whitehouse, “For Some Manufacturers, There Are Benefits to Keeping Production at Home,” Wall
Street Journal, January 22, 2007.
28Timothy Aeppel, “Otis Shifts Work Closer to Home,” Wall Street Journal (Online), October 7, 2011.
M06_FARN0095_03_GE_C06.INDD 178 13/08/14 1:39 PM
ChAPTer 6 Production and Cost Analysis in the Long Run 179
The substitution of machinery for labor in the fresh-fruits industry discussed
above was a response to increased global competition facing American farmers.
In the early 1990s, Florida orange farmers had overplanted, and there were large
bumper crops in Brazil where harvesting costs were about one-third as high as in
Florida. These changes gave Florida growers the incentive to invest over $1  million
per year into research in mechanical harvesting to reduce their costs. By the
1999–2000 harvest, this investment resulted in four different types of harvesting
machines working commercially.29
The grocery industry, which is very competitive, has tried to cut costs through
reduced bag usage. Supervalu Inc. expected to save millions of dollars each year
through a bagging program that put more items in each bag or omitted the bag
altogether. The company’s rigorous program prohibited double bagging and the
use of bags for large items with handles and also emphasized the use of plastic
bags, which cost 2 cents per bag compared with 5 cents for a paper bag. Supervalu
Inc., which convened a company-wide task force to study bag use in 2008, could
encounter customer resistance from the use of plastic bags or complaints over the
increased weight of the bags.30
U.S. airlines, operating under extreme competition and facing huge increases in
the cost of fuel, are trying to gain control over the costs of plane parts either by
searching for less expensive suppliers or by determining how to make the parts
themselves at significantly lower costs. Continental Airlines estimates that it has
saved almost $2 million per year by making its own parts such as tray tables and
window shades.31
The airlines have also used new aviation software to help minimize their fuel
costs and the overflight fees charged by countries for using their airspace. These
fees, which cost the world’s air carriers $20 billion per year, are usually based on
takeoff weight and distance travelled. The software, which calculates multiple
scenarios, balances the overflight fees with additional fuel costs if an alternative
route is less direct. United Airlines expects that once the software...