Study GuideIntroduction toPublic PolicyByRobert G..docx

Study Guide
Introduction to
Public Policy
By
Robert G. Turner, Jr., Ph.D.
About the Author
Robert G. Turner, Jr., Ph.D., has more than 20 years of teaching
experience. He has taught seventh grade, worked as a
curriculum
developer for the Upward Bound Program, and taught sociology,
anthropology, and honors seminars at the university level. As a
professional writer, he has written nonfiction books, journal and
magazine articles, novels, and stage plays.
Copyright © 2013 by Penn Foster, Inc.
All rights reserved. No part of the material protected by this
copyright may be
reproduced or utilized in any form or by any means, electronic
or mechanical,
including photocopying, recording, or by any information
storage and retrieval
system, without permission in writing from the copyright owner.
Requests for permission to make copies of any part of the work
should be

mailed to Copyright Permissions, Penn Foster, 925 Oak Street,
Scranton,
Pennsylvania 18515.
Printed in the United States of America
All terms mentioned in this text that are known to be trademarks
or service
marks have been appropriately capitalized. Use of a term in this
text should not be
regarded as affecting the validity of any trademark or service
mark.
INSTRUCTIONS TO STUDENTS 1
LESSON ASSIGNMENTS 9
LESSON 1: PUBLIC POLICY:
WHY WE DO WHAT WE DO 11
LESSON 2: MARKETS AND GOVERNMENT 45
LESSON 3: TOOLS FOR ANALYSIS 75
LESSON 4: MAKING POLICY 117
RESEARCH PROJECT 127
SELF-CHECK ANSWERS 133
iii
C
o

n
t
e
n
t
s
C
o
n
t
e
n
t
s
1
In
s
t
r
u
c
t

io
n
s
In
s
t
r
u
c
t
io
n
s
INTRODUCTION
Welcome to your course in public policy. For many thousands
of years, the concept of public policy was essentially
nonexistent.
Most of us lived in tribal societies composed of relatively small
populations. There was no concept of “public” in tribal worlds.
Daily life was largely face-to-face and, quite often, the name
people had for their society simply meant “the people” or the
“human beings.” The word Cheyenne, for example, simply
means “human beings.” The face of the world for indigenous
people like the Inuit of the Arctic, the Kwakiutl of the Pacific
Northwest, or the Navajo (Dine) was the natural world itself.
Understanding and cooperating with nature was the name of
the game when it came to survival.

However, the rise of civilizations changed all of that in signifi-
cant ways. As population densities and numbers increased in
the first urbanized civilizations, the mode of production was
intensive, organized agriculture that produced food surpluses.
The food surpluses, in turn, permitted a far more complex
division of labor. While most people in the early civilizations
were peasant farmers, food surpluses permitted the rise of
whole categories of people who weren’t peasant farmers.
Among the new social classes of people there were artisans,
soldiers, merchant traders, priests, and bureaucrats. The lat-
ter class represented the interests and policies of social elites,
such as imperial or monarchial dynasties and their ministers
or agents. In general, policies were made at the top and
outcomes
were often problematic for all those folks trying to make a
living
day by day—a number of whom were slaves. Today, in light
of the powerful forces of advancing technology, ideals of dem-
ocratic governance, and sophisticated modes of communication,
social orders are imposed on natural environments, sometimes
in rather destructive ways. In our world of glass, steel, inter-
nal combustion engines, asphalt, and electric power plants,
our environments are essentially artificial. Meanwhile, the
division of labor is so complex that most of us have little
understanding of what others do to make a living. Physicists
talk to physicists, and cops talk to cops. Journalists hop
about in search of a story. Factory-work environments are
a far cry from country club environments. And the people
who inhabit these environments aren’t likely to comprehend
Instructions to Students2
the life of a Maine lobster fisherman or a West Virginia coal
miner. And so, we begin to grasp the baffling complexity

that faces public policy makers.
Today, the word public is still an abstract concept, but it
can be loosely defined as all those people in a complex soci-
ety who either benefit from or are negatively impacted by
policy decisions. For in our world, public policies frame and
organize all of our social institutions—economic, political,
and social. Education, health care, sports, the military, law
enforcement, and the many other fields of human action and
discourse require communal dialogue, sorting out differences,
and reaching compromises on just how our lives should be
conducted and organized.
OBJECTIVES
When you complete this course, you’ll be able to
n Define public policy, and explain the five steps in
the policy-making process
n Describe and contrast the characteristics of for-profit,
nonprofit, and government institutions
n Describe and explain issues related to balancing private
and public concerns in a world characterized by scarcity
and uncertainty
n Explain and describe 10 factors that may challenge
collective efforts to reconcile differences regarding
public policy
n Explain how the complexities of the real world are sim-
plified through the use of conceptual models, such as the
law of supply and demand or the idea that individuals
seek to maximize their utility

n Describe the concept of moral hazards given that infor-
mation is often incomplete and that human behavior
isn’t necessarily rational
n Explain and discuss the characteristics of externalities—
negative or positive—as these may be affected by property
rights, transaction costs, and other factors
Instructions to Students 3
n Explain concepts of strategic interaction as these
are expressed in game theory as exemplified by the
prisoner’s dilemma
n Explain what’s meant by dynamic human behavior as
illustrated by things like runs on banks or collective-
action problems wherein accepted individual behaviors
result in negative group outcomes
n Discuss and explain how indicators, like the poverty
threshold, or indexes, like the human development index
(HDI), are used to assess levels of social welfare
n Describe and explain conceptual tools for making inferences
about social welfare, including efficiency, deadweight loss,
equity, and distinguishing absolute from relative poverty
n Differentiate between parliamentary and presidential
systems with respect to the legislative process
n Discuss and explain societal approaches to allocating
scarce resources and, in particular, the characteristics
of markets

n Describe the characteristics of market failure in relationship
to enforcing property rights, lowering transaction costs,
promoting competition, and ameliorating externalities
n Explain and compare redistribution and paternalism in
respect to providing a social safety net
n Describe, explain, and apply descriptive statistics in the
context of gathering and measuring information related
to public policy
n Describe, explain, and apply principles of probability
sampling and statistical inference in the context of
public policy decision making
n Describe, explain, and apply linear regression analysis in
the context of assessing information related to public policy
n Discuss and describe stakeholder analysis, mechanisms
for changing behavior, and the elements of the policy
process as aspects of effective public policy
COURSE MATERIALS
This course includes the following materials:
1. This study guide, which contains an introduction to
your course, plus
n A lesson assignments page with a schedule of study
assignments you’ll complete during the course
n Assignment introductions emphasizing the main
points in the textbook

n Self-checks to help you assess your understanding
of the material
n A research project and instructions for completing it
2. Your course textbook, Introduction to Public Policy, which
contains the assignment reading material
YOUR TEXTBOOK
Your textbook, Introduction to Public Policy, by Charles
Wheelan,
contains the material on which you’ll be tested. Success in
your course depends on your knowledge of the textbook. For
that reason, you should take some time to look through the
textbook from front to back to get a sense of how the material
is arranged. Here are some of the key features of your textbook:
Front Matter
n Brief Contents are found on pages vii–viii.
n Contents are found on pages ix–xix. Studying the contents
of the five sections of the textbook will give you a quick
overview of the learning challenges you’ll encounter in
this course.
n The Preface, found on pages xxi–xxiii, will give you an
overall sense of the author’s approach to understanding
the challenges and promises of forging effective public
policies.

Chapter Features
n Each chapter begins with an engaging exploration of a
topic related to the contents of the chapter.
n A chapter outline follows the introductory essay. Read it
carefully to understand the learning goals of the chapter.
n Within the text, key terms appear in bold type.
n Supplements under the heading “Policy in the Real World”
offer case studies covering specific public policy issues.
They’re found in every chapter.
n A Conclusion ends the main text of each chapter. You’ll
want to read it carefully as you review your understand-
ing of the chapter.
n A “For Discussion” feature follows the conclusion. It will
challenge you to think about a specific public policy
issue in considerable detail.
n Questions follow the discussion feature. They can help
you think critically and creatively about specific public
policy concerns.
n Lastly, a list of Key Concepts is found at the end of
each chapter.
End Matter
n A Glossary is found on pages 549–568. This course will
introduce you to a great number of concepts and terms.

So, beyond using the glossary as a reference when you’re
in mid-read and fuzzy about a term, you may also find it
useful to simply scan the glossary once or twice as a
form of review.
n An Index, found on pages 569–582, can help you pinpoint
names, concepts, and topics.
A STUDY PLAN
Think of this study guide as a blueprint for your course. You
should read it carefully. To receive the maximum benefit from
your studies, follow this procedure:
1. In this study guide, read the introduction to Assignment 1.
This is the first reading assignment of Lesson 1. Pay
attention to the new ideas and concepts introduced,
and carefully note the pages in your textbook where
the reading assignment begins and ends.
2. Skim the assigned pages in your textbook to get a general
idea of their contents.
3. Then, read the assigned pages in the textbook. Try to see
the “big picture” of the material during this first reading.
4. Next, go back and carefully study the assigned pages in
your textbook. Pay careful attention to all details, includ-
ing the illustrations, charts, and diagrams included in
the textbook. Take notes on the important points and
terms in a notebook, if you wish.

5. At the end of the reading assignment, review what you’ve
learned by completing the self-check questions in this study
guide. Write the answers on a separate piece of paper,
if you wish. Try to answer the questions on your own
without looking them up in the textbook. Don’t worry
about making a mistake. The purpose of answering
these questions is to review the material and to help
you recognize the areas that you may need to study
again. After you’ve answered the self-check questions,
check your answers with the answers in the back of
this study guide to confirm that you answered the
questions correctly. If you answered any questions
incorrectly, review the material for that topic until
you’re sure that you understand it. Note that these
questions are provided only for you to review your
learning. You won’t be graded on them in any way.
Don’t send your self-check answers to the school.
6. Repeat Steps 1 through 5 for each of the remaining
reading assignments in the lesson.
7. When you’ve finished reading all of the assigned textbook
pages for the lesson and you’re sure that you’re comfort-
able with the material, complete the examination for that
lesson. The examination contains a number of multiple-
choice questions. You may go back to your textbook to
review material at any time when you’re working on the
examination. When you’re finished with each lesson, take
the examination as soon as you’re ready. Don’t study
another lesson until you’ve completed the examination.
8. Repeat these steps until all lessons have been completed.

9. Complete the Research Project and submit it for grading.
Remember that you may ask your instructor for help whenever
you need it. Your instructor can answer your questions,
provide additional information, and provide further explanation
of your study materials. E-mail your questions to your instruc-
tor, and he or she will see to it that you receive the needed
information.
Now you’re ready to begin Lesson 1.
Good luck with your course!
Remember to regularly check “My Courses” on your student
homepage.
Your instructor may post additional resources that you can
access to
enhance your learning experience.
NOTES
9
A
s
s
ig

n
m
e
n
t
s
A
s
s
ig
n
m
e
n
t
s
Lesson 1: Public Policy: Why We Do What We Do
For: Read in the Read in the
study guide: textbook:
Assignment 1 Pages 11–15 Chapter 1, pages 3–31

Examination 501783 Material in Lesson 1
Lesson 2: Markets and Government
Lesson 3: Tools for Analysis

Lesson Assignments10
Lesson 4: Making Policy
Research Project 50178700
Note: To access and complete any of the examinations for this
study
guide, click on the appropriate Take Exam icon on your “My
Courses”
page. You should not have to enter the examination numbers.
These
numbers are for reference only if you have reason to contact
Student
Services.
11
L
e
s
s

o
n
1
L
e
s
s
o
n
1
Public Policy:
Why We Do What We Do
ASSIGNMENT 1
Read Assignment 1 in your study guide. Then, read Chapter 1
on pages 3–31 of your textbook, Introduction to Public Policy.
Imagine yourself living near a national park that gets lots of
snow during the winter. You’re a fan of outdoor camping and
backpacking; you respond to the “call of the wild.” So, how
might you feel about snowmobiles? Would you adore riding
them across snowy wastes? Would their noise bother you?
Would you view their riders as barbaric invaders of nature’s
precious solitude?
If you’re responsible for crafting public policy, how would you
go about reconciling the range of attitudes about snowmobiles
in national parks? Of course, at this point, that’s not a fair

question. But this scenario does tell you one thing: Making
public policy is intensely challenging. Understanding how
and why is the whole point of this course.
Defining Public Policy
To quote your textbook, “Public policy is the process by which
a society makes and enforces decisions on what behavior is
acceptable and what is not” (page 7). Basically, this formalism
tells us that public policy is all about what we, collectively, are
willing to allow other people to do. Consider these questions:
Are you in favor of legal abortion? Are you in favor of a livable
minimum wage? Do you feel your tax burden is too high, about
right, or too light? Do you agree that America has the right to
wage preemptive wars against foreign states, such as Iraq and
Afghanistan, in pursuing a global fight against terrorism?
You’ll notice two kinds of questions here. The issues of
abortion
and same-sex marriage may involve communal discourse and
debate that might lead to legislative actions. Such debates
can have a lot to do with how, collectively, we may decide on
what we think other people should be allowed to do. On the
Introduction to Public Policy12
other hand, the issues related to raising revenue (taxes) and
waging war are pretty much beyond the control of ordinary
citizens. Regardless of how people feel about tax rates, they’ll
still have to pay national, state, and local taxes. Regardless of
how people feel about preemptive wars, government foreign
policy will be what it will be, except for the possibility of mas-
sive, organized dissent. That said, it’s still the case—in every
case (at least in principle)—that public policies should aim at

making things better for people.
The Policy Process
Familiarize yourself with the five steps of the policy process:
1. Identify a social goal. Improving public education, reducing
poverty, or improving public infrastructure could be a
few, among the many, social goals you might imagine.
2. Diagnose the problem. If the issue is poverty, we’ll seek
data that helps us identify possible causes of poverty.
3. Identify the appropriate institution for action. Having iden-
tified a key cause, like unemployment, we might decide on
ways agencies of the government might offer incentives
to business for hiring more workers.
4. Evaluate the substance and politics of competing policy
options. Is it better to expand public welfare services (like
Medicaid or the food stamps program) or cut funding for
such programs in the name of balancing the budget?
5. Implement, enforce, and monitor the policy change.
During the Clinton administration, public policy programs—
national, state, and local—called for increasing the number
of cops on the beat to reduce crime and enhance public
safety. With such programs in place, enforced by government
mandates, public policy experts went about monitoring
the results, such as looking at changes in crimes rates.
Note: Refer to the graphic on page 11 for an overview of the
five steps in the policy-making process.

Lesson 1 13
For-Profit, Nonprofit,
and Government Institutions
The key idea to keep in mind here is that all of the institu-
tions mentioned in your textbook, for-profit firms, nonprofit
firms, and the government, influence and are influenced by
public policy.
For-profit firms exist to make money. They aim to keep their
expenses lower than their revenues. Most of our daily needs
and jobs come from the private sector, and private sector
operations can thrive only when they’re meeting the needs
of the people they employ and the customers they serve.
Nonprofit firms offer goods and services to people that aren’t
provided by the private sector. In this way, they contribute to
the overall social welfare. For example, university endow-
ments can be used to finance the education of students who
can’t afford to pay tuition costs.
Government institutions are unique in that they can compel
people to do what they would prefer not to do. This is true of
local, state, and national governments because, in general,
governments maintain a monopoly on force. Western states
embrace representative democracies. Too much government
can squelch and limit free enterprise and human freedom in
general. Too little government can lead to a “war of all against
all,” civil corruption, and the unraveling of the social order.
Balancing Private Life
and Public Policy
With respect to the economic face of public policy, here’s the
key idea: Our collective decisions, expressed as public policy,

create the framework for the private and corporate production
of goods and services. In short, we need public policy. On the
other hand, policy makers must always try to find the balance
between liberated human expression and the collective good.
Private, individual freedom must be balanced against the
rights and welfare of communities and society as whole.
The Art of the Possible: Life in a World
of Scarcity and Uncertainty
Most people who’ve lived on planet Earth for a while under-
stand this: Much of what we value is scarce and life is full of
unforeseen contingencies hidden in clouds of uncertainty. A
blizzard arrives, people get snowed in, and the power bill goes
through the roof.
On a larger scale, it’s simply a fact that many amenities peo-
ple might want are beyond their financial means. That’s why
families need budgets: Incomes are finite, and we can never
buy everything we might want. Yet, even our budget estimates
may be hedged about with uncertainty.
Fields of uncertainty noted in your textbook include
n Scientific uncertainty. Scientists must deal with a great
swarm of data to figure out just how climate change
will affect conditions like sea-level rise and spreading
desertification.
n Human unpredictability. Humans can be identified by
their fingerprints, because each of us has a distinctive
fingerprint pattern. In the same sense, actual individual

responses, for example, to a new zoning policy, are
inherently subject to uncertainty.
n Deliberately hidden information. As your textbook points
out, the invasion of Iraq was based on the assumption
that Saddam Hussein’s government was harboring and
producing weapons of mass destruction. In fact, no such
weapons were ever found. So the questions arise: Was the
intelligence faulty or was information deliberately with-
held from the public to carry out a preconceived plan?
n The sheer complexity of life. Your textbook discussion here
focuses on the immense complexity of economic outcomes
in the context of financial meltdowns and international
trade agreements. But you’ve probably guessed that life
is bafflingly complex simply from trying to figure out why
you bought a car you couldn’t afford or why your boss
failed to smile while passing your cubicle.
Take some time now to review what you’ve learned by
completing Self-Check 1.
Lesson 1 15
Self-Check 1
At the end of each section of Introduction to Public Policy,
you’ll be asked to pause and
check your understanding of what you’ve just read by
completing a “Self-Check” exercise.
Answering these questions will help you review what you’ve
studied so far. Please complete
Self-Check 1 now.

Indicate whether each of the following statements is True or
False.
1. In the world of public policy, adverse effects are more likely
to be significant when an
intervention has a profound effect on people’s lives.
2. The private sector is composed of for-profit firms.
3. The first step in the policy-making process is identifying the
appropriate institution for
implementing the policy.
4. Policy makers can make decisions only if they have complete
information.
5. In a nonprofit organization, surplus revenues are retained by
the organization.
Check your answers with those on page 133.
ASSIGNMENT 2
on pages 32–66 of your textbook, Introduction to Public Policy.
Public Policy Success:
Life Is Better Now
Overall, even given reversals in some areas of public welfare
in the last few years, it can be said that life is better for many,
if not for all. In the developed world, on average, people live
longer than they did several decades ago. Per-capita income

(adjusted for inflation) is considerably greater now than it
was in 1970. In many areas of the economy, people work
fewer hours per week than they did in the 1960s and 1970s.
But these are overall, general trends. They don’t account for
things like the off-shoring of jobs that’s currently undermining
the health of the American middle class. Furthermore, calls for
austerity sparked by the international financial crisis of 2008
have stirred a hornet’s nest of protest across the European
Union. In America, the “Occupy” movement protested the
accelerating gap between the superrich and ordinary wage
earners.
So Far to Go
A host of problems remain. The staggering cost of health care
isn’t declining; it continues to increase. Mounting tuition costs
are resulting in burdensome student loan debt, now approach-
ing one trillion dollars. American prisons are overflowing even
as crime rates have been in decline for the last couple of
decades. In the face of such problems, a variety of challenges
face public policy makers. In preview, here’s a quick look at 10
challenges elaborated in your textbook.
1. What is “better”? What’s good for some may be bad for
others. For workers and wage earners in general, a liv-
able minimum wage will make the world a better place.
For private sector corporate interests, raising the cost
of labor is bad news. CEOs seek maximum return on
investment (ROI) for owners and shareholders.
Lesson 1 17
2. Disagreements over basic values. Regarding the sharp
contention over abortion, imagine a balance-beam scale.

In one pan is the “weight” of opinion that a zygote
(a fertilized egg) is a person with person rights. In the
other pan is the “weight” of opinion that women should
control their own bodies. The word weight reminds
us that values are subjective, not objective “facts.”
3. Trade-offs among basic values. Again, as illustrated by
the abortion issue, the Supreme Court decision in Roe v.
Wade (1973) made abortion legal. However, despite that
decision, based on the Constitution’s concept of the right
to privacy, pro-life sentiments continue to be expressed
as legislative measures.
4. “Side-effects” and organized interests. After mass mur-
ders, such as those at Virginia Tech and a Colorado
theater, the National Rifle Association (NRA) continues
to lobby for the sale of all kinds of guns based on the
Second Amendment. Would restricting gun ownership
eliminate or even reduce such incidents? Does the NRA
in effect protect the profits of arms manufacturers by
protecting the people’s right to bear arms in spite of
public sentiment from others favoring restrictions on
access to firearms?
5. Balancing the present against the future. An example is
balancing demands for energy against rapidly accumu-
lating greenhouse gases in Earth’s atmosphere. Should
the use of fossil fuels (coal, natural gas, and petroleum)
be reduced to protect future generations from the
ecological damage, such as habitat destruction,
species die-off, and global warming?
6. Collective-action problems, or the “tragedy of the commons.”
Collective-action problems occur when accepted individ-
ual behavior results in harm to the common good. The
tragedy of the commons is a prime example. If you’re not

quite sure what the tragedy of the commons is all about,
just consider threats to food-fish populations. In many
coastal areas of the planet, people are heavily dependent
on harvesting ocean resources for food. However, as indi-
viduals and firms pursue their aims, fish stocks in many
areas are literally vanishing.
7. Nondemocratic safeguards. These happen when policies
deliberately create institutions that bypass the will of the
majority in the name of social efficiency. According to polls,
most Americans favor some kind of universal health care
scheme, such as Medicare for all. However, government
policies typically favor the special interests of for-profit,
private health insurers, pharmaceutical firms, and for-
profit hospital chains.
8. The challenge of changing human behavior. Changing
human behavior for the better isn’t easy. In any society,
once patterns of attitudes, values, and behaviors are
deemed “normal” and “virtuous,” they tend to persist.
9. Progress breeds new challenges. As many have noted,
based on historical records, once a weapon is invented,
it will be used. So far, no nation has employed nuclear
weapons, save for the United States at the end of World
War II. However, as demonstrated in the work of Ike
Jeanes and others, as proliferation accelerates, the odds
that nuclear weapons will be used increases. Today, the
United States, the Russian Federation, France, the United
Kingdom, China, India, Pakistan, North Korea, and Israel
possess employable nuclear weapons. Thus, statistically,
the global nuclear threat at present is actually greater

than it was during the Cold War.
10. Status-quo bias. This tends to keep us stuck with old
ideas, even when they no longer serve the best interests
of either individuals or firms. In a nutshell, people who
benefit from social institutions as they are will be power-
fully inclined to maintain their advantages.
Reconciling Our Differences
Values, Facts, and Theories
Values are strongly held beliefs. Indeed, they’re core beliefs
that shape a person’s view of self, others, and the world at
large. Values are based on assumptions that may or may not
be consistent with reality—whatever that may turn out to be.
But because they define one’s identity, they tend to be strongly
Lesson 1 19
defended when questioned or challenged. That understanding
can help us realize why some values, such as those associated
with a religious faith, can spark violent rivalries.
A theory is a model that more or less adequately explains the
relationships of a set of observed facts. Newton’s (or
Einstein’s)
theory of gravity is based on mathematical relationships
among measurable variables, such as velocity, mass, and
acceleration. All valid science-based theories can be modified
as new observations unveil new facts. Theories are models,
not “the truth.”
Sources of Disagreement

The four sources of disagreement given in your textbook are
as follows:
1. Differences in basic values. Sound public policy formula-
tion must not be based on the policy makers’ values. For
example, the issue of capital punishment must be viewed
in terms of “facts on the ground,” such as public sentiment,
the political atmosphere, and cost-benefit assessments.
2. Disagreements over facts or theory. Good public policy
research requires sound data and sound data analysis.
Nevertheless, studies may vary when it comes to the
question of, say, the deterrence value of capital punish-
ment.
3. Different interests. Even soundly conceived public policy
won’t please all of the people all of the time, even if the
benefits of a policy are apparent in the overall picture. For
instance, someone may decline Social Security benefits
because he or she sees them as an unwarranted form
of charity.
4. Disagreements over things that are at present unknowable.
Existential questions (Does God exist?), questions about
the future (such as global warming), and “what if” (coun-
terfactual) questions (What if the South had won the
Civil War?) may have answers—at least regarding exis-
tential and future scenarios. But we can’t know what
they are. Even so, assumed answers to such questions
can indeed influence public policy formulations.

Self-Check 2
False.
1. Only occasionally can we predict the outcomes of a policy
with complete certainty.
2. There is no single, quantifiable measure of public policy
success.
3. The so called “tragedy of the commons” is a collective-action
problem.
4. A fact is to an objective reality as a theory is to a general
principle based on available data.
5. Disagreements over basic values are most often evident with
respect to social issues,
like abortion or gun control.
Lesson 1 21
ASSIGNMENT 3
on pages 69–105 of your textbook, Introduction to Public
Policy.
Simplifying the World
The Role of Models

The most basic model for microeconomics—having to do with
the ebb and flow of collective consumer behavior—is the law of
supply and demand. As the supply of peanut butter increases,
the price of peanut butter declines, and people will buy more
peanut butter—but not indefinitely. Once the demand for
peanut butter reaches its market limit, an equilibrium price
is established. Further increases in the production of peanut
butter won’t raise the price and will result in wasted inven-
tory. Of course, that simple model is, indeed, a simplification.
Be sure to study the graphic illustration on page 73.
The Ceteris Paribus Assumption
That term ceteris paribus is Latin for “all other things being
equal.” For example, in the supply and demand model, we
assume, for purposes of analysis, that nothing changes in the
world except the demand for and the supply of peanut butter.
Assuming all things being equal isn’t an offense against logic.
In any case, good models have these characteristics:
n Good models allow us to study complex phenomena
by putting aside extraneous detail. The good model will
focus on the most important factors that lead to such
things as poverty, poor educational outcomes, or the
exorbitant costs of health care.
n Good models produce theories that can be tested by way
of analyzing data or facts.
n Good models point the way toward policies that will have
the most impact on problems, like low worker productivity,
obesity in children, or income inequality related to gender.

Individuals Seek to Maximize Their Own Utility
In the world of economists, utility has to do with people’s
perceived usefulness of particular attitudes, values, and
behaviors for providing personal well-being. You might put
it this way: People will pursue maximum happiness. To state
that idea formally, we can say that individuals will seek to
maximize their utility. However, the concept of utility is a
theoretical term. It can’t be directly observed or measured.
That’s the case because what one person sees as enhancing
their well-being may be viewed as an obstruction to happiness
by another person.
Firms Seek to Maximize Profits
A basic purpose of a corporation, a partnership, or any private
enterprise is getting maximum outputs relative to inputs of
time, knowledge, machines, and labor. Relative to public policy,
that means one can safely assume that firms won’t voluntar-
ily do work that will be unprofitable in the long run. Further,
policy makers can properly assume that the profit motive is
such a strong motivator that, absent some form of regulation,
companies will be tempted to perform in ways that are harmful
to society at large.
The Power of Incentives
An incentive can be any practice or policy that alters behavior.
Joe may be encouraged to work harder on his algebra home-
work if Dad promises him a cash reward for getting an A in
algebra. That’s a positive incentive. Your textbook recognizes
four vital points to consider when thinking about the power
of incentives:
1. Incentives motivate and explain an extraordinary range of

human behaviors. As demonstrated in the work of Nobel
Laureate economist Gary Becker, many human behav-
iors are motivated by economic factors. Individuals will
make decisions about investment in education and skill-
set training in much the same way that a business will
make decisions about allocating resources to things like
marketing, new product development, or reorganizing job
specifications.
Lesson 1 23
2. One of the most powerful tools for changing behavior is
changing incentives. From a policy-making point of view,
for example, sin taxes can raise the price of cigarettes,
thus providing an incentive that will reduce the demand
for cigarettes. As public health is improved by reduced
percentages of smokers, public health costs may also
be reduced.
3. Rational individuals and firms will seek to avoid any
outcome
that makes them worse off. Your textbook illustrates this
principle by examining some of the ways married couples
in China have tried to get around that country’s “one-child”
policy. But the main point of this principle can be gener-
alized. Given that any policy will yield winners and
losers, losers will look for any and every feasible scheme
they can use to thwart or evade an “unfriendly” policy.
4. Policies that fail to anticipate how rational individuals and
firms will respond can have serious unintended consequences.
There’s a general law in the social sciences called the law
of unintended consequences. If policies fail to recognize
how rational individuals and firms may respond to policies,

there may be unintended consequences. Examples are
offered in your textbook.
Moral Hazard
The moral hazard concept is illustrated at some length in
your textbook. But, basically, it refers to institutional situa-
tions wherein protective provisions of insurance, such as car
or flood insurance, may invite a reduced sense of responsibility
and responsible diligence on the part of a policy holder. For
example, if you know your car is insured for near or complete
replacement, you may be less inclined to exercise precaution-
ary measures, like locking your car when you’re parked in
an urban neighborhood. Fundamentally then, a moral
hazard is a financial hazard for insurers.
Incomplete Information
A situation wherein one party to a transaction has more
information than the other party represents an asymmetry
of information. For those who craft public policy, this sort of
information imbalance is relevant because it obstructs the
proper functioning of a market. “Lemon laws” to protect con-
sumers who get stuck with improperly manufactured cars
and regulations requiring packaged or canned food to include
a list of ingredients are examples of policies meant to address
information asymmetry.
Principle–Agent Problem
Imagine that you’re the owner of a small publishing business.
You employ a chief editor, two editorial assistants, and an

administrative assistant. Being human, you expect your employ-
ees to support and further the interests of your business.
However, you discover that one of the editorial assistants,
Drake, is padding his expense account while bad-mouthing
you to friends who work for a rival publishing company. In
an employee–employer relationship, you, the owner, are con-
sidered the principle and your employees are agents of your
firm. This imaginary scenario illustrates two characteristics
of a principle–agent problem.
First, the agent—Drake—is clearly motivated to behave in ways
inconsistent with your interests. Second, because you have
all you can do to sort through submitted manuscripts and
queries, you have no way of closely monitoring the behavior
of your editorial assistants. And there’s the problem.
Your textbook offers you some interesting insights into prin-
ciple agent problems that were inherent in the Wall Street
financial crisis of 2008.
Adverse Selection
The concept of adverse selection is most readily understood
through an example. Let’s assume that health insurers aren’t
able to gather as much information about policy applicants’
health as is known to the persons seeking health care coverage.
Here, we immediately see a case of asymmetrical information.
Lesson 1 25
Adverse selection enters the picture as people with the most
precarious health profiles most actively seek coverage. When
a high percentage of policy holders are at risk for serious
medical care, the insurance companies are placed at an actu-

arial disadvantage. As a result, premiums may rise across
the board for all the company’s policy holders. And, as an
added unintended negative consequence, policy holders may
be driven out of the market, thus increasing the total number
of uninsured persons.
Mechanisms for Overcoming
Information Problems
In a market economy, information problems can wreak havoc.
If, for example, a home buyer isn’t informed that lead-based
paint has been used throughout the house, he or she may be
committing to a mortgage that purchases serious health risks
for children and pets. In these kinds of contexts, a number of
governmental policies have been established to protect con-
sumers. Nutrition and ingredient information is required on
packaged foods. The Food and Drug Administration (FDA)
provides appropriate policies for labeling pharmaceuticals.
But given that the enforcement of well-intentioned policies
may not always suffice, there are other ways to overcome
information problems. Your textbook discusses four ways
to overcome information problems:
1. Branding. You may not be able to tell an alternator from
a fan belt, but if you’ve had good experiences with Hondas
or Fords, the brand itself will reassure you. Companies
spend lots of time, energy, and money building up brand
images.
2. Signaling. A “signal” like a college degree or Phi Beta
Kappa key may have no intrinsic cash value, but it
can convey a message. When employers are in search
of competent, capable employees, such signals may
make all the difference.

3. Certification. Information provided by expert third parties
can do a lot to overcome information deficits on the part
of individuals or organizations. Car and Driver magazine,
the 75-year old grandfather of consumer information,
Consumer Reports, and the Underwriters Laboratory are
invaluable resources when it comes to providing accurate,
timely information to people who want to know.
4. Screening. Car insurers offer a good example of screening.
As drivers establish a sound record based on a pronounced
absence of accidents, DUIs, and traffic violations, they
may become increasingly eligible for reduced policy
premiums.
How Rational Are We?
Here you’ll learn a bit about a budding specialty in the domain
of economics—behavioral economics. You’ll discover some
odd,
fascinating experimental results. You’ll learn a bit about the
way human risk assessments may be unrealistic. You’ll get a
few clues as to the kinds of mistakes people make in setting
money aside for retirement. In short, all things considered,
you’ll be challenged to think again about the accuracy of the
assumption that people are rational actors striving to maximize
their utility.
Lesson 1 27
Self-Check 3

False.
______ 1. For policy analysts, the law of supply and demand is
considered a model.
______ 2. The most serious adverse selection problems occur in
the food-processing industries.
______ 3. According to research findings, revenge offers
satisfaction even if it doesn’t make one
better off.
______ 4. The principle-agent problem illustrates one kind of
imperfect information problem.
______ 5. According to Becker, the fact that human labor can
be monetized defines the concept of
human capital.
______ 6. Perverse incentives are associated with the law of
unintended consequences.
ASSIGNMENT 4
on
pages 106–138 of your textbook, Introduction to Public Policy.

Externalities
A stark example of costly negative externalities is the polluting
runoff from factory farms. The runoff from factory farms is
contaminating a large percentage of aquifers in the Midwest
and elsewhere. Aquifers are underground strata of clay, shale
and other natural formations that accumulate and store water.
That water, drawn up from wells, in turn, supplies the fresh
water needs of millions of people.
On the other hand, there can be positive externalities. A
group of investors may decide to construct the Meadowbrook
shopping mall in Blankenship’s town square. Of course the
investors will hope to profit from their venture. But in this
case, there can be a flow of positive side effects. New jobs will
appear. Property values near the mall will increase. Taxes
derived from increased local incomes can be made available
for community needs, like playgrounds, public parks, and
recreational centers.
With respect to public policy, the key to negative externalities
may be regulation. But with respect to positive externalities,
public policy may be aimed at providing incentives for investors
to create Meadowbrook-like ventures in other towns and
localities.
Property Rights
For a market system to perform effectively, the legal concept
of private property must be clearly defined and enforced. In
short, the concept of private property is fundamental in a
free-enterprise system. When we think about private property,
we mostly think of land, buildings, and other kinds of “hard”
assets. But private property also extends to intellectual
property,

such as patents, copyrights, trademarks, brands, original lit-
erary works, film option rights, and so on. In effect, anything
that can be ascribed value may be defined as private property.
All of this is relevant to public policy because it’s important
to understand how groups go about allocating resources.
Lesson 1 29
Transaction Costs
Transaction costs are related to public policy in two impor-
tant ways: (1) Transaction costs can be a serious obstacle to
transactions that could make all the parties better off. For
example, in New England, some towns and localities straddle
the Canada–United States border. If the owner of a farm that
straddles the border wants to build a new barn, that project is
likely to be obstructed by the need to reconcile distinct zoning
ordinances on either side of the border. (2) Good public policy
can aid in lowering transaction costs. Courts can lay out
clear stipulations for enforcing contracts. Government water
authorities can assist community commissions by providing
information that would be hard to gather by private citizens.
The Coase Theorem
Ronald Coase received a Nobel Prize in 1991 for figuring out
how private property and transaction costs influence that
structure and dynamics of modern economies. There are dif-
ferent interpretations of the Coase Theorem. Your textbook
focuses on the idea that negative externalities like pollution
can be abated or ameliorated if the market is characterized
by clearly defined property rights and low or nonexistent
transaction costs. Presumably, this would be because a non-
polluting company being impinged on by an adjacent polluting

company will be advantaged in terms of relative social
efficiency.
Put another way, when a nonpolluting company contends
with a polluting company, the nonpolluting company can buy
out and shut down the polluter because it offers more value
to the economy at large.
Strategic Interaction
This section introduces you to game theory. In the context of
public policy, game theory examines ways in which strategic
interactions among economic agents produce outcomes (pref-
erences and utilities) that might not have been expected by
any of the agents. The complex mathematical formulations
that drive game-theory models in many areas, from strategic
corporate planning to strategic foreign policy determinations,
are beyond the scope of this course. However, the logic of game
theory can be represented by the prisoner’s dilemma game.
The Prisoner’s Dilemma
Imagine this scenario: Jake and Calvin have been charged
with armed robbery. At police headquarters, the two men are
interrogated separately. They have no opportunity to interact
with each other. While being interrogated, each of the men
has two options: confess or stay quiet. However, each of the
culprits understand that the outcome of his choice—confess
or don’t confess—will depend on whether the other man con-
fesses or doesn’t. Jake and Calvin may be failed bank robbers,
but they’re not stupid. As you study Figure 1, you’ll see that
if both Jake and Calvin confess, each will do 20 years in the
slammer. If Jake confesses and Calvin doesn’t, Jake will get a

light 3-year sentence for cooperating, and Calvin will be
slugged
with a 30-year sentence. If neither Jake nor Calvin confesses,
the case against the two will remain largely circumstantial
and both will end up with 5-year sentences. Finally, if Calvin
confesses and Jake doesn’t, he’ll get the 3-year sentence
while Jake gets 30 years.
Remember the primary idea here. In a prisoner’s dilemma
game or in similar games, the agents (actors) produce out-
comes that are unexpected by the players. In the prisoner’s
dilemma, self-interest dictates that each player will adopt a
dominant strategy: confess. Come clean, and go for a plea
bargain for being so nice and cooperative. Yet, from an out-
side perspective, that’s clearly not the best strategy. The
heaviest aggregate prison term, 20 + 20 = 40 years is what
these guys get when they both choose the confession strategy.
Clearly, the best strategy, in terms of aggregate time in
prison, occurs when neither man confesses.
CALVIN
Confesses Doesn’t Confess
JAKE
Confesses 20, 20 3, 30
Doesn’t Confess 30, 3 5, 5
FIGURE 1—An Example of the Prisoner’s Dilemma
Lesson 1 31

An Arms Race
Something like a prisoner’s dilemma happens in an arms
race. Take two irritable nations—Nation X and Nation Y. They
share a common border and both are moderately armed in the
name of self-defense. Say Nation X begins to acquire guns,
tanks,
and bombers, on a large scale. Nation Y becomes uneasy. If
Nation Y responds with an arms buildup, the likelihood of war
is likely to increase. On the other hand, if Nation Y decides
against an arms buildup, the country becomes vulnerable; it
could be overrun. The ministers and dignitaries of Nations X
and Y could meet somewhere and rationally determine that
war is a lousy idea. It’s expensive. The arms race scenario
could be cancelled in favor of diplomacy, joint trade ventures,
and other agendas of common interest. Nations X and Y
could decide on a positive-sum scenario.
Zero-Sum, Positive-Sum, and Negative-Sum Games
In economic contexts, strategic interaction “games” can be
sorted into three categories.
In a zero-sum game, the size of the resource pie is fixed.
Imagine that players are competing for a larger slice of a fixed
budget. Under such a scenario, if the department of high-
ways manages to increase their share of the budget, funding
for other priorities, like education, will be reduced. One party
may grab more of the pie, but overall, the sum of the slices of
the resource pie remains the same.
In a positive-sum game, the size of the resource pie can be
increased, given certain conditions. For example, in a contest
between workers and management, the size of the pie might
be increased by finding ways to cut nonlabor costs and
increase productivity. If that works out, both parties benefit.

In a negative-sum game, the payoff gets smaller over time.
For example, a union strike may aim at better pay and work-
ing conditions even as the market share for an industry is
contracting. The “game” will be aimed at grabbing the largest
possible slice of the contracting resource pie.
Dynamic Behavior
Human behavior is dynamic. Individuals make choices, day
by day, hour by hour. As that happens, group behavior is
altered. If the Joneses place a plastic deer on their front lawn,
others in the neighborhood may buy lawn ornaments. Or the
neighborhood association might meet to condemn plastic deer
and other sorts of lawn decor. In any case, you’ll want to be
able to explain three kinds of dynamic behavior.
A negative feedback loop is also referred to as a vicious cycle.
Your textbook offers the example of a bank run, which was
fairly common before the Roosevelt Administration introduced
the Federal Deposit Insurance Corporation (FDIC). The FDIC
insures bank deposits (for up to $250,000 currently) and, thus,
keeps most banks solvent as economic tides wax and wane.
A positive feedback loop is the opposite of a negative feed-
back loop. Maybe you recall the film Pay It Forward. The
film’s protagonist, a young boy, decides to do good deeds to
inspire others to act with compassion and kindness. The film
ends in tragedy, even though some others decide to “pay it
forward.” Socially positive behavior, economically speaking,
can inspire others to pursue socially positive behavior.

An equilibrium situation occurs when system changes, such
as price changes or changes in the labor market, don’t upset
the system’s overall inputs or outputs. To illustrate how the
market system can be “self-correcting,” your textbook uses
the example of an increase in the cost of tomatoes. The mar-
ket will respond as people buy fewer tomatoes in favor of
alternative vegetables. The decrease in demand for tomatoes,
in turn, will tend to restore the former equilibrium price level
for tomatoes. Supply and demand will be “rebalanced.”
Collective-Action Problems
At this point, you’re familiar with the concept of collective-
action problems with respect to individuals. Here, the focus
shifts to groups. Thus, if we can assume that individuals will
rationally pursue behaviors that will make them better off, so
will groups. Of course, that assumption fails when people are
engaged in a prisoner’s dilemma or similar situation.
Lesson 1 33
On the other hand, experiments and studies have demonstrated
the reality of the key collective-action problem—shirking.
Individuals who share in the benefits of collective action with-
out doing their share of the work are called free riders. Free
riders are parasitic. And if more and more people decide to be
free riders, the benefits offered by a group will be diminished,
even for the shirkers.
A phenomenon called the Ringelmann Effect illustrates
collective-action dynamics. In 1913, a French agricultural
engineer, Maximilien Ringelmann, conducted experiments
based on rope-pulling contests. He found that as people were
added to each team, individual effort declined as a precise

function of group size. And so, when you learn from your
textbook
that free rider problems increase as groups get larger, you’ll
now know just how this has been empirically demonstrated.
You can add that to the other two reasons that encourage
free riders—efficiency and monitoring problems. As groups
get larger, individuals will perceive their contribution (or non-
contribution) to the group’s goals as negligible. Furthermore, as
groups get larger, it gets increasingly difficult to monitor indi-
vidual behavior. Two heads may be better than one, but 100
or 1,000 heads may simply be a road to productive inefficiency.
Note: In your textbook, you’ll want to pay careful attention to
the ways in which the tragedy of the commons, a concept devel-
oped by Garrett Hardin, illustrates collective-action problems.
Your textbook offers you six strategies that may serve to alle-
viate collective-action problems. For example, regulations,
like rules for fisheries and fishermen, can be helpful, at least
if the individuals involved can grasp the need for the rules.
Government services can provide a public good through services
like park maintenance, fire-fighting, and sanitation services,
all of which are paid for through taxation. Be sure to think
about all six of these strategies to consider when or how they
might be right for a particular situation.
Nobel Ideas: Elinor Ostrom
and Governing the Commons
Elinor Ostrom, a Nobel Laureate in economics, wrote a book
titled Governing the Commons: The Evolution of Institutions
for

Collective Action. That title should give you a hint as to the
concerns she has addressed. Ostrom compels us to understand
that users who depend on shared resources have developed
effective ways to protect that resource for the common good.
She
proposes seven principles for effective common-pool resource
management. Before you get to the list of seven, here are
three that stand out:
1. Clear rules should define who has what entitlement.
2. Adequate mechanisms should be in place for resolving
disputes and conflicts.
3. Governance of the resource should be democratic.
It’s nice to know that solutions for some of our most pressing
problems are being tried and tested.
Lesson 1 35
Self-Check 4
Match each statement with the correct term. Not all of the terms
will be used.
______ 1. Strategic interactions include an arms race or the
prisoner’s dilemma.
______ 2. People who depend on a shared resource often
create rules and other mechanisms that preserve

and protect the resource.
______ 3. This is the legal right to exercise control over
some resource.
______ 4. Externalities can be corrected without government
intervention if property rights are clearly defined
and transaction costs are low.
______ 5. There are effective strategies for alleviating
collective-
action problems given particular situations.
______ 6. When an organization has a fixed budget,
funding one program will reduce funding
for some other program.
a. Zero-sum game
b. Coase Theorem
c. Transaction costs
d. Property rights
e. Game theory
f. Negative-sum game

g. Common-pool resources
h. Regulate, privatize,
selective incentives
ASSIGNMENT 5
on
What Is Poverty? And Who Is Poor?
Becoming more aware of one of this world’s most pressing
problems—poverty—will sharpen your focus on the main text.
Study the comparative poverty levels for the mainland United
States, Alaska, and Hawaii in the table on page 139. Then
note the ambiguity inherent in the term poverty. You’ll see
that the “market basket” concept of poverty was based on
rough estimates provided by the Department of Agriculture
decades ago. The estimates summed up the amount of money
needed to provide a person with an adequate, nutritious diet.
For a family of three, that estimate was multiplied by three, and
so on. However, you’ll see that significant variables weren’t
accounted for in this calculation, such as government subsidies,
like food stamps or the fact that standards of living, after
accounting for inflation, have increased over time. Moreover,
you’ll discover that commonsense revisions for estimating
poverty thresholds proposed by the National Research Council
have never been adopted.
Measuring Social Welfare

From an individual’s perspective, the previously introduced
concept of utility can be thought of as a good. People will be
guided by the Freudian pleasure principle; they’ll try to maxi-
mize the “goods” in their lives while minimizing the “bads.”
But, given the way poverty rates are estimated, how can a
policy maker aim to maximize social welfare without any
objective measure of social welfare?
Indicators
Ideally, indicators of social welfare should be easily measurable
and strongly correlated to the underlying variable of interest,
such as poverty. In that context, even though the concept of
poverty is fuzzy, it can be useful. For example, the measure
can help policy makers determine who should be eligible for
Lesson 1 37
government assistance programs. It can also give us a sense
of trends. If poverty rates are increasing or decreasing, we
can look for reasons why this is the case. Meanwhile, we can
use comparative indicators, like life expectancy, to examine
relative well-being in different countries.
Indexes
Indexes combine several indicators that would seem, logically,
to contribute to social well-being. The human development
index (HDI), developed by a Pakistani economist, is a prime
example. The HDI looks at
n Life expectancy at birth

n The adult literacy rate
n The population percentage of enrollment in primary,
secondary, and tertiary schools
n The per capita gross domestic product (GDP)
Being able to reduce the social well-being measure to a
single number is the upside of an instrument like the HDI. The
downside is the subjective uncertainty inherent in assigning
a weight to any of the HDI indicators. If, for example, the GDP
per capita measure is given greater weight, HDI comparisons
across countries would be altered to some extent.
Note: In the table on page 147, you’ll note that the United
States isn’t included among the top 10 highest HDI ratings.
Norway, Australia, Iceland, and Canada are at the top. Reasons
for this include wide variations in social well-being across
regions and social classes and the exceptionally large wealth
gap between the wealthiest people and the middle and lower
classes in the United States. On average, the high HDI coun-
tries feature fairly homogeneous populations and, typically,
strong social safety nets.
You may wonder how a rural farmer in Zaire or Bolivia can
live on less than $2.00 a day. The answer is simple: They can’t.
In such places, absolute poverty may be alleviated to some
extent by immediate access to subsistence resources and
informal trade and barter. But these variables exist “outside”
the parameters one applies in a money-measured economy.
The Philosophy of Well-Being

The essence of this section boils down to exploring a continuum
of economic systems based on distinct philosophies of well-
being. A continuum is proposed with two bookends—Marxism
and libertarianism. The Marxist ideal was “from each according
to his abilities to each according to their needs.” However,
following the Bolshevist Revolution of 1917, the actual
economic
system that emerged in the Soviet Union imposed central
economic planning by way of brutal police-state tyranny,
particularly under the iron-fisted rule of Josef Stalin.
Ideologically,
more humane communist governance has emerged here and
there. Your textbook outlines the fascinating development of
the communist-ruled state of Kerala in India. Interestingly,
while that state ranks poorly on the GDP measure of the HDI,
it performs admirably in areas like life expectancy and infant
mortality. That fact suggests that hard and fast boundaries
between communist and capitalist ideals aren’t easily drawn.
For example, in Western European countries like France,
Germany, Norway, and Finland, socialist policies in respect
to education, workers’ rights, and health care, exist side by
side with robust market-driven economies.
Be that as it may, it’s appropriate to view libertarianism as the
antithesis of communist (or socialist) ideologies. Libertarian
economic philosophy, expressed in the ideas of Friedrich
Hayek and Milton Friedman, argue strongly against undue
government interference with the market. For libertarians,
people pursuing private property and individual self-interest
in the marketplace generates the highest level of personal
freedom and the most effective system for allocating resources.
Note: Economics, like psychology, biology, and physics, is
guided by theories. Recall that theories are models based on
sets of observations. Theoretical models are valid only if they

can be revised or rejected on the basis of new information or
alternative conceptions of basic things, like efficient or ineffi-
cient market mechanics. However, because humans aren’t
omniscient; their observations will be filtered through basic,
often unquestioned, assumptions. For example, the concept of
“economic man” maintains that people are motivated by self-
interest guided by rational calculations that weigh advantage
Lesson 1 39
against disadvantage. Other approaches to the production
and allocation of resources may be based on different
assumptions about human motivation. To explore that idea,
be sure to read about the Bhutan’s concept of gross national
happiness on pages 148–150.
Tools for Making Inferences
about Social Welfare
Efficiency
The extent to which resources are used to generate produc-
tive outcomes is the extent of efficiency. If crop rotation and
erosion control generates lush harvests of corn, wheat, or
soybeans, these practices are economically efficient. If we
use our time efficiently, we’re more likely to be productive.
Note: Be sure to examine the three interesting examples of
inefficiency offered to you on page 155.
If there’s one topic in this section that may lead to confusion,
it’s
the ideas about resource allocation attributed to the Italian
economist Vilfredo Pareto. Be prepared to understand that

the allocation of resources is Pareto-efficient when it’s not
possible to make anyone better off without making someone
else worse off. By contrast, where we find Pareto-inefficiency,
it’s possible to make one or more person better off without
making anyone else worse off. Clearly, Pareto-efficiency
doesn’t
claim that life is fair and equitable. It does mean that where
people pursue economic self-interest, there must be winners
and losers. In any case, it’s suggested that you carefully pur-
sue your textbook’s elaborations on these ideas to sort out
their meaning. Meanwhile, keep in mind that your textbook’s
discussion here is meant to help you grasp the idea that, in
spite of Pareto’s principle, the most efficient tool we have for
allocating resources is the market.
Deadweight Loss
Simply put, deadweight loss refers to the net result of economic
inefficiency. The example expanded on here is the issue of
the minimum wage. For some, the minimum wage expresses
the ideal that workers should earn wages sufficient for a
modest but livable lifestyle. For others, a minimum wage cre-
ates deadweight loss because employers may not profitably
pay the minimum wage to dishwashers and similar workers.
As your textbook sums up the matter, a minimum wage, at
whatever level, will exclude job seekers willing to accept low
wages, thus creating some degree of deadweight loss.
Equity
The measure of a public policies’ fairness is its level of equity.
While keeping in mind that most people favor fairness, you’ll

want to recognize three different kinds of equity.
Horizontal equity has to do with the extent to which people in
similar circumstances are treated the same way. For exam-
ple, in the workplace, people who do the same kinds of jobs
should receive the same level of pay. In academic venues,
people who get the same score on a test should get the same
grade.
Vertical equity can point up or down. For example, looking
upward, the level of one’s tax burden should reflect the level
of one’s income. Looking downward, the poorest people should
receive the higher levels of government subsidies.
Intergenerational equity relates to public policies that should
remain consistent from generation to generation. A public
policy discussed here is Social Security. The Social Security
fund comes from direct payment of payroll taxes. So here’s
the equity problem: As our society ages and more and more
people reach retirement age, the payroll taxes on younger
people may have to be raised if they’re to receive equivalent
Social Security benefits when they retire. Is that fair? If not,
what are the alternatives?
Lesson 1 41
The Efficiency-Equity Trade-Off
Economic efficiency may be perceived as equitable or
inequitable. Social mechanisms perceived as equitable (fair)
may be seen as efficient or inefficient. Prepare to think about
this conundrum as you tackle this section. As you do, be ready
to ponder the public policy challenges involved in balancing
equity and efficiency. For now, consider these points:

n An efficient market rewards only people who can offer
something translatable into monetary value. So, an effi-
cient market may well produce paradox when it comes to
equity. At one end of the reward spectrum, the efficient
market may leave some people with no income at all. At
the opposite end of that spectrum, a few may be rewarded
at levels which must seem inaccessible to ordinary people
struggling with bills and taxes.
n A market economy will distribute resources efficiently
based on any person’s initial endowment. Your initial
endowment will include your talents, interests, education,
and social class. In our society, initial endowments vary
enormously as a function of social class, race, ethnicity,
and gender.
n Inequity provides powerful incentives. People who want
to reduce the gap between what they have and what they
want are encouraged to work hard, innovate, create, and
compete. On the other hand, equity may be too much of
a good thing. For example, generous unemployment com-
pensation may discourage a person’s active pursuit of a
job, thus harming the economy.
Absolute versus Relative Well-Being
Absolute poverty is measured in terms of basic needs, includ-
ing food, shelter, and health care. Relative poverty has to do
with how people fare in comparison with other people within
their neighborhood, work place, or society at large. For public
policy makers, an absolute poverty measure provides means

for assessing the extent to which a society is meeting the needs
of a polity. By contrast, measures of relative poverty can be
fuzzy. Hard-to-measure psychological factors enter the picture.
For example, a sense of relative deprivation may be quite dif-
ferent for a hedge fund manager, a social service worker, a
cop, or a plumber. Furthermore, individuals’ attitudes and
values can be radically different. For some, making do with a
modest income may be seen as responsible citizenship. For a
brain surgeon, being excluded from the “Ferrari-and-yacht”
class may be viewed as bitter evidence of relative poverty.
When you’ve reviewed Lesson 1 and you feel confident that you
understand the material, complete the Lesson 1 examination.
Lesson 1 43
Self-Check 5
Fill in the blank with the best word to complete the statement.
1. Under _______ governance, the Indian state of Kerala is
distinctive for healthy indicators of
social welfare in spite of modest wealth.
2. As an illustration of _______ equity, if Family A has the
same income as Family B, the two
families should have the same tax burden.
3. A combination of indicators is a/an _______.
4. In a market economy, one’s life chances will be related to
one’s _______, which may include
talent, health, parental income, and old-fashioned luck.

5. An allocation of resources is _______-efficient if it’s
impossible to make one individual better
off without making another individual worse off.
6. As a measure of inefficiency, _______ loss is observed when
the loss of welfare of one party
exceeds the gain in welfare of another party.
NOTES
Markets and Government
ASSIGNMENT 6
Read Assignment 6 in this section of your study guide. Then,
read Chapter 6 on pages 177–213 of your textbook, Introduction
to Public Policy.
Aggregating Preferences
To aggregate means “to gather together.” In the political
process, aggregating preferences means gathering together
and organizing the diverse opinions and beliefs of a group to
forge some kind of consensus. In this context, the study of
comparative politics explores how different kinds of political
systems effectively or less effectively aggregate preferences
around the world.
Direct and Representative Democracy

Direct democracy requires direct participation and decision
making in and for a community. The first experiment in
democracy, the citizen assembly of Athens, was a direct
democracy which, in Plato’s view, had all the finesse and
character of a mob. (On a side note, Socrates was condemned
to death by this “mob” for corrupting the youth of Athens.) In
America, direct democracy thrived in the immediate
postcolonial
environment due to the influence of citizen assemblies. Echoes
of
that era remain in the traditions of New England town meetings.
However, direct democracy becomes unwieldy as populations
grow larger. One can hardly expect the citizens of Trenton,
New Jersey, to gather in some venue like a sports arena to
debate issues, resolve differences, and pass laws. So, simply
as a practical matter, methods of representative democracy
were established early on.
45
L
e
s
s
o
n
2
L
e

s
s
o
n
2
Methods of Assigning Representation
In a representative democracy, the rules for allocating repre-
sentation in a legislative body will have a profound effect on
the relative power of the different groups being represented.
The framers of the U.S. Constitution were deeply concerned
with this issue. Based solely on proportional representation,
the larger states, like Virginia and Pennsylvania, would drown
out the voices of small states like Delaware and Rhode Island.
The final resolution of this contention proposed proportional
representation in the House of Representatives and equal
representation in the Senate. Namely, there would be two
senators from each state, regardless of its population size.
Of course, there are other approaches to assigning
representation.
In an at-large legislature, say for a county or city, the mem-
bers of a council will be directly elected by the voter-eligible
citizens of that polity. In such a system, all of the elected
council members will represent all of the citizens. In the case
of single-member legislative districts, as illustrated by the U.S.
House of Representatives, people can elect only one candidate

to represent their congressional district.
On pages 184–187, your textbook offers five hypothetical
cases to illustrate likely political outcomes under five types
of voting rules. You should pay special attention to the fourth
case, in which the electoral districts have been subject to
gerrymandering. This term derives from the actions of then
Governor of Massachusetts, Elbridge Gerry. In 1812, he
redrew the boundaries of the state’s electoral districts to
favor the interests of his Democratic–Republican Party prior
to a state senate election. Today, efforts of both Republicans
and Democrats to redraw congressional districts to “lock in”
advantages to their party have effectively biased election out-
comes in a number of states.
Deliberately Nondemocratic Institutions
Within democracies, some institutions are deliberately insulated
from the slings and arrows of electoral politics. The concept
of the separation of powers gets a lot of attention here. The
political philosopher Montesquieu invented the concept during
Lesson 2 47
the era of the Enlightenment. It was adopted by those who
framed
the U.S. Constitution in 1789. Distinguishing a legislative,
executive, and judicial branch of government was meant to
restrain the power of the central government in relationship to
the states. In that context, however, the democratic process
was “diluted” to some extent, largely to assure greater free-
dom of action on the part of high-level government officials.
To that end, Senators (mostly drawn from the upper classes)
are elected to six-year terms as opposed to two-year terms, like

the members of the more populist House of Representatives.
The president, as the head of the executive branch, is freed
up to either embrace or ignore populist demands if he or she
can manage election to a second term. Federal district and
Supreme Court justices, appointed by the president, hold
their office for life.
Arrow’s Theorem
One can assume that, in a democracy, public policy options
will be related to social welfare concerns. Therefore, all voting
systems are meant to aggregate the preferences of many indi-
viduals into an option preferred by the whole. But is that
actually possible in every instance? According to the work
of Nobel laureate economist Kenneth Arrows, it isn’t. If three
or more options are on the table, no social welfare function
(purpose) can aggregate individual preferences into a collective
group preference without violating one or more basic condi-
tions that should characterize democratic decision making.
The basic conditions are as follows:
n Unrestricted domain. The process should take account of
all rational individual preferences and allow them to be
ranked in preferential order. If, for example, there are
three abortion-procedure options (A, B, and C), they
can be ranked in some way, such as B, C, and A.
n Completeness. The voting must deliver a clear answer,
such as B as the clear preference.
n Transitivity. If B is preferred over C, it must also be
preferred over A. (If a number, N-1, is greater than N-2,
and N-2 is greater than N-3, then N-1 must be greater
than N-3.)

n Pareto optimality. If every individual in a group prefers B
over C, then the group (collectively) must also prefer B
over C.
n Nondictatorship. The social welfare function can’t simply
reflect the preferences of one individual while disregarding
the preferences of the remaining voters.
n Independence of irrelevant alternatives. If B and C are the
dominant options (as in Republicans or Democrats), then
votes for alternative A (such as a third party) shouldn’t
affect the final showdown in the preference contest
between B and C.
Note: Arrow’s theorem is expressed through complex mathe-
matical assumptions and algorithms. Simply try to get a
general feel for these ideas.
Here’s the key idea derived from Arrow’s theorem: Democratic
decision making won’t automatically aggregate individuals
into group preferences. Instead, the key factor is the design
of institutions. For example, with respect to policies regarding
abortion (trimester rule, medical procedure, and so on), the
option agreed on will have more to do with legislative rules
and procedures than with the aggregation of individual
preferences.
The Legislative Process
Parliamentary versus Presidential Systems
Parliamentary systems, like those in the United Kingdom,
Canada, and Western Europe, are more common than presi-

dential systems. In parliamentary systems, the effective head
of the government is the prime minister. There are exceptions.
For example, the chief executive in the Russian Federation is
an elected president who, in turn, appoints a prime minister
with the approval of the Duma (the bicameral Russian legisla-
tive body). For Americans who have access to C-Span, the
nature of most parliaments is well illustrated in the proceedings
of the British House of Commons. In the United Kingdom, the
prime minister is elected by the members of the Commons,
Lesson 2 49
and, in most cases, the prime minister is the head of his or
her party. Margaret Thatcher, for example, was the head of
the Conservative (Tory) Party. Tony Blair was the head of the
New Labour Party. Once elected, in a sort of medieval ritual,
the queen charges the new prime minister with the duty of
forming a government. The prime minister then appoints
assorted ministers to administer the government.
In a fair number of cases, as in Italy, the prime minister’s party
may not hold a majority of seats in the parliament. In such
cases, the prime minister must form a coalition government,
which may quite often be politically unstable. Also, in the
United Kingdom and elsewhere, a prime minister may be
voted out of office by the Commons MPs through a vote of
no confidence.
The British upper house, the House of Lords, is presently
composed of “peers of the realm,” whose seats are inherited
as a right of noble title, and appointed members. However,
under Tony Blair, a movement began to eject entitled and
appointed lords in favor of democratically elected members.
Traditionally, the lords may propose legislation, but more

often, they serve as a critical overseer of legislation originating
in the Commons. An appointed body within the House of
Lords, informally called the Law Lords, serves as the equiva-
lent of a supreme court in the British judicial system.
Perhaps the greatest strength of a parliamentary system is
providing a legislative voice for minority parties, and also a
capacity to remove an ineffective prime minister who has lost
the confidence of the people through a vote of no confidence.
Rules, Committees, and “Gatekeepers”
The American constitutional republic is composed of the three
branches of government mentioned earlier. In your textbook’s
discussion of the American system, a great deal of emphasis
is placed on the nature of the rules and procedures in the
House and Senate. The point of that emphasis is to help us
understand that these rules and procedures have a major
impact on the formulation and enforcement of public policy.
For example, the House Rules Committee exercises control
over the legislative calendar. Decisions by that committee
determine the order in which bills are presented to the House
for consideration as well as the number of amendments that
may be proposed or adopted. The Rules Committee is one
among 25 standing committees in the House. The immensely
powerful Senate Finance and Senate Armed Services committees
are among the 20 standing committees in the Senate.
Most of the work of the U.S. Congress, or in state legislative
bodies for that matter, is carried out by deliberations within
committees. For example, most bills (proposals for a law)

submitted to the House or Senate will be sent on to a com-
mittee which, in turn, is enabled to hold hearings and/or
revise and “polish” the bill (referred to as marking up a bill)
before allowing it to either “die in committee” or be reported
out to the floor for general deliberation. In the House, the
elected Speaker of the House, representing the majority
party, is responsible for sending bills to one of the various
House committees. So, in effect, the Speaker of the House
and the various committees act as gatekeepers who deter-
mine which bills will be considered and in what order.
After the vice president, the Speaker of the House is next in
line to ascend to the presidency. Therefore, he or she has an
immense influence on legislative actions in the House. There
is no equivalent role in the Senate. The Senate’s majority
leader certainly serves as a gatekeeper. However, the role
of the President Pro Tem of the Senate, held by the vice
president, is mainly ceremonial. He or she may cast a
deciding vote in the event of a tie, but otherwise is seldom
even in attendance for Senate deliberations.
Note: Be sure to study the graphic on page 195. It will help
you understand how legislation is passed in the U.S. system
of governance.
Perhaps the greatest strength of a presidential system is
provided by a strong executive who can stand apart from the
legislative body. The greatest weakness of the U.S. system is
the risk of legislative “gridlock” when the president and the
Congress represent opposing political parties.
Lesson 2 51
Understanding Political Outcomes

The Median Voter Rule
Most people who pay attention to politics will recognize
evidence
of the median voter theory. Namely, to gain a majority in an
electoral contest, politicians will strive to attract the voters
who hold centrist public policy views. If, for example, the
issue is funding education initiatives, a politician will try to
attract voters who are in favor of moderate education-budget
increases. That is, they stand between the extreme views that
would reject all such spending on one hand or opt for major
education-budget increases on the other hand.
However, in the view of public policy analysts, the median
voter rule can apply only given a set of unequivocal and
specific assumptions. They include the following:
n There are only two candidates. If third-party interests are
influencing public opinion, the rule won’t apply.
n Voters have single-peaked preferences. That is, the
median voter position is expressed by a significant
majority of voters.
n Voters care about just one issue. For example, in some
electoral districts, the education-funding issue may be
competing with voter concerns about crime rates or the
need to invest in roads and bridges.
n All voters are equally informed and engaged. Poll-based
research suggests that many voters are either uninformed,
indifferent, or both.
The Power of Organized Interests

In the real world of politics, the median voter rule is often
contradicted by actual electoral outcomes. In fact, in many
cases, voter preferences, pro or con, seem to have less to do
with electoral outcomes than what’s referred to as the intensity
of preference. “Intensity” in this context refers to how much
winners or losers desire some preferred outcome. You can
think of this idea in terms of an old adage, “It’s the squeaky
wheel that gets the oil.” Or, put another way, the loudest
voices attract the most attention.
With respect to how organized interests impact public
policy decisions, you’ll want to become familiar with
these assumptions:
n Organized interests manipulate the political process to
compete for benefits, such as tax breaks, subsidies, and
tariff protection, or to avoid costs, such as taxes and
expenses related to regulations.
n The intensity of preference with respect to specific
issues rises as a function of how much is at stake,
monetarily speaking.
n The cost of organizing a group aimed at influencing a
policy position increases with the size of the group to
be organized. In particular, as group sizes increase,
free rider and collective-action problems also increase.
Based on these observations, we can deduce that small
organized interests have an advantage for two reasons.
(1) Given any level of capital benefit, the costs of organizing
are spread out across a much larger segment of the popula-

tion. For example, if domestic sugar interest demands a
price subsidy, the cost of that subsidy will be spread out
across millions of taxpayers with minimal influence on indi-
vidual tax burden. (2) Smaller groups are both easier to
organize and more likely to use resources efficiently.
Coalitions and “Logrolling”
It’s often said that “Politics is the art of the possible.”
Basically, that means ideal legislative outcomes are nice
when they happen, but half a loaf is better than no loaf at
all. That often makes sense because governing very large
groups inevitably involves compromises. And so, legislators
make deals. “I’ll support your appropriations bill if you sup-
port the dam project I want for my congressional district.”
In other words, “I’ll scratch your back if you scratch mine.”
This is an example of logrolling.
Lesson 2 53
Rent Seeking
The process by which political interests use the powers of
government to gain advantages is called rent seeking. That
rather odd term comes from the concept of economic rent,
which basically has to do with securing an advantage in the
market system. Your textbook’s discussion of this topic uses
the example of licensure. Most states require people to attain
licenses to practice law or medicine. However, licensure is
also required for all kinds of other occupations, ranging from
public school teachers and accountants to plumbers and
manicurists. Licensure policies benefit the state in at least
two ways. First, licensing boards gain immediate revenue for
the state from fees required to sit for the licensing exams.

Second, by restricting entry into a profession, the smaller
number of qualified practitioners will lead to increased market
values for services rendered. Furthermore, the public should
benefit from higher levels of proficiency on the part of
licensed professionals.
Strategic Points of Entry
Individuals, groups, and corporate organizations will seek the
best ways to initiate and establish advantageous policies.
The obvious direct route is encouraging legislatures to pass
desired laws that will be endorsed by an executive authority,
such as a president, a governor, a mayor, or a county board
of supervisors. However, there are alternative strategic entry
points for pursuing policy initiatives. They include the
following:
n Referendum. California law permits the electorate at large
to propose and vote directly on policy initiatives. Thus,
laws can be passed by way of direct democracy.
n Executive order. Illustrating the direct opposite of a public
referendum, presidents and governors, informed by
staff and advisors, may issue executive orders. During
World War II, Franklin Roosevelt’s infamous Executive
Order 9066 permitted authorities to round up and
deport Japanese American citizens to detention camps.
n Agency rules. Agencies of the executive branch, such as
the FDIC, are empowered to create and enforce what are
called quasi-judicial policies. Your textbook mentions the
authority of the Environmental Protection Agency (EPA)

to identify and regulate toxic and harmful industrial
wastes and set clean-water standards.
n The courts. In some instances policy initiatives can
be introduced through legal challenges, mainly in the
context of reversing standing policies. A famous example
is the Supreme Court’s 1950 decision in Brown v. Board
of Education of Topeka. The decision reversed earlier
court rulings that supported discriminatory segregation
in public schools and public facilities.
Lesson 2 55
Self-Check 6
1. A/An _______ figures out ways to influence government
policies to alter tax codes and
regulations to his or her advantage.
2. When conditions seem favorable, politicians will cater to the
political center to maximize their
votes in a senatorial election. This idea illustrates the _______
rule.
3. _______ are the pork that members of Congress insert into
bills.
4. According to _______ theorem, when a group is making a
decision among three different
alternatives, no voting system will meet generally accepted
criteria for a fair election.

5. As a student of _______ politics, one would study the ways
political systems vary in
different countries.
6. In the domain of political coalitions, the concept of _______
is illustrated by the adage
“I’ll scratch your back if you scratch mine.”
7. The _______ rule stipulates that only legislation germane to
unemployment insurance
can be introduced in a bill dealing with unemployment
insurance.
ASSIGNMENT 7
on
Allocating Scarce Resources
Allocating resources is a basic challenge in any society. And,
as it turns out, a propensity to trade and barter is a basic
type of interaction in any society. Even simple tribal societies
have characteristically engaged in trade, often using some
kind of an exchange medium like salt or cowrie shells. Money
(based on precious metals) was first used in lieu of live animals
or vegetable produce in Babylonia around 1500 BCE. Paper
money was first used in China during the Song dynasty
(960–1127 CE). So there’s nothing new about either markets
or money.

Features of Markets
Modern societies use fiat currency. It has value only because
international accords, like those that gave birth to the World
Bank in 1944, have accepted national currencies as legal
tender. (The U.S. dollar is, for now, the international reserve
currency.) Be that as it may, markets provide immensely effi-
cient mechanisms for allocating most goods and services. That
doesn’t mean that markets serve all of a society’s needs. In
every modern society, there are contentious policy disputes
over areas of human life that should be excluded from mar-
ket considerations in favor of universal equity. Many feel that
health care, basic education, and adequate shelter should be
considered basic human rights. There’s also a sort of “black
hole” in societies where values are all but universally mone-
tized. For example, most would agree that healthy children
are vital to a society’s viability. Yet child raising isn’t consid-
ered to have intrinsic (monetary) value. That’s why, in some
Western European countries like France, government subsi-
dies afford benefits and extended maternity leave to new
Lesson 2 57
mothers, and even new fathers. In any case, all markets
share a number of key features that can help us understand
both the strengths and potential weaknesses of this
immensely efficient system for allocating resources:
n All markets bring together utility-maximizing consumers
with profit-maximizing firms. The essential quality of this
mutual attraction is voluntary exchange. Day by day,
people make opportunity cost calculations as they decide
whether or not to spend money on X, Y, or Z, which

amounts to maximizing utility. This can be thought of as
getting the most for your money according to how you
derive utility. A principle of this feature is that as prices
decrease (for desired goods or services), demand will
increase, ceteris paribus. This is called the aggregate
demand curve. Study the graphic on page 220.
n Price is the mechanism that brings supply of a good or
service into equilibrium with demand. You’re now familiar
with the supply and demand curve. So that feature of
markets should be no mystery to you. But keep in mind
that price considerations dictate investments on the sup-
ply side. As prices increase, supply is increased ceteris
paribus. If the market value for soybeans is stronger than
the market value for corn, you’ll plant more acreage in
soybeans and less acreage in corn. That sort of phenom-
enon illustrates the aggregate supply curve. Your textbook
supplies you with elaborate examples of the ways supply
and demand meet as an equilibrium price is established.
Study the graphic on page 221.
n Markets are a powerful force for innovation and progress.
People have been inspired to “invent the better mouse-
trap” mostly by visions of wealth and, maybe, 15
minutes of fame. Furthermore, to the extent that
economic competition is healthy, innovations aimed at
improving product quality and consumer services tend
to contribute to the overall quality of daily life.
n Regulating markets, or otherwise limiting the degree to
which prices are allowed to bring supply and demand into
equilibrium, will also dull the mechanism by which markets
“heal” themselves. Basically, interfering with the mechanisms
that produce the supply and demand curve restricts the

flow of voluntary transactions that make markets efficient
allocators of goods and services. Yet, taking the “big picture”
perspective, it has sometimes been necessary to estab-
lish caps on prices and regulate supply. During World
War II, the war effort required restricting consumer demand
by rationing war-vital resources like rubber, gasoline,
and various kinds of food stuffs, like butter and beef. On
the other hand, the war years produced innovations like
Spam, the first computers, advanced radar technology,
and commercially viable jet aircraft. Study the graphic
on page 227 to visualize the way price caps produce
shortages.
n Market outcomes are amoral. Markets are driven by
self-interest, and self-interest may or may not be in
the public interest. The same market mechanisms that
deliver cornflakes, automobiles, and fresh spinach to
consumers also provide prostitution and illegal drugs
in response to the principles of supply and demand.
Meanwhile, one’s socioeconomic place in a capitalist
society, as well as one’s chances of succeeding within
that system, depends on the initial endowments you read
about earlier. Regardless of the efficiencies of markets,
market outcomes are related to the impulses of human
behavior. These unknowns are summed up in the final
characteristic of markets.
n The same features of markets that make our lives steadily
better also make it difficult to stop socially pernicious
behavior for which there are markets, whether it is
selling guns, drugs, sex, or even human organs. This fact
is illustrated by the less than successful “war on drugs.”
Consider this paradox: To the extent that a desired com-

modity is deemed illegal, it will inevitably produce an illegal
market. The market for cocaine has killed thousands and
corrupted any number of governments. America’s experi-
ment with prohibition had the unintended consequence
of financing organized crime families. The mafia families
acquired enough financial clout to seriously impact polit-
ical power equations in cities like Chicago and New York,
not to mention Las Vegas.
Lesson 2 59
Elasticity
For economists, elasticity refers to the degree to which supply
and demand relationships will respond to changes in price.
The higher the price elasticity is, the higher the sensitivity
of demand in response to a price change. As prices increase,
consumers will buy significantly less of a given product or
service. Similarly, the higher the price elasticity is, the higher
the sensitivity of supply in response to a price change.
Producers
will significantly reduce output quantities of a given product or
service. The general formulas for price elasticity are found on
page 231 of your textbook. Percentage values are plugged into
the formulas to calculate demand or supply price elasticity.
So, arithmetically, price elasticity is a percentage ratio. You’ll
want to understand that when supply or demand are all but
impervious to changes in price, we are observing an inelastic
market relationship.
So, what’s the point? If economists or public policy analysts
can calculate price elasticity, they can evaluate or even pre-
dict market stability. Home buyers, investors, economists,
and public policy analysts like to know these things.

Creative Destruction
The Austrian economist, Joseph Schumpeter, invented the
concept of creative destruction. He was referring to the manner
in which innovations in the means of production have the
effect of outdating older skills and technologies. As personal
computers delivered their digital blessings to writers, reporters,
data analysts, and people searching for information, type-
writers became junk and IT became a familiar acronym.
Cassettes were replaced by CDs, which in turn were replaced
by MP3s. As breakthroughs in cybernetics made industrial
plants robot-friendly, assembly line skills of yore became
dated, along with rather large numbers of automobile factory
workers. That is the nature of creative destruction. Market
economies are dynamic. And, as has always been the case
in a market economy, there have been, and will be, winners
and losers.
The voices of the losers tend to be louder than the voices of
the winners, even if there are more winners than losers. The
political process tends to be more responsive to small,
organized
interest groups seeking protection from creative destruction.
As a result, public policy makers are often charged with pro-
viding safety nets for the dispossessed. Often these days, people
on the downside of international free trade agreements, such
as the North American Free Trade Agreement (NAFTA), are
the people calling for compensation.
Productivity

Economists agree that productivity is the single most impor-
tant determiner of our material standard of living. Thus,
since standard of living is a major concern of public policy
analysts, you’ll want to have a good understanding of all the
things that promote or inhibit productivity.
In a market economy, goods and services are derived from
processing raw materials and information. Here’s a simple
example: Back in George Washington’s day, hemp was a mar-
ketable agricultural product, not unlike cotton or tobacco.
Especially in New England, hemp fibers were woven into all
kinds of rope vital to shipbuilding and maritime trade. So
we can trace a rough “chain of trade”: Raw hemp à Hemp
market à Rope fabricators à Maritime market. Money flowed
along this chain as long as there was a demand for hemp and
hemp products. For rope fabricators, productivity (and prof-
its) were dependent on how efficiently various fabricating
techniques could turn inputs (hemp fiber) into finished rope
(outputs) for standing rigging, running rigging, and ratlines.
At every stage of this “chain,” information was the creator of
value. Farmers had to know how to grow hemp. Factors (raw
product wholesalers) required knowledge of the market. Rope
fabricators had to have skills based on knowledge passed down
from generation to generation, as well as a fine understanding
of maritime markets and ship outfitting requirements.
Modern society is based on a highly complex division of labor
wherein trends in almost any industry will impact many other
industries. But our simple example still applies. Production,
whatever its level of efficiency, is based on supply, demand,
and information. And in that context, we identify a basic
Lesson 2 61

principle: A society that can, overall, produce more goods and
services from a stock of available inputs will have a higher
level of consumption and, thereby, a higher standard of living.
As you study this chapter, be sure you come away with an
understanding of the following concepts:
n Diminishing marginal returns. Imagine you own a catering
business. Your market is confined to a travel radius that
takes in the town of Clyburn, Indiana. But, at least you own
the only catering business in town. You run your small
business with the help of three employees, all of whom,
like you, can purchase and prepare food, prepare a venue,
drive the company van, act as servers, and faithfully pro-
mote your services to the local public. Presently, your firm
is moderately profitable, but you would like to increase
your profits. To do that, you decide to lease a second
van to increase the number of customers you can serve,
expand your menu selections, and spend more money
on advertising. But here’s your dilemma. As you add to
the span and scope of your business, your costs also
increase. In fact, you’ll reach a point where increasing
your menu selections will require hiring an additional
employee even as your food-purchasing costs (and hours
on the job) increase significantly. Here’s the lesson you
learn: At a certain point, adding more menu selections
and more customers actually diminishes your return on
investment. Moreover, each unit of added input will
eventually lead to diminishing returns on each unit of
added output. The moral of the story might be this:
Tailor your business to produce an optimal output rela-
tive to inputs. What applies to small business, may also
apply to the crafting of public policy.
n Human capital. As you become more and more savvy
about how to prepare for a catered affair, attract new

customers, and get the best deal on food, each hour of
your labor becomes more productive. This smarter, wiser
approach has what acquired knowledge economists call
human capital.
n Marginal product of labor. This is the incremental output
produced by each added unit of labor, holding all other
inputs constant. Roughly, that translates as workers will
be paid what they’re worth to the firm. More specifically,
where there is a competitive labor market, employees will
earn wages and salaries that reflect their marginal prod-
uct of labor. To see how that works, study the examples
offered in your textbook. As you do that, note that it can
be difficult to ascribe a specific marginal product of labor
to a particular individual when strong levels of produc-
tivity are the product of group efforts. Also, be sure to
follow the discussion of just how a society may become
more productive.
Specialization and Gains from Trade
This section of your assigned chapter is focused on how
job specialization and trade encourage and facilitate wealth
creation. At the outset, one should understand that, in a
modern society, a complex division of labor is crucial to high
levels of productivity. But, in turn, that complex division of
labor can exist only because people can become skilled and
efficient in performing specialized tasks. (The U.S. Department
of Labor’s Dictionary of Occupational Titles includes over
12,000 job descriptions—and that list is likely to be changing
month by month.) Meanwhile, trade, based on specialization
within nations, expands the arena of wealth creation literally

across the planet. However, not all trade is equal. In some
cases, a nation may have an absolute advantage. In other
ways, a nation may benefit from comparative advantage.
Absolute Advantage
Here’s the general idea: If a nation specializes in what it does
best, it will have a trade advantage that invites wealth cre-
ation. For example, to cite the example elaborated on in your
textbook, Saudi Arabia has an absolute advantage in produc-
ing oil. Iowa, on the other hand has an absolute advantage in
the production of corn. Therefore, Iowa has no advantage in
producing oil and Saudi Arabia isn’t likely to see corn grow
on the arid expanses of the Arabian Peninsula. All things
Lesson 2 63
being equal and accounting for transport costs, Saudi Arabia
will derive an absolute benefit from trading oil for corn and
Iowa will have an absolute trade advantage in getting oil in
exchange for corn. But that’s only part of the story. Given
this trade relationship, total production of both goods will be
higher because demand for both goods has been increased.
Comparative Advantage
Trade advantage is said to be relative for a person, firm, or
country when it has the lowest opportunity cost for produc-
ing some good or service. Recall that opportunity cost can be
defined as the value of doing X over the next best option, Y.
The profitability of doing X minus the profitability of doing Y
equals the opportunity cost of doing X. For example, Can-Do,
Inc. produces home security systems. As a second choice, the
company could produce garage door openers with the same

raw materials. So, in this case, the opportunity cost of pro-
ducing home security systems is equivalent to what the firm
might reap from producing garage door openers.
Note: Your textbook illustrates the concept of comparative
advantage by examining comparative opportunity costs in
China and the United States when it comes to focusing on
the production of either televisions or jet engines. It will take
you a while to sort out the numbers, but the bottom line will
be the same: Both China and the United States will benefit
from doing what they do best.
Self-Check 7
Match each statement with the correct term. Not all of the terms
will be used.
______ 1. Prices aren’t allowed to bring supply and
demand into equilibrium and allow the
market to “heal” itself.
______ 2. Innovations in technology make old skills
less valuable.
______ 3. You’ll want to determine the ratio of outputs
to inputs.
______ 4. At Motel Zee, for most of the year, the supply

of motel rooms exactly matches the demand
for rooms.
______ 5. Holding all other factors constant, you calculate the
incremental output produced by one additional unit
of labor.
______ 6. As an example, percentage change in quantity is
divided by the percentage change in price.
a. Creative destruction
b. Market equilibrium
c. Comparative advantage
d. Regulated market
e. Marginal product of labor
f. Elasticity
g. Maximized utility
h. Productivity
Lesson 2 65

ASSIGNMENT 8
on
Market Failure
Human behavior, as is the case with all complex systems, is
inherently unpredictable, like the weather. Human-behavior-
driven markets emerge from immensely complex systems
composed of countless transactions. Therefore, market behavior
may be described in general terms at either the macro or the
micro level, as we might predict either local or global weather
patterns in response to an El Niño event. However, the assump-
tion that humans are rational actors when it comes to weighing
self-interest is just that—an assumption. Like all science-based
assumptions, it must contend with uncertainty. Actual human
behavior may be reflexive (simply reactive) semi-conscious,
driven by emotions, and/or flatly irrational. Therefore, markets
may fail to be efficient allocators of resources for reasons that
can’t be entirely predictable.
On the other hand, economists can identify ways in which
government policies may, at least potentially, play a con-
structive role in preventing market failures or dealing with
the effects of market failures.
Creating and Enforcing Property Rights
The concept of ownership is basic to a market economy. For a
market to be an efficient allocator of resources, private property
rights must be clearly defined and effectively enforced. For
example, people won’t enter into voluntary transactions if
there’s
uncertainty about who owns the goods to be exchanged. At

the whimsical level, think about naïve visitors to New York
City being offered a great deal on the Brooklyn Bridge. At a
more serious level, consider potential investors in a real
estate development project, like a shopping mall. No rational
investor will be willing to put up capital if there’s a shred of
uncertainty about who owns the land where the mall is to be
constructed. In a similar way, no one is likely to invest in a
newly invented product if the inventor doesn’t hold a patent
right to his or her invention. In that context, think about the
importance of intellectual property. Ideas expressed in patents,
novels, screenplays, trademarks, textbooks, brand logos, and
works of art are examples of intellectual property.
Here’s the bottom line: For markets to perform effectively, the
government must identify and enforce private (or corporate)
property rights.
Lowering Transaction Costs
Government policies can offer vital assistance in lowering
transaction costs. Consider the following:
n Currency valuation. Governments in tandem with inde-
pendent agencies, like the Federal Reserve or national
banks, establish and maintain a sustainable value for
currencies like the U.S. dollar, the British pound sterling,
or the euro. Imagine how hard it would be to conduct
business transactions without an assurance that a dollar
is worth a dollar.
n Infrastructure. Public investments in roads, bridges, port

facilities, railways, canals, dams, and so on are absolute
necessities for lowering the cost of doing business.
n Information. Government agencies can establish and
coordinate market standards. Government mandated
labeling of food products (lists of ingredients, expiration
dates, and so on) provides useful information to consumers,
which spares them the effort of gathering information
on their own. If people can trust that a 16-ounce can of
tomatoes actually contains 16 ounces of tomatoes, they’ll
shop with confidence and profit margins will reflect that
confidence. In a similar way, mandated safety and hygienic
standards for restaurants facilitate consumer confidence
and the profit interests of restaurant owners.
n Regulation. In a healthy economy, some voluntary
transactions must be protected by laws. For example,
immense sums are put at risk through the sale or pur-
chase of equities (stocks, bonds, and so on). If these
kinds of transactions weren’t regulated by the Securities
and Exchange Commission (SEC), Bernie Madoff-type
Lesson 2 67
predators might wreak more havoc in financial markets
than the system could bear. Furthermore, consider these
questions: Would you put your money in a bank if you
didn’t know that banks operate under strict regulations,
such as limits on fractional reserves? Would you trust
your bank to protect a $100,000 deposit if that deposit
wasn’t insured by the FDIC? Probably not. That’s why
during the Great Depression, the favored personal “vault”
was under your mattress.

As has been mentioned, most of a bank’s assets are evaluations
of “accounts payable” represented by loans to customers.
However, under regulations managed by central banks, the
ratio of loans put out at interest to money stacked in vaults
or deposited in central banks can’t exceed a set fraction of
the bank’s actual cash reserves. As of 2006, the required
ratio in the United States was 10 percent on transaction
deposits (checking accounts and savings accounts), but
zero on time deposits like CDs. Roughly speaking, that
means that more than 90 percent of a bank’s assets can be
in the form of money lent out at interest. In any case,
fractional reserve policies are meant to assure depositors
that they can withdraw or deposit money without risking
the loss of their funds.
Promoting Competition
In an ideal free market economy, all the players are equally
informed and all have access to capital. As Adam Smith
argued in his famous book, An Inquiry into the Nature and
Causes of the Wealth of Nations (1776), concentrations of
wealth constrain the “invisible hand” that otherwise converts
individual’s pursuits of self-interest into a common good
shared by all.
Needless to say, this isn’t an ideal world. And, in fact,
oligopolies,
like the “big three” automakers (Ford, Chrysler, and General
Motors) effectively and collectively were able to set prices and
effectively implement economies of scale that produced a sort
of “consumer heaven” from about the end of World War II
(1945) to the early 1970s. The rise of the global economy put
an end to all that as a cascade of corporate mergers, many of
them transnational, created de-facto oligopolies in media,
energy, publishing, and so on.

Oligopolies occur when a handful of corporations dominate
an industry. For example, today, among nine top-tier media
companies, five giant corporations maintain dominant control
over media—Time Warner, Disney, Bertelsmann, Viacom, and
Rupert Murdoch’s News Corporation. However, it’s assumed
among public policy analysts that these sorts of oligopolies still
provide consumer benefits through competition for customers.
In any case, public policies aimed at breaking up monopolies
(also called trusts) are crucial to maintaining a vital and viable
free market. A major principle of free markets is competition.
Ideally, competition spurs innovation and market efficiencies
that increase overall benefits afforded to consumers. A com-
pany that controls an entire market, as did Standard Oil
under John D. Rockefeller in the 1920s, has market power
because it can decide unilaterally on production goals aimed
at maximum profit. Monopolies have pricing power because
they can set prices above marginal return precisely because
they have no competitors. This draining of consumer resources
amounts to deadweight loss to the economy as a whole. Keep
in mind that when companies collude to act like a monopolist,
they’re called cartels. A cartel, such as a drug cartel, is a
de facto monopoly. Illegal cartels can generate a lot of dead-
weight loss, simply in terms of funds directed to law
enforcement and international border surveillance.
Note: Be sure to study the graphics on pages 257 and 258 to
understand how an open market (for lemonade) is distorted
by a firm that has a monopoly (on lemonade).
Ameliorating Externalities

At this point, you’re familiar with the concepts of negative
and positive externalities. Presently, the most prominent, and
global, negative externalities have to do with the environmen-
tal destruction produced by industrial effluents on the one
hand and aggregate private behaviors on the other hand. For
example, steel production is directly related to coal produc-
tion. Both of these industries produce negative externalities
resultant from greenhouse gas emissions. Developed nations
are heavily dependent on mostly privately owned motor vehi-
cles. Cars, trucks, and buses produce many metric tons of
carbon dioxide pollution year by year.
Lesson 2 69
Here, you’ll encounter a lengthy exploration of just how, in
economic terms, the marginal cost of production of steel is
related to the social marginal cost of production. After you’ve
studied the textbook’s treatment of this topic, you’ll want to
make this distinction. The private marginal cost of manufac-
turing steel is the cost borne directly by the steel company.
The social marginal cost of steel production is the private
marginal cost, plus the cost borne by the whole society
Note: Studying the graphics on pages 262–266 will help you
visualize the extent of deadweight loss as you compare dis-
tinctive relationships of each kind of cost in terms of supply
and demand curves.
Recall the Coase Theorem. It predicts that parties dealing
with a negative externality can reach a private agreement to
produce an efficient outcome if (1) the relevant property rights
are clearly defined and (2) the transaction costs are low. If it
turns out that these conditions can’t be met, public policy tools
are available for dealing with either a positive or a negative

externality. Respecting negative externalities, governments
can regulate behavior through legal sanctions imposed on
people or firms that violate clean air standards. Respecting
positive externalities, governmental bodies can stipulate
standards that encourage pro-social behavior. For example,
homeowners associations can require homeowners to meet
landscaping standards.
A Pigovian tax (or subsidy) can be used to induce market par-
ticipants to embrace pro-social behavior. Such a tax will be
based on the difference between the private marginal cost of
a behavior and its social marginal cost. For example, taxes
on cigarettes and alcohol (“sin taxes”) can reduce behavior
that generates social costs. Under so-called cap and trade
schemes, a regulatory body can set a cap on the total of, say,
carbon dioxide emissions allowed within an industry or group
of industries. With the cap in place, individual industries can
buy or sell carbon emission permits such that the average of
net carbon emissions stays within the aggregate cap allowance.
In sum, governments can exercise coercive power to encourage
or require outcomes that markets aren’t likely to deliver.
Providing Public Goods
Some goods and services that may enhance social welfare
may not be offered by the market. The reason for that fact
also characterizes two properties of public goods:
n Nonrivalry. Nonrival goods or services may be offered to
any number of additional consumers without further cost.
That is, the cost of the service doesn’t increase over time.
A lighthouse once built can be used over and over again

based on the initial cost of building the lighthouse.
n Nonexclusivity. A lighthouse once built can’t exclude ships’
navigators from using it, regardless of the fact that they
didn’t contribute to its cost. The same principle goes for
most public parks. While some citizens may contribute
to park maintenance through taxation, many others,
such as tourists or visiting relatives can’t reasonably
be kept from enjoying the park.
Note: Study the 2 � 2 matrix on page 272. It will help you
understand how concepts of exclusivity/nonexclusivity–rivalry/
nonrivalry explain the way goods and services are provided
in a modern economy. Note the distinctions between private
goods, club goods, common-pool resources, and public goods.
Providing a Social Safety Net
While you study this and the final section of this chapter, keep
in mind that there’s no objective measure of what economists
call utility. What’s considered a “good” by one individual may
be considered a “bad” by another individual. For example, the
utility of smoking (pleasure and anxiety reduction) may be
seen as outweighing the potential health costs (lung cancer
and circulatory disorders). For landless peasants, the utility
of land reform may seem obvious. Starvation has very low
utility. Yet, for landowners, the utility of land reform, which
might mean breaking up large land holdings in favor of many
smaller land holdings, may be seen as a great evil. In any
case, there are two basic approaches to providing a social
safety net.
Lesson 2 71

Redistribution
Within a given society, policies of redistribution aim at trad-
ing some degree of economic efficiency for increased human
welfare based on equity. In America, the primary mechanism
for redistributing wealth is taxation. Taxes are used to support
public education, Social Security, and social-aid programs like
rent subsidies and the food-stamps program.
Wealth redistribution schemes are typically controversial.
That’s largely because ideas about social equity tend to be
divided along entrenched ideological lines. At one extreme is
the libertarian, “survival of the fittest,” position associated
with laissez-faire (unregulated) capitalism. Rights of private
property are paramount and there should be no restrictions
on how people use their property, save infringement on any-
one else’s property rights. At the other extreme is the radical
socialist ideal that rejects capitalism in favor of the Marxist
aphorism, “From each according to their abilities to each
according to their needs.” Somewhere in the middle are ideo-
logical positions that would base social equity on some kind
of human rights platform. For example, all members of a
society should have access to the basics of life, including
food, shelter, and basic health care.
Paternalism
Basic to the concept of paternalism is the assumption that
people will abuse freedom. They must therefore be protected
from themselves. People will smoke cigarettes, drink too much,
drive recklessly, drop out of high school, experiment with
drugs,
eat too much, and so on. So in effect, paternalistic policies are
meant to improve overall social welfare by preventing people
from engaging in behavior that will harm society as a whole.
Seat belt laws, motorcycle helmet laws, required warnings on

cigarette packages, and regulations meant to reign in predatory
“pay-day” loans are examples of paternalistic policies.
The Limitations of Government
A modern market economy can’t exist without government
agencies. Courts are needed to settle contract and property
disputes. Government funded infrastructures, such as roads,
bridges, and railways, are needed to keep transaction costs
under control. Law enforcement agencies are required to
protect property rights.
On the other hand, policies exercised by government agencies
have their downside. In general, when it comes to efficient use
of resources, government agencies tend to be less efficient than
private firms. More importantly, there are no self-correcting
mechanisms that can cure bad government. For example, the
administrators of government bureaucracies are frequently
tempted to overstress agency needs at budget time, thus
leading to sometimes extraordinary cost inflations. In other
instances, private contractors hired by government agencies,
such as the military, may negotiate “sweetheart” contracts
that would never be accepted between private firms, such as
auto parts suppliers and auto manufacturers.
Ultimately, the public policy challenge is finding the balance
between public welfare and economic efficiency. All govern-
ment isn’t good government, but too little government can
sometimes be a recipe for great harm to the common good.

Lesson 2 73
Self-Check 8
1. To deal with a negative externality, such as second-hand
smoke, you can recommend a/an
_______ tax, which will be equal to both the private and social
marginal costs of smoking.
2. Governments can provide infrastructure, information, and
judicious regulatory measures to
reduce _______ costs.
3. Aiming to improve overall social welfare, _______ policies
attempt to prevent individuals from
participating in behaviors that will harm them sooner or later.
4. Because it controls the whole market, a/an _______ can
exercise market and _______ power,
thus effectively regulating supply and demand.
5. Both nonrivalry and _______ are illustrated by a lighthouse.
6. Inventions, ideas, and works of art are examples of _______
property.

NOTES
75
L
e
s
s
o
n
3
L
e
s
s
o
n
3
Tools for Analysis
ASSIGNMENT 9
on

The introductory discussion for this chapter is a guided tour
of analytical approaches to two public policy related issues.
Is global income inequality growing or decreasing? And, how
large is America’s national debt? Give this discussion some
undivided attention before you turn to the rest of the chapter.
To introduce this chapter, our focus will be on basic measuring,
sampling, and statistical concepts. You may already be fully
familiar with research methods and the principle concepts
of statistical analysis, or you may not be familiar with these
concepts. In either case, if you can increase or refresh your
understanding of concepts like the normal curve, correlation,
measures of dispersion, and probability sampling, you’ll find
it easier to follow explorations into the ways in which they
may be applied to public policy analysis.
Descriptive Statistics
The Mean
Imagine that 20 students have taken a math exam. Under a
standard scoring system, the students’ scores range from 2
to 10. The array of scores from low to high looks like this:
2 3 3 4 4 4 5 5 6 6 6 6 6 7 7 8 9 9 10 10
The mean, or average, score will equal the sum of all the scores
divided by the number of observed scores. In this case that
will be 120 divided by 20 � 6. The mean score is 6. Here the
students as a whole are the unit of observation. The data
points, or observations, are individual scores. Here it is more
formally:
∑X/N = μ

The sum (∑) of X (all units of observations) divided by N [the
number of units (scores, people)] equals the mean. If the set
of scores for the 20 students is all you’re interested in, the
set of 20 scores is your study population. In that case, N �
the number of scores and μ (mu) stands for a population
parameter (measure), which in this case is the mean.
If the scores for these 20 students is a sample drawn at ran-
dom from 200 students, then you would write:
(“x-bar”)
Here, n � the sample population size and x-bar � the sample
mean. The distinction is important because statistical calcu-
lations based on population parameters are a bit different
from calculations based on sample parameters.
The Median
The median is the midpoint of an array of observations
when the observations are arranged in either ascending or
descending order. If the total number of scores is an odd
number, the median is simply the score at the midpoint of
the array. If the total number of observations is an even
number, the average of the two midpoint scores is the
median. In this case, the two scores in the middle are 6
and 6 and 6 + 6/2 � 6. But if the two scores in the middle
were 6 and 7, the median would be 6 + 7/2 � 6.5.
Note: Another measure of central tendency is the mode. In our
sample data array, the most common score is 6. Therefore, 6
is the mode for thus array.

Statisticians divide arrays into percentiles, deciles, and
quartiles to think about where scores or observations fall
in a data array. For our hypothetical data array:
MEDIAN
2 3 3 4 4 4 5 5 6 6 ↓ 6 6 6 7 7 8 9 9 10 10
1st quartile 2nd quartile 3rd quartile 4th quartile
Mean or Median?
Sometimes, the value of a population or sample mean will be
skewed by outliers. Outliers are relatively small numbers of
observations that are distant from the rest of the distribution.
Σ X n x/ =
Lesson 3 77
For example, let’s say that the typical range of incomes for
a population of 100,000 is $4,000 to $6,000. However, 500
people have incomes ranging from $50,000 to $200,000. In
such a case, the mean will be skewed to the right. Therefore,
the mean income will be a poor indicator of the actual income
distribution in the total population. In such cases, breaking
down populations by quintiles, quartiles, and percentiles
ordered around a median will give us a better picture of
income distribution.
Measures of Dispersion
Variance and standard deviation are the key measures of

dispersion. Dispersion is the degree to which observations
(data points) are spread out in a data array. A basic clue to
how statisticians view dispersion is thinking about the nature
of a bell curve, also called a normal curve (Figure 2). As it
turns out, normal curves are very common in nature. If you
measure the precise diameter of, say, all the acorns you can
find around a grove of oak trees, those diameters will be
spread out from smaller to larger following a normal curve.
IQ scores follow the normal curve. The height of male adults
within a population will follow the normal curve. So will the
height of female adults, but the mean values will be different.
–4 –3 –2 –1 0 1 2 3 4
FIGURE 2—A Bell Curve
For any set of N observations X1, X2, X3 . . . Xn, with mean μ,
variance � [(x1 � μ)
2 + (x2 � μ)
2 + (x3 � μ)
2 � . . . (xn � μ)
2]/N
For any set of N observations X1, X2, X3 . . . Xn, with mean μ,
standard deviation � σ (sigma) � the square root of [(x1 � μ)
2
� (x2 � μ)

2 � (x3 � μ)
2 � . . . (xn � μ)
2]/N
Put simply, to find the variance, you subtract the mean from
the value of each observation—expressed as an absolute value—
add all these up, and divide by N. To find the standard
deviation, you simply find the square root of the variance.
Finding the square root of the variance allows you to visual-
ize areas under the curve.
Specifically, about 68 percent of observations for some
population (N) will be found between one standard deviation
above the mean and one standard deviation below the mean.
About 95 percent of observations will be found between two
standard deviations above the mean and two standard devia-
tions below the mean. About 99.7 percent of observations
will be found between three standard deviations above
the mean and three standard deviations below the mean
(Figure 3).
–4 –3 –2 –1 0 1 2 3 4
68%
95%
99.7%
FIGURE 3—The Spread of
Standard Deviations Above
and Below the Mean
Lesson 3 79
Correlation

As opposed to constants like π, variables are any kind of
phenomena with a range of values that can be measured.
Height, weight, IQ, and income are variables that can be
measured fairly precisely. Subjective attitudes can be meas-
ured on some kind of a quantified scale. For example, a
questionnaire item might be posed as follows.
If “Strongly Agree” is coded as 5, where “Agree” = 4,
“Undecided” = 3, “Disagree” = 2, and “Strongly Disagree” = 1,
we can correlate an attitude variable with quantitative values
like income or, for that matter, other kinds of attitudinal values
like approval ratings of Congress or attitudes about abortion.
For example, if we measure height and age variables for a
small population of middle school students, we’ll find that
height and age are positively correlated. That is, in general,
kids get taller as they get older.
Often, when we explore the correlations between variables, we
can assume that one variable precedes the other. For example,
increasing height is dependent on getting older. Thus, we may
want to think of height as a dependent variable (Y) and age
as an independent variable (X). We’ll then postulate that X
precedes Y and make a note that “X à Y.” However, this
notation doesn’t mean that we’ve determined that X causes Y.
In other cases, we may detect negative correlations. For
example,
as protein deficiency increases, cognitive abilities decline. In
other cases, we find that there’s no correlation at all or one
so weak as to be negligible.
You should keep in mind two key ideas as you contemplate the
nature of correlation. Mathematically, a perfect correlation is
+1 and a perfect negative correlation is –1. However, in actual
research, most correlations are suggestive rather than definitive.

Item 3: Government is the problem, not the solution. (Please
respond by checking only one of the following.)
Strongly Agree ____ Agree ____ Undecided ____
Disagree ____Strongly Disagree____
That is, even relatively strong correlations leave much to be
desired. For example, a positive correlation like 0.8 may seem
convincing. But, in fact, 0.80 squared (which returns us to
variance) accounts for only about 64 percent of the variance
in the relationship between X and Y. That leaves 36 percent
of the variance unexplained. Above all, remember that corre-
lation doesn’t establish causality. The only research method
that can reasonably identify a causal relationship, such that
X causes Y, is a carefully designed experiment.
Absolute versus Relative Figures
Data such as dollars spent, petroleum consumed, or bushels of
corn produced can be analyzed as absolute figures on their own.
Analyses based on comparisons look at ratios. For example,
using absolute figures, we can examine the ratio of health
care spending per capita to measures like longevity and
infant mortality.
Much of your textbook exploration here focuses on public
expenditures for education. For example, we can use absolute
figures (money spent) to compare relative investment in educa-
tion in different countries. As it turns out, average expenditures
per student in the United States, in absolute terms, ranks

near the top internationally according to the Organisation
for Economic Co-operation and Development (OECD). On
the other hand, when we look at spending on education
relative to GDP, the United States ranks thirty-seventh
among OECD nations.
In short, what we find depends on where and how we look.
To take another angle on the education issue, given how
much is spent on education in the United States, why is it
that students in less wealthy countries often outperform
American students in areas like mathematics and science?
Percentage Change
Percentage change is a relative statistic because it has meaning
only to something else. For examples, a 25 percent decrease
in price will have meaning only if it applies to some product
or service, like fishing rods or a carpet cleaning service. A
Dow Jones average increase of 6 percent has meaning only
Lesson 3 81
with respect to the end of one trading period and the Dow
average at the opening of the immediately succeeding trading
day. What you will find in this subsection are some tips on
how to appraise percentage changes with respect to things
like stock prices, profit reports, and equity trends.
Expected Value
An expected value (EV) can help us make decisions during
times of uncertainty. The basic idea is conveyed by an exam-
ple involving a single die. What are the odds that a toss of
the die will yield a 1, 2, 3, 4, 5, or 6?

Where the probability in each throw is 1 chance in 6,
EV � 1/6 [1] �
1/6 [2] �
1/6 [3] �
1/6 [4] �
1/6 [5] �
1/6 [6]
The sum of probabilities is then 1/6 �
2/6 �
3/6 �
4/6 �
5/6 �
6/6 � 21. So, EV �
21/6 � 3.5.
That figure, 3.5, is the average outcome if you throw the die
repeatedly. That may not strike you as terribly exciting infor-
mation. Nevertheless, precisely this sort of probability sorting
is used by insurance companies to calculate expected loss.
Using this sort of calculating can help the insurance company
establish actuarially fair premium rates.
Sources of Data
Many things public policy makers might like to know can be
derived only from rigorously controlled experiments on actual
human beings. But that’s simply not an option. Imagine, for
example, researchers trying to identify the effects of child

abuse.
In a controlled experiment, the researchers would have to
purposely abuse children in an experimental group to com-
pare them to the children in the control group. Clearly, this
sort of scenario is out of the question. Therefore, public policy
analysts have to depend on other sources of data. These
include data sets and representative samples.
Data Sets
Cross-sectional data can be compared to a snapshot.
Researchers will examine a set of observations of individuals,
countries, or some other unit of observation at a specific
time. For example, we might be able to compare academic
achievement scores for children with a history of child abuse
and similar children who have no such history based on a
sample taken in the fall of 2011.
Longitudinal data (also called time-series data) can track
observations over a period of time. For example, we might
study trends in poverty rates in rural Kentucky from 1930
to 2011.
Panel data combines cross-sectional and longitudinal data.
For example, in a study of juvenile crime, we might combine
observations and interviews with 100 at-risk working class
teenagers every six months for a period of five years. Panel
studies are excellent for tracking changes in attitudes and
behaviors over time. However, they’re also expensive.
Natural experiments can be conducted unobtrusively. For
example, researchers can observe and report on differences

in behavior and academic achievement among students in a
private school and a public school. Or economists might be able
to observe and compare indexes of productivity in a worker-
owned company and a similar privately owned company.
Representative Samples
It’s not normally feasible to study every person in a town, every
firm in an industry, or every instance of product malfunction.
Instead, researchers can draw probability samples designed
to be representative of the study population. The topic of sta-
tistical sampling will be elaborated in Chapter 10. For now,
the simplest example of a representative sample is a random
sample in which every unit (person, firm, or product) has an
equal chance of being selected. Random number tables,
found in any text on research methods, can be used to draw
a random sample such that the sample n is representative of
the study population N.
Lesson 3 83
Cautions regarding statistical sampling include the following:
n Selection bias. A data selection bias occurs when the
sampling process can generate only a nonrepresentative
sample process. Your textbook gives a good example. If a
polling firm uses a telephone survey to determine typical
attitudes of homeless people, this won’t work—most
homeless people don’t have phones.
n Self-selection bias. A marketing firm selects a random
sample for conducting a marketing survey. All those con-
tacted are offered a discount on stainless steel cookware.
The result is that only people with an interest in cookware

are likely to answer the survey questions.
n Survivorship bias. When survey forms are lost or damaged
before a study is concluded, the study findings will be,
to some extent, distorted or inaccurate.
n Positive-finding bias. Especially in the pharmaceutical
industry, there’s a tendency to publish research findings
that endorse the efficacy of a new drug as opposed to
studies that question the efficacy of the new drug. Sadly,
there’s a lot of pressure on researchers to highlight the
good news and suppress the bad news.
The Time Value of Money
The simplest way to prepare for this section is by getting
familiar with a couple of simple formulas.
Future Value
You’ll want to know the value of a $1,000 deposit five years
from now, assuming the interest rate will be 6 percent. Here’s
the general formula:
FV � PV(1� r )n
Here, FV is future value, PV is the present value of $1,000, r
is the rate of return, and the superscript exponent, n, is the
number of periods, which in this case is five years. The number
1
in the parenthesis is added to the rate of interest, 0.06 percent,
to make the calculation easier. (Otherwise, 0.065 would equal

0.06 � 0.06 � 0.06 � 0.06 � 0.06 � 0.00000007.) Here,
PV � $1,000 (1.06)5
Therefore, 1.065 � (1.06) (1.06) (1.06) (1.06) (1.06) �
1.3382255, or ~1.34.
So, FV � $1.000 � 1.34 � $1,340
In five years, at 6 percent interest, $1,000 will be equal
to $1,340.
Present Value
Public policy analysts sometimes have to go backwards to
convert future value to present value. For example, what
would be the cost of a bond that will pay out $10,000 ten
years from now if the interest rate is 7 percent? We can apply
a bit of algebraic manipulation to FV = PV(1 + r )n, to yield
PV � FV/(1 � r )n
PV, FV, and r mean the same as before, except that now,
n � 10 years in the future.
PV � $10,000/(1.07)10
Thus, as you’ll see in the textbook example, PV �
$10,000/1.97 � $5,100
Lesson 3 85
Self-Check 9

Fill in the blanks with the best word to complete the statement.
1. Using _______ figures, as opposed to absolute figures, to
make comparisons among
U.S. states, we calculate the proportion of money spent on
education relative to the
gross domestic product for each state.
2. In the data array 2, 2, 3, 4, 5, 6, 6, 7, 7, 8, the median is
_______.
3. A/An _______ bias occurs when the process used for
collecting data produces a
nonrepresentative sample.
4. Statistically speaking, the main purpose of figuring out the
variance of a data set is being able
to calculate the _______ of that data set.
5. In the domain of pharmaceutical research, there’s a tendency
for researchers to avoid publish-
ing findings that question the effectiveness of a new drug in
favor of findings that affirm the
effectiveness of that drug. This sort of phenomenon is referred
to as _______ bias.
6. _______ data is also referred to as time-series data.
ASSIGNMENT 10

Policy.
Sample Populations
As you’ll discover in your textbook discussion, remarkably
accurate assumptions about behaviors or attitudes within
a specific population can be determined by relatively small
samples. Indeed, in some cases, a small sample, properly
selected, may be superior to a large sample that may be
more vulnerable to survivor bias.
Simple Random Sampling
Imagine 1,000 differently colored marbles in a big jar.
There are 250 red marbles, 250 white marbles, and 500
yellow marbles. We decide to draw a sample of 20 marbles.
The sample-selecting person is blindfolded. He or she reaches
into the jar, takes out one marble, and hands it to a recording
person, who records the color of the marble and replaces it in
the jar. The jar is then tumbled, to assure that the marbles of
different colors remain well randomized. This procedure is
repeated 20 times. Based on the laws of probability, the chances
are quite good that the sample distribution of red, white, and
yellow marbles will be fairly close to that of the jar. That is, the
odds are good that the sample will be composed of 10 yellow
marbles, 5 red marbles, and 5 white marbles.
However, in actual research, drawing a sample in which each
unit of a study population will have an equal chance of being
selected can be quite difficult. For example, in a study of
college
students’ attitudes toward hazing, we use a random number
table to select a 10 percent sample. However, if we use a
phone survey approach, we may not be able to reach all the
people selected for various reasons. Some potential respon-

dents will be put off by the caller ID and ignore the message.
A few of the selected students may have left the campus for
some reason. Any of these reasons may produce some degree
of sampling bias. Sampling bias occurs when errors or mis-
judgments on the part of researchers, or simple bad luck,
distort the survey findings.
Lesson 3 87
Stratified Sampling
Imagine we want to study comparative attitudes about unions in
a
small town, recognizing that social class may be related to
views
on unions. Our study population is the male adult population of
the town. Through available data on income, level of education,
and
occupation, we determine that 60 percent of the male population
is working class, 30 percent is middle class, and 10 percent is
upper middle class—lawyers, dentists, business owners, and so
forth. To make sure our sample is representative of the town’s
adult population, we’ll survey 60 working class men, 30 middle
class men, and 10 upper middle class men. That’s an example
of stratified sampling.
Cluster Sampling
Let’s say we want to study national attitudes about organic
foods
in upper-crust America. Job one will be locating up-scale zip
codes
in general. Job 2 will be defining a specific cluster of up-scale
gated

neighborhoods. Job 3 will be drawing a random sample from,
say,
400 gated neighborhoods which comprise our defined cluster. If
our budget is meager, we may randomly pick only one
community
from our cluster. That community will be the target for our sur-
vey. If our budget is substantial, we may, by random selection,
choose 5 percent (20) of these communities as survey targets.
Oversampling
If we want to make comparisons among two or three samples,
statistical analysis will require adequate sample sizes. For
example, there will have to be at least 30 study units (n � not
less than 30) in a pair of samples. Otherwise, we can’t apply
statistical techniques like analysis of variance (ANOVA) tests.
For this reason, we may draw a proportionally larger sample
from a smaller subgroup to permit such tests. However, when
we oversample a group, some kind of weighting will be applied
when statistical tests are performed to avoid biased findings.
For example, if we’re comparing attitudes about Hollywood for
a large group of “below-the-time” people, such as carpenters,
electricians, and caterers who work on movie productions,
versus
highly paid talent (a much smaller group), analysts might
multiply
the aggregate talent attitude measures by a coefficient (like 0.5)
to reduce their impact on the overall findings.
Drawing Statistical Inferences
Not surprisingly, public policy analysts must often draw

inferences from sample data. In statistical terms, that means
using the tools of statistical inference. Here, we’ll provide
some basic ideas about these tools.
The Sample Mean and the Central Limit Theorem
Mathematically, you calculate the sample mean in the same
way you do a population mean, except that you use a formula
that reads
A sample mean (x-bar) is the sum of the sample observations
(x) divided by the number of observations (n). To calculate
the sample standard deviation(s), you’ll first find the sample
variance by subtracting the sample mean from the value of
each sample observation—expressed as an absolute value—
add all these up, and divide by n – 1. To find the sample
standard deviation, you simply find the square root of the
sample variance.
Note: You’ll use n – 1 rather than n for mathematical reasons
having to do with the properties of a sample mean relative to
a population mean.
The central limit theorem states that the sample means will be
distributed around the population mean as a normal curve.
Of course, you won’t know the value of a population mean
or a population standard deviation from looking at a sample.
Let’s say you draw 15 samples of n = 30 from a statistical
population, N = 1000. It’s unlikely that any of your sample
means will be the same. But the array of sample means will
be distributed as a normal curve, regardless of the shape of
the actual underlying population. That means that as the
number of sample means is increased, their central tendency
will gather around the actual population mean.
Note: Consult the graphic on page 339 of your textbook.

Σ x n x( ) =/
Lesson 3 89
The Standard Error
The standard error measures the dispersion of all the sample
means you’ve gathered. Specifically, for any population with
a mean, μ, and a standard deviation, σ, the standard error is
expressed as σ divided by the square root of n, which is the
size of the sample population.
Given that we still don’t know the actual standard deviation of
the study population, what do we do? We improvise. That is,
we calculate the standard error of the distribution of sample
means by substituting s for σ. Thus, we calculate the stan-
dard error of our distribution of sample means by dividing s
by the square root of n, which, of course, is the sample size.
We can now make inferences based on the properties of the
normal curve. Namely, about 68 percent of our sample means
will be found between one standard error above the population
mean and one standard error below the population mean.
About 95 percent of our sample means will be found between
two standard errors above the population mean and two
standard errors below the population mean. About 99.7
percent of our sample means will be found between three
standard errors above the population mean and three stan-
dard errors below the population mean. See the graphic on
page 342 of your textbook.
Confidence Intervals

A confidence interval is simply the likelihood that the actual
population mean will be found within a defined span under
the normal curve. Let’s say, for example, that we’re sampling
a study population to determine the average income of flight
attendants for Southwest Airlines. It turns out that our calcu-
lated standard error is $100, given an estimated population
mean of $10,000. Choosing the 95 percent confidence level,
we can state (with 95 percent confidence) that under the cen-
tral limit theorem, the typical annual salary for Southwest
Airlines flight attendants will be 1.96 standard errors above
and below the mean. Therefore, the typical salary for the
flight attendants will be between $9,796 and $10,196. When
you think about confidence levels, be aware that the likelihood
that we’ll be off the mark increases as we increase our confi-
dence level.
Polling
Polling analyzes representative samples to calculate confi-
dence intervals regarding the proportions of people who will,
for example, vote for Candidate D versus Candidate R among
likely voters. The emphasis would be on sampling likely voters
because, otherwise, the poll won’t be accurate. In studying your
elaborated textbook treatment of this topic, note two things.
The formula for calculating the standard error for a sample
proportion is different from the one we use for figuring out
the standard error for quantitative samples. In the formula,
the letter p with a symbol over it, called “p-hat,” stands for
“proportion.”
standard error of a sample proportion �

Most pollsters aim for a 95 percent confidence level, which in
the context of polling is called the margin of error. The topic
of error margins is actually quite complex. For example, when
you hear a newscaster report a poll as accurate within plus
or minus 3 percentage points, that assertion may be derived
and interpreted in different ways. Be that as it may, you can
get a feel for the polling concept of margin of error with
examples
laid out in your textbook. The math is simple, but patience is
called for. Take some time to follow the logic. Meanwhile,
here’s
a simplified example to consider.
Imagine that you’ve gathered 100 representative samples of
likely voters in a Nebraska Senate race. If you apply the 95
percent confidence level, you can expect five of these samples
to fall outside two standard errors to either side of the mean of
your 100 means. Thus, the margin of error can be viewed as
accurate within plus or minus 5 percentage points. Meanwhile,
one other idea to keep in mind is that error margins are likely
to be reduced as sample sizes get larger.
Decision Trees
Decision trees are logic maps or algorithms used for making
decisions in the presence of uncertainty, if you can estimate
the probabilities of contingencies or “payoffs” resulting from
possible courses of action. If, for example, you’re considering
installing a solar panel on your home, you would want to
p(1 – p)/n^ ^
Lesson 3 91

consider all relevant variables, such as initial cost, unit
efficiency
(savings on power bills), maintenance costs, averages of
sunny versus cloudy days where you live, aesthetic factors,
and so on. In any kind of decision tree, your objective is
weighing the probabilities of positive returns versus costly
contingencies.
Note: The best way to get a handle on this concept is through
studying your textbook’s discussion and paying special atten-
tion to the illustration of a decision tree on page 350.
Hypothesis Testing
Most statistical inference is based on hypothesis testing. A
hypothesis may be thought of as an educated guess that can
be confirmed or rejected based on statistical analysis. Here’s a
simplified example: You reasonably suspect, based on available
data, that juvenile delinquency (JD) rates within neighborhoods
are correlated to average differences in neighborhood income
levels. Let’s call juvenile delinquency rates your dependent (Y)
variable since you postulate that average income level (X)
encourages juvenile delinquency and, therefore, precedes
delinquent behavior. You’ll pose two hypotheses.
H0: JD rates are unrelated to income levels. (X and Y aren’t
correlated.) That’s your null hypothesis.
Your alternative hypothesis, which you hope to prove by
rejecting the null hypothesis, would be
Ha: JD rates and income levels are negatively correlated. (As
income increases, JD rates decline.)
You gather representative samples from 20 neighborhoods

that roughly represent a spectrum from wealthy neighbor-
hoods to lower income neighborhoods. You pair income levels
and JD rates for each neighborhood to create a scatter plot
with 20 X-Y junction points. After deciding that the correla-
tion coefficient should be statistically significant at the 0.05
level (five chances in 100 that the finding is due to random
fluctuation) to reject the null hypothesis, you calculate the
correlation coefficient.
Note: Significance levels for correlation coefficients are usu-
ally found in an appendix at the back of any basic statistics
textbook.
Regarding our example, you might be surprised to learn
that several classic studies found that delinquency rates
for lower class and middle class teenagers were roughly
the same. However, lower class kids are far more likely to
be prosecuted and incarcerated. Now there’s an interesting
challenge for people who formulate public policy.
Type I and Type II Errors
When testing a hypothesis, two basic kinds of errors can occur.
Put simply, a type I error occurs when we reject a hypothesis
that is, in fact, true. For example, in our study of delinquency
rates and income levels, if we reject the alternative hypothe-
sis and embrace the null hypothesis, we have made a type I
error if, in fact, delinquency rates and average neighborhood
income are inversely related. By contrast, a type II error occurs
when we accept a hypothesis as true when, in fact, it is false.
For example, if we reject the hypothesis that delinquency rates
and average neighborhood income are inversely related, such

that delinquency rates decline as neighborhood income
increases,
we’ve fallen prey to a type II error.
Let’s consider another example. A jury will embrace a type I
error if it wrongly rejects the null hypothesis that John is not
guilty. As a result, the innocent John will be treated as a
criminal. If the jury rejects the not-guilty null hypothesis
when John is actually guilty, the jury has accepted a false
negative and John will skip town, a free man, due to a
type II error.
An important insight to look for under this topic is that
public policy makers must tread a fine line between the
risk of making either a type I or a type II error. Strive to
understand how and why this is the case.
Lesson 3 93
Self-Check 10
1. If your selected confidence level is 95 percent and n = 100,
you know that _______ of your
sample means will fall outside the range of two standard
deviations to either side of the mean
of our sample means.
2. In trying to decide on the risks versus the advantages of
making an investment in a new
milling machine, you can estimate the probable outcome of
alternative courses of action using
a/an _______.

3. If you’ve gathered a/an _______ sample, each person in your
study population has an equal
chance of being selected.
4. A/An _______ error occurs when a null hypothesis is rejected
even though it’s true.
5. Following the _______ limit theorem, the standard error is a
measure of the dispersion of any
number of _______ means.
6. If, in the process of hypothesis testing, you reject the
_______ hypothesis, you’ve accepted
the null hypothesis.
ASSIGNMENT 11
Policy.
A discussion of the Framingham Heart Study leads this
chapter. That study has persisted for the last 50 years and
has contributed immensely to understandings of health
issues, like the relationship of coronary disease to exercise
and diet. You’ll want to study it because it illustrates the
considerable potency of regression analysis.
Linear Regression

In general, regression analysis aims at explaining how one or
more independent variables (X1, X2, X3) helps explain some
dependent variable (Y), like heart disease, attention deficit
disorder, unemployment, or poverty. For example, if poverty is
our dependent variable, we might explore the relationship of
poverty to factors like, age, gender, and socioeconomic status
(SES).
Establishing a Linear Relationship
From your earlier exploration of correlation, you’ll have some
insight into the concept of bivariate regression analysis.
Consider
any scatter plot, such as the one on page 369.
Now visualize a straight sloping line that best describes the
obvious positive correlation of height and weight as children
mature over time. Of course, your visualization would be
subjective. Someone else might come up with a different line.
The solution is offered by applying an ordinary least squares
(OLS) equation to the data you’ve gathered. You won’t have
to figure that out by hand. Statistical procedures, like those
available through SPSS (Statistical Package for the Social
Sciences) will calculate a line that best fits your height-weight
scatter plot points. It will do that by minimizing the distance
between each observation point and the fitted line. The verti-
cal distance between a given observation and the best-fitted
regression line is called an error term.
Lesson 3 95
The slope of the line will be represented by a number, like 3.5
or 2.2. The bigger that number—called the regression
coefficient—

the steeper the pitch of the regression line. Regression
coefficients
show us the best relationship between a dependent and an
independent variable. They also allow us to predict the height-
weight relationship of added observations, such as those for
any child not included in the original sample.
Note: Although your study guide doesn’t cover many of the
mathematical equations in this chapter, you should still study
the quantitative examples offered in your textbook. You should
try to grasp an overall feel for how these analytical tools
are used.
Multiple Regression Analysis
Multiple regression analysis allows us to quantify the rela-
tionships between a dependent variable and more than one
independent variable. The general multiple regression equation
looks like this:
Y � a + b1x1+ b2x2 + b3x3 . . . + bixi + e
Here, Y is the dependent variable; a is an intercept; x1, x2, x3
. . . xi are independent variables and e is the error term. The
term b stands for the regression coefficient. For each inde-
pendent variable, b (times a value x) allows us to quantify the
relationship between the independent and dependent variable,
ceteris paribus.
At this point take note of two ideas. First, we can easily
graph a bivariate regression relationship because it can be
represented two-dimensionally. However, multivariate regres-
sion relationships are extremely difficult to visualize because
they are multidimensional. Second, for reasons that will be
considered in your textbook, in multivariate analyses we dis-
tinguish between variables with a range of values and binary

variables, referred to as dummy variables. Sex is binary (if we
exclude cross-gender people). Smoking may be classified as a
dummy variable if we give it a binary definition, such as
“smoker” or “nonsmoker.”
Interpreting Regression Results
Sign, Size, and Significance
The sign of a variable value—plus or minus—is important.
If income has a positive (+) relationship (positive correlation)
with our dependent variable, longevity, we need to know that.
A negative value means that a variable (like income) is nega-
tively correlated to the dependent variable such that as
income increases, longevity decreases. (That would seem an
unlikely outcome. But, in fact actual global longevity studies
have identified the longest-lived people in remote, relatively
impoverished places like Bolivia and Kazakhstan.)
The size of the relationship matters. You would probably agree
that a higher correlation is better than a modest correlation if
we’re looking for likely causal factors. In multiple regression
analysis, the quantitative product of b times x is more impor-
tant if b (the coefficient of the independent variable) is larger
rather than smaller. (By analogy, a steeper regression line is
more relevant to the relationship between a dependent variable
and a given independent variable than is a shallow regression
line.)
For public policy purposes, the statistical significance of a
relationship is important. In that context, a researcher will
draw on the procedures required in hypothesis testing. An

alpha (significance level) where p = 0.05 (5 chances in 100) is
better than p = 0.5 (5 chances in 10). If public policy is to be
effective, resources should be directed toward the factors with
the most weight.
Goodness of Fit
Beyond sign, size, and significance, there is a measure of the
overall degree to which a regression analysis helps explain an
outcome. Specifically, the R2 measure of goodness of fit tells
us how much variation in the dependent variable is explained
by the regression equation relative to the total variation in the
dependent variable. For example, R2 could tell how much of the
variation in worker productivity at Company Z is accounted
Lesson 3 97
for by our multiple regression analysis. Meanwhile, the
statistical significance of R2 can be calculated by way of the
F-test, as explained on page 385 of your textbook. Notice
that the F-test is based on the principles of hypothesis
testing.
Note: The textbook discussion of linear transformations, cover-
ing pages 385–389, is recommended but optional. Questions
about this material won’t appear on your lesson exams or in
the self-check for this chapter.
Common Regression Errors
n Confusing correlation with causation. Regression analysis
is based on correlations. As previously noted, only
properly designed experiments can demonstrate or
prove causation.

n Endogeneity (reverse causality). In the real world, most
variables don’t stand all alone in their own cell-like space;
they interact. For example education can lead to certain
occupations or working in an occupation may lead to
expanded education. So, sometimes a postulated inde-
pendent variable will turn out to be best considered as
a dependent variable.
n Omitted variable bias. Sometimes an independent vari-
able stands as a sort of proxy for a variable that’s been
omitted from a multiple regression analysis. For example,
reported frequent headaches and above-normal blood
pressure may be “stand-ins” for a variable omitted from
consideration—job stress.
n Multilinearity. The ideal in multiple regression analysis
is isolating the effect of different independent variables.
However that objective may be compromised if, say, two
of the independent variables are so closely correlated that
it’s all but impossible to identify their specific impact on
the dependent variable.
n Data mining. One can load up a multiple regression
equation with all kinds of variables that may have little
or nothing to do with likely factors producing variation in
the dependent variable. Further, as more and more “junk”
variables are loaded into the equation, the chances of
spurious relationships increase. A spurious relationship
amounts to a false correlation. For example, ice cream
consumption can be positively correlated with levels of
street crime. But, in fact the causal factor related to both

crime rates and ice cream consumption is temperature.
Both crime rates and ice-cream eating increase during
hot weather.
Lesson 3 99
Self-Check 11
1. Among errors that can occur in regression analysis,
endogeneity refers to reverse _______,
such that Y causes X where Y is the dependent variable.
2. In interpreting regression analysis findings, we’ll consider
the _______ ascribed to a given
independent variable to determine if it is either negatively or
positively correlated to the
dependent variable.
3. In the general equation, y = a + bx, the term _______ is the
point where a regression line
intercepts the y axis.
4. It’s true that the _______ error of a regression coefficient
will decrease as the sample size
increases, all things being equal.
5. _______ squares is the most common method for fitting a
regression line to the set of
observations in a bivariate regression analysis.
6. In a bivariate regression analysis, the vertical line between a
data point representing any

height-weight junction (data point) and the regression line is
called the _______ because
it is a measure of the portion of the dependent variable that
remains unexplained by the
regression equation.
ASSIGNMENT 12
Policy.
Basic Tools of Cost-Benefit Analysis
In purely economic terms, comparing the utility of two or
more public projects, such as building roads versus laying
railways versus building a dam, it makes sense that the cost-
to-benefits ratio should be positive. But if that’s the case,
how will we go about determining those ratios? There are no
easy answers to that question. But one can begin by becoming
familiar with the tools of cost-benefit analysis.
Establishing Common Units (Monetizing)
A fundamental tool is called monetizing. It’s fundamental
because a basic challenge of cost-benefit analysis is assigning
a common value denominator, usually dollars, to all kinds of
conceivable costs and benefits. In that way we can establish
and assign costs and benefits to such diverse things as human
safety, reduced transaction costs, environmental externalities,
and improved quality of life. That can seem to propose an

insurmountable menu of difficulties. Given inevitable con-
tentions over basic economic and human values, how, for
example, does one assign dollar values to quality of life, or for
that matter, human safety? Well, at least we can begin with a
basic premise: The goal of public policy is making the best
use of society’s finite resources.
Present Discounted Value (Net Present Value)
To monetize is to assign a common denominator of value. But
even before we attempt that task, we have another problem
we can call the time value of money. In fact, we explored that
issue in Chapter 9. But we can recall the upshot of that issue
quite simply: The value of money in hand is greater than the
value of money to be received at some future time. Reasons
for this include the cost of delayed satisfaction, investment
opportunities that might vanish while we wait for “our ship
to come in,” and, of course, inflation.
Lesson 3 101
Note: You can benefit from reconsidering the formula for net
present value, N/(1 + r)t where N is the nominal or future
value of costs or benefits, r is the discount rate, and t is the
exponent given in numbers of periods (years usually) that will
pass before the cost or payment will come due.
Evaluating Uncertainty
All projected costs and benefits may be subject to uncertainty.
The value of a public investment in roads may falter if light
rail investment increases. The cost of maintaining public
water treatment plants may increase as result of fertilizer
runoff adjacent to factory farms. Changes in technology may

reduce the value of parking meters if more people begin to
telecommute. In light of these kinds of contingencies, public
policy planners must calculate the probabilities of possible
things that can go wrong, or right.
Note: Study your textbook illustration of how cost uncertainty
may be addressed. If two cost estimates differ significantly,
they can be reconciled by mathematically evaluating the
weighted average of two possible cost outcomes.
Opportunity Cost versus Cash-Flow Accounting
For accountants, the cost of your time on the golf course
amounts to cart costs and greens fees. The cost of getting
your license renewed is the license renewal fee. In short what
matters is cash flow. Cost-benefit analysts take a different
point of view. Whatever you have to give up to spend time on
the golf course will be the opportunity cost of that choice. The
time you lose waiting in line at the Department of Motor
Vehicles
is the opportunity cost of getting your license renewed. In
cost-benefit analysis, cash flow isn’t irrelevant, but the
opportunity costs of inputs and outputs are always taken
into consideration. For accountants, money is money. For a
cost-benefit analyst, time is money.
Economic and Noneconomic
Values (Use and Nonuse Values)
In the context of public policy formulation, cost-benefit
analysts
have to be skillful jugglers. Imagine four balls they have to

keep
in the air. The Economic Cost-or-Benefit ball is emblazoned
with a dollar sign. It’s all about dollars and good sense. The
Noneconomic Cost-or-Benefit ball is like a translucent sphere.
Now you see it, now you don’t. It’s all about intangible gains
or losses related to things like self-esteem, motivation, and
one’s views as to what’s important in life. The ball stamped
Use Value has to do with how a resource can be used directly
for recreational or commercial purposes. The ball stamped
Nonuse Value is similar in appearance to the Noneconomic
Cost-or-Benefit ball—intangible. Here we find values that may
attach to a resource one may not actually use. The preserva-
tion of historical antiquities may be more a value than a used
resource. The value of preserving National Parks for future
generations is one of the finest “nonuse” ideas Americans
ever came up with.
Even though the distinctions among these four concepts may
be hard to determine, public policy formulators must strive to
weigh and balance each ball, always remembering that the
goal of public policy is making the best use of a society’s
finite resources.
Note: The concluding “For Discussion” feature at the end of
this chapter will help you better understand the cost benefit
juggling act.
Making a Decision Using
Cost-Benefit Analysis
Consider these principles:
n If, for any proposed project, the total expected social bene-
fits outweigh the total expected social costs, it should be
adopted because it will expand that society’s net resources.

n When weighing project options, policy makers should, as
a rule, choose the one with the largest net present value
(NPV).
Lesson 3 103
n Policy makers must decide which parties, such as the
people in Watts or the residents of Cornwall County, will
have standing. The parties who have standing are the
parties we’ll consider in our analysis of a proposed project.
n In some cases the project with the greatest NPV won’t be
favored by those who have standing. For example, risk-
averse policy makers may adopt an alternative option with
a lower NPV if outcome probabilities are more certain.
n In some cases policy makers will turn thumbs down on a
project with a large NPV because of distributional effects.
For example, a sales tax that will provide needed revenues
in a locality may be rejected because the poor and needy
will be cruelly burdened by such a tax. (In such cases we
say the tax policy is regressive.)
Valuing Inputs and Outputs
Economists and cost-benefit analysts share a common idea
about the value of anything at all. It’s summed up in the
willingness to pay (WTP) principle. That principle asserts that
the most accurate measure of the value of anything from a
taco to a toaster to a luxury car is how much any person is
willing to give up to get it. That certainly makes sense in the
marketplace. But the WTP principle can also be applied to
things that aren’t goods or services with price tags. In this
section, keep an eye out for the following ideas.

Tradable Goods
When a project will rely on resources that can be bought and
sold in the marketplace, the best measure of social cost is
the cost of those inputs. However, market prices aren’t a
good guide to social cost when there are distorting factors.
These may include negative externalities, monopolies, tax
rates, or the impacts of regulation.
In some cases, opportunity costs will be greater than account-
ing costs. The example used in your textbook is conscription
for military service. An extreme historical case in that context
is offered by the impacts of World War I. The generals on all
sides used Napoleonic tactics against barbed wire, machine
guns, and massive artillery barrages. Casualties on all sides
were appalling. Among the belligerents, particularly in
Europe, entire generations of young men were bled dry. The
“opportunity costs” reckoned in terms of the extermination
of future scientists, skilled artisans, actors, artists, and
entrepreneurs were incalculable.
Shadow Pricing
The process by which an opportunity cost is assigned to
resources with no obvious market value is called shadow
pricing. As a visual metaphor, imagine the unknown costs of
a present-day controversial practice—“teaching to the test.”
Imagine a shadow cast by that educational philosophy. What
might be the opportunity costs associated with graduates’
diminished capacities for critical thinking or creatively

“thinking
out of the box” to come up with socially beneficial innovations?
Where market feedback is absent, a process called contingent
valuation can help policy analysts estimate price values asso-
ciated with a proposed program or project. Specifically,
carefully
crafted surveys, guided by probability sampling, can gather
vital information from publics or parties that would be impacted
by a policy. However, surveys of this sort must provide the
following:
n An adequately detailed scenario or description of a pro-
posed program, such as measures to protect endangered
species or revitalize a town’s commercial center
n Items that require survey respondents to indicate
willingness to pay
n Relevant demographic data, such as age, gender, race,
income, and political philosophy
The latter information will permit analysts to use, for exam-
ple, regression analysis to isolate and correlate demographic
variables to attitude variables regarding a proposed project.
Lesson 3 105
Value of a Statistical Life
In the context of conducting a cost-benefit analysis, when
human lives are assessed as a cost or benefit, policy makers
must assign a monetary value to each life. That sum is called
the value of a statistical life (VSL). Contingent valuation is one

approach to identifying a VSL. But there are alternatives.
n Revealed preferences. Assume we can identify people’s
willingness to pay for goods or services that reduce the
probability of infirmity or death. We can use that data to
make inferences based on willingness to pay for things
like health insurance, home security systems, or even
pricey organic food over processed foods. In that way we
can infer people’s monetary valuations of measures that
will protect a statistical life.
n Hedonic market analysis. Assume that workers who
work at risky jobs will expect a premium (a bit like
combat pay) added to their base wages. If this is the
case, we can use regression analysis to estimate the
risk premium, or compensating differential that must
be offered to induce people to work in risky occupations,
like high steel construction. As noted in your textbook,
these kinds of analyses may wobble a bit, given that
some risky occupations, like coal mining, rely on low
SES workers who have limited bargaining power.
Choosing a Social Discount Rate
Native American wisdom proposed that an appropriate use of
resources will consider the people’s future needs unto the
seventh generation. Considering present-day global challenges
like climate change, species die-off, and soil depletion, that
formula seems like an unrealistic ideal. Nevertheless, how to
go about allocating resources for the immediate future, much
less future generations, presents thorny problems for cost-bene-
fit analysts. The essence of that thorny issue is how to go
about establishing a social discount rate. The social discount
rate is used to discount costs and benefits to the present
time, just as we would determine an NPV where we’re con-
cerned only with a strictly monetary cost-benefit analysis.

However, a social discount rate must also take public senti-
ment into account. Indeed, as you study this topic, pay
attention to how different social philosophies may influence
ideas about how best to establish a socially acceptable and
monetarily viable discount rate.
Note: Study the table on page 430 to think about the value of
$100 at different points in the future using different discount
rates. Also, to think about the utter lack of consensus as to
appropriate social discount rates in regard to climate change,
study the discussion and the graphic analysis on page 433.
Sensitivity Analysis
As you round out your study of this chapter, make sure you
understand two concepts.
n Sensitivity analysis. Cost benefit analysts will look at
the results of applying a range of discount rates to a
given problem—like maintaining clean water standards.
A program design in which the social benefits outweigh
the social costs regardless of which discount option is
applied would be ideal. The default option is choosing a
social discount rate that appears to be better than the
alternatives. The term sensitivity used in the context
sensitivity analysis refers to the extent to which a program
appears viable or not, given some range of discount rate
alternatives.
n Cost-effectiveness analysis. Imagine that the goal of
reducing CO2 emissions is taken as paramount. There’s

no real alternative. Either CO2 emissions are curbed or
the biosphere becomes terminally hostile to human life
as we know it. No dollar amount can weigh the cost of
nonaction. In such a case, a range of policy solutions is
evaluated. But now the goal is identifying a program that
will most immediately and effectively reduce CO2 emission,
regardless of the cost.
Lesson 3 107
Self-Check 12
1. A cost-benefit analysis always values inputs and outputs in
terms of their
associated _______ costs.
2. We can employ a/an _______ market analysis to determine
the compensating differential
in pay required to induce workers to take jobs that involve an
increased likelihood of injury
or death.
3. _______ to pay offers the simplest measure of what anything
is worth.
4. To conduct a/an _______ analysis, we’ll look at several
different estimates of future energy
costs to complete our cost-benefit analysis of a proposal to
install solar panels on the roof of
our office building.
5. In respect to the concept of _______ goods, the price of

anything that can be bought or sold
in a competitive market is the best measure of the social costs
of those inputs.
6. In the process called _______ pricing, we assign an
opportunity cost to resources, like clean
air, that have no obvious market price.
ASSIGNMENT 13
Policy.
You should come away from this chapter with a good overall
sense of how one goes about evaluating the outcome of a
policy-
driven program, intervention, or project. In that context, you’ll
find that a lot of the chapter material elaborates concepts that
have already been introduced. Consider that a plus. Learning
is enhanced when we can explore ideas of any sort from a
different angle.
The Treatment Effect
The impact of a policy or program is called the treatment effect.
If, for example, a fitness program is mandated for public
schools, we would assess the program by assessing the
extent to which indicators of fitness have improved in the
wake of the program. If, overall, kids can do more sit-ups,
run faster, and show improved body-fat to weight ratios, we

could justly claim that the program is doing what it was
intended to do.
In this example, we have a clear comparison group if we have
data on kids’ physical fitness indicators prior to the initiation
of the new fitness program. On the other hand, if we have lit-
tle or no data on kids’ fitness prior to the implementation of
our program, we’ll have a lot more trouble assessing the
value of our fitness program.
Confounding Factors
This section elaborates on the nature of factors that can dis-
tort or totally invalidate a research design. Consider the
following:
n Selection effect (selection bias)
n Distinguishing causation from correlation (or just
coincidence): Be sure you can differentiate the possibilities:
A causes B. A and B are correlated, but A doesn’t neces-
sarily cause B. B may cause A (reverse causality). Some
Lesson 3 109
omitted variable, C, may be responsible for both A and
B. A and B are related as a result of coincidence; B just
happens to show up when you observe A.
n Dropout effect (survivorship bias)
n Evaluator or participant bias
Evaluation Strategies

Randomized Experiment
As your textbook points out, the properly conducted experi-
ment is the “gold standard” in all kinds of research, including
research to evaluate the outcome of public policy programs,
projects, or interventions. If you have a good understanding
of an ideal experimental design, you’ll have a handy tool for
understanding the shortcomings of alternative research
designs. Imagine this ideal research scenario.
An experiment is conducted to assess the treatment effect
of a drug, “Nova 22,” on hospitalized patients suffering from
emphysema. At a large general hospital it proves feasible to
use probability sampling to select two matched samples of
n = 20. All subjects are male. All subjects are between the
ages of 45 and 55. All subjects developed their symptoms
while engaged in manufacturing activities that exposed work-
ers to asbestos particulate.
Sample 1 is designated the control group. Sample 2 is
designated as the experimental group. Once a day, nurses
administer a plain white tablet to each person in both the
control and experimental group. All subjects are informed
that the drug is intended to improve respiratory functioning
(make for easier breathing). However, while the subjects in
the control group are actually taking a placebo (a sugar pill),
only the subjects in the experimental group are ingesting
doses of Nova 22. The medication regimen lasts precisely 20
days. Respiratory function (RF) for each patient is measured
immediately following day 20 and at the same time of day.

Your hypotheses are as follows:
HO: The difference in RF between the Control and the
Experimental group � 0
HA: The difference in RF between the Control and the
Experimental group = 0
To eliminate possible bias, the study is double blind. That
is, neither the nurses administering the medication nor the
patients themselves know which patients are in the control
or experimental groups.
Assume that RF can be precisely measured on a scale from 1
to 10, such that any reading greater than 5 is considered
meaningful (significant) improvement in ease of breathing.
Finding: The average RF reading for the control group � 3.8
with a range of 2.5 to 6*. The average RF reading for the
experimental group turns out to be 8.5 with a range of 7.8 to 9.
The null hypothesis is rejected.
* The modest improvements in RF among patients in the
control group might be said to result from what’s called a
placebo effect. All the patients understood what the drug
was intended to do. Expectations and assumptions can and
often do produce physiological effects.
Conclusion: Nova 22 produces a significant and precisely
measurable increase in respiratory functioning.
Now let’s return to the real world. Let’s say that the problem
you want to address is domestic abuse. Your public policy
concern is reducing rates of domestic abuse. To address that
issue you’ll review all the available literature and data on
domestic abuse in search of hypotheses that can be tested.

You know that previous studies have shown that socioeco-
nomic status (SES), a history of child abuse, and financial
insecurity are related in the incidence (frequency) of domestic
abuse. But can you imagine carrying out any kind of controlled,
double blind experiment to establish precisely measurable
causes of an increased rate in domestic abuse? Maybe you
can (without spending a very large sum of money on the
research). If so, you may become a legend in the field of
public policy analysis.
Lesson 3 111
Natural Experiments
An example of a natural experiment was offered to you earlier
and another is provided in your textbook. The basic idea is
that occasions may arise when a study population can be
naturally divided into a control and experimental group. It
might turn out, for example, that there are two middle
schools in Claxon County, one at either end of the county,
such that kids from School A seldom if ever interact with stu-
dents from School B. Both schools serve populations with
similar demographics. But for no particular reason, one class
of seventh graders at School A viewed a documentary film on
the Civil Rights Movement. The film wasn’t used at School B.
If you were interested in the attitudinal impact of this film on
student views on racism, you could select a seventh-grade
class at School B that’s fairly well matched to the class that
viewed the film. At both schools, you administer a reliable
and valid survey designed to assess attitudes toward racism
to each class. If you find that the seventh graders who viewed
the film at School A are significantly more likely to embrace
ideals of racial equality, you would reject your null hypothe-
sis in favor of alternate hypothesis. Namely, exposure to the

documentary film increases sensitivity to racial issues.
Nonequivalent Control Group (Nonrandomized)
A nonequivalent control group design (NEGD) is the most
common research design in the social sciences. Imagine
that at Grace Middle School, to assure a satisfactory
teacher–student ratio, there are two eighth grade classes.
Based on observation, the characteristics of the students
in Class 8A and Class 8B are at least roughly equivalent.
A standard pre-test–post-test design is adopted to assess
learning outcomes based on a new approach to teaching
math. A “map” of the research design looks like this:
Class 8A: Treatment group of size N à Pre-test à Treatment
à Post-test
Class 8B: Control group of size N à Pre-test à Post-test
All the students in both classes take a standardized math-
achievement test. However only the students in Class 8A
(the experimental group) are exposed to the new approach
for teaching math—the treatment. At the end of the experi-
ment, students in both classes take the math-achievement
test again. If the post-test scores are significantly higher for
the students in Class 8A, the superiority of the new approach
to teaching math is supported.
So, what’s wrong with this design? Nothing, not if Class 8B is
truly equivalent to Class 8A. But there’s the rub. The students
in the two classes weren’t assigned to the experimental or
control groups by random selection. That’s why an NEGD is

called a nonequivalent control group design. The researchers
can’t be absolutely sure that the two groups are equivalent.
Therefore, to the extent that the two groups aren’t equivalent
the findings may be distorted or flat-out wrong. The antidote
for this dilemma is replicating the experiment in different
places over time.
Time-Series Analysis
A researcher can use longitudinal data to conduct a time-
series analysis. The objective will be to assess outcomes of an
intervention before and after its implementation. Suppose the
issue is gang violence. The intervention we’ve implemented
involves coordinated efforts of police officers and psychiatric
social workers. Now if it turns out that gang violence is
noticeably reduced after the intervention, we can reasonably
assume that our program has been effective. On the other
hand, it can be difficult or impossible to show that the inter-
vention has, in itself, reduced gang violence. For example,
neighborhood parents may have decided to take an active
role in the matter or another project, the creation of a neigh-
borhood recreation center, may be causal factors.
Also, it’s possible that response to the intervention may take
awhile. For example, social psychologists are aware of what
are called sleeper effects. A person who has been exposed to
arguments against racism may not begin to revise his or her
attitudes for some period—which, oddly enough, turns out to
be about 21 days.
Lesson 3 113
Note: You should read the feature on pages 462–463, “Twin
Studies: Are Psychopaths Born or Made?” These days, the

issue of psychopathic or sociopathic behavior in high places
is getting a fair amount of attention. The social costs of psy-
chopathic behavior, illustrated by the scandals surrounding
powerful corporations like Enron and WorldCom, are far from
negligible. Pensions are lost. Jobs vanish, and public confidence
in institutions is seriously compromised.
Multiple Time Series (Difference in Difference)
Suppose Tarrytown and Whitesburg are Midwestern towns.
Both towns are located in the “corn belt” and the demograph-
ics of the two towns are similar. An intervention program is
launched to reduce teen delinquency rates in Tarrytown but
not in Whitesburg. Two years later the teenage delinquency
rate has risen by 12 percent in Whitesburg but only 2 percent
in Tarrytown. Note here that every three months we analyze
longitudinal data in both towns. In this context, we can think
of the teen population of Whitesburg as our control group
and the teen population of Tarrytown as our experimental
group. At the end of our two-year study we can assess the
so-called difference in difference for the teen delinquency
rates in the two towns. In sum, we’ve established that there
is a 10 percent difference between the delinquency rates in
the two towns that can, reasonably, be attributed to our
intervention.
Discontinuity Studies
Imagine a flock of bright high school graduates applying to
Calumet University. All hope to qualify for the university’s
Advanced Placement (AP) program. Data has shown that
graduates who get into the AP program earn significantly
higher salaries than students who don’t make the cut. The
qualifying score on the AP academic proficiency test is 90.
Researchers want to compare the impact of the AP program
on future earnings for those who make the cut and those

who don’t. So how can we do that?
Here’s how: Over a period of years, we examine the income
records for those who barely made the cut and those who
came close, but not quite close enough. We do so under the
assumption that psychometric and academic achievement
measures for applicants who score 87, 88, or 89 on the exam
will be very similar for those who score 90, 91, or 92. We can
then compare long-term income levels of graduates who just
made the cut or who just missed it. If the average income
difference is statistically significant, we have a measure of
the relative utility of the AP program among graduates of
Calumet University.
Multivariate Analysis
At this point, you’ll understand that the purpose and the power
of multivariate analysis is isolating a factor (independent
variable) that may be associated with the outcome (dependent
variable) while controlling for other factors that may effect
that outcome. Your textbook illustration designates Y in the
multiple regression equation as an assessment of changes in
sexual behavior resulting from a sex-education regimen of some
sort. For example, intervention in the form of a sex-education
program may be associated with a longer delay before onset
of sexual behavior. In the context of that discussion, it’s
noted that in this sort of analysis a particular kind of dummy
(bivariate, “yes or no”) variable will indicate whether or not the
individual participated in the sex-education program. That
variable will be called the treatment dummy. If the (regression-
line slope) coefficient on the treatment dummy, bnxn, is correct,
it will measure the relationship between participation in the

sex-education program and the onset of sexual activity, holding
other factors constant.
Two problems with multivariate analysis are identified. First,
recalling that numbers must always be interpreted, causality
may run in the opposite direction, such that the dependent
variable is causally influencing one or more dependent vari-
ables. Or, because variables may interact, it may be the case
that causality is running in both directions; that is, parental
role models may influence the student’s decision to enroll in
the sex-education program while peer attitudes are influenced
by individuals in that group who elect enrollment in that
program.
Lesson 3 115
Second, where we are focused on the observed relationship
between, say, onset of sexual activity, and sex education, it
may turn out that variation on both variables is caused by an
omitted variable. For example, strong religious commitment
may be the chief causal influence with respect to both of
these variables. Or, put another way, both A and B are
caused by C.
Self-Check 13

1. An antipoverty program has been implemented. Your effort to
evaluate that program includes
a multiple-regression analysis. If your equation was properly
conceived and includes a binary
independent variable called the _______ dummy, the coefficient
on that variable will measure
the relationship between those who were participants in the
program and those who were not.
2. The “gold standard” of program evaluation is the randomized
_______.
3. As a public policy analyst you want to find out if the
outcome of a program was actually caused
by the implementation of the program. In other words, you are
tackling the _______ problem.
4. Trying to distinguish causation from correlation, you might
determine that some _______ variable
has thrown the study off-track, such that the apparent
relationship between variables X and Y
is spurious, given that both are causal effects of a variable Z.
5. Using longitudinal data to assess the “before and after”
effects of a public policy intervention
is referred to as a/an _______ analysis.
6. The problem with using the approach from question 6 is that
before and after findings won’t
allow you to prove _______.

117
L
e
s
s
o
n
4
L
e
s
s
o
n
4
Making Policy
ASSIGNMENT 14
on
What we call institutions are, ultimately, shared ideas that
guide the ways in which we think and behave. Ritual head-
hunting and head-shrinking technologies are institutionalized

among the Jivaro of the Orinoco Basin. The institutionalized
concept of family among the Hopi of the American Southwest
normalizes a matrilineal kinship system in which you get
your name from your mother who, with her sisters, will own
the house you live in. Based on Enlightenment ideals, the
concept of freedom of expression is institutionalized in the
U.S. Constitution’s First Amendment.
In general, in any society, four generalized concepts are first to
appear as a society evolves—family, economy, some mode of
governance, and religion. The enormous array of institutions
in a modern society are “spin-offs” from these basic shared
concepts. In terms of how we see the world and how we
interact,
the dominant institutions in a modern Western or western-
ized society include some kind of representative governance
interacting with an economy informed by mass production
and mass society, generally organized around the concept of
sovereign nation states.* Within this dominant pattern, family
and kinship norms organize interpersonal relationships, as will
some form of what we may call a “cosmic” philosophy. These
idea-entities include long-established organized religions, sec-
ular philosophies like Marxism or Neo-liberalism, cults of all
kinds, and narrowly defined sects like the Old Order Amish.
*The modern concept of sovereign nation states originated with
the Treaty of Westphalia (1648) which ended the Hundred
Years’ War in Europe. The idea of ethnically and culturally
homogeneous nations identified with delineated geographic
borders was refined in France following the French Revolution
of 1789.

Public policy in a modern state is primarily focused on recon-
ciling the interests of the many faces of mass society—often
viewed as either publics or markets—with the inherent interests
of a for-profit market system as moderated by the institutions
of governance. In general, the most potent and influential
institutionalized form in the market system is the corporation.
The most potent institutions of American governance are
federal,
state, or local legislatures, courts, or executive bodies that
include regulatory agencies. This trio represents the American
constitutional ideology of checks and balances provided by the
executive, legislative, and judicial branches of government.
In the United States, the notion that a corporation is a legal
person with person rights originated from interpretations of
an 1886 federal court decision. Most recently, the idea that
corporations have person rights has been greatly expanded
by Supreme Court decisions declaring that money is a form
free speech and the 2010 ruling in Citizens United v. Federal
Election Commission declaring the right of corporations to
make essentially unlimited political campaign contributions
under the First Amendment.
The Role of Institutions
Ideally, institutions must enact three roles:
1. Gather together (aggregate) the preferences of people that
constitute a group. A group in this sense can be called a
public. Working mothers, residents of a neighborhood,
extractive industry workers (miners), and likely voters
can be thought of as publics.
2. Implement and enforce communal decisions. The thorny
issue here is that, by definition, public policy involves
changing behaviors, often through coercion. Think about

the Civil Rights Movement and civil rights legislation of
the 1960s to get the point.
3. Protect minorities and dissenting views. Nothing resem-
bling a representative democracy can exist if dissent is
forbidden or suppressed. By the same token, a system
of governance that ignores the rights and concerns of
minorities creates oppressive majorities and, thereby,
defeats the ideal of government of, for, and by the people.
Lesson 4 119
Why Can’t Government
Operate Like a Business?
Your textbook offers five reasons why business practices may
be
antithetical to the welfare of a society as whole. But basically,
the matter boils down to a simple observation. Private sector
enterprises are driven by the profit motive. That’s fine for
allocating a wide range of goods and services that can be
understood in terms of supply and demand measured in
dollars and cents. However, some kinds of goods and services
required for the general welfare, like low-cost public housing
or unemployment insurance, have to be administered on a
nonprofit basis, simply to keep costs under control. At the
same time, appropriate regulatory measures, like mandated
food safety standards, must be provided by public agencies
that are accountable to the public, not the commercial
enterprises that must meet those standards.
Institutions and Outcomes
The prime directive for people who craft public policy is

improving people’s lives. And so it follows that the main pur-
pose of institutional analysis must be aimed at enhancing
and protecting people’s lives. In turn, that means figuring out
how and why institutions produce outcomes that either aid or
detract from the public welfare. For example, if roads, bridges,
and other aspects of a country’s infrastructure are deficient,
what policies might be initiated to cost-effectively upgrade the
country’s infrastructure?
Note: As you study the main points under this topic, be sure
to give some special attention to the Policy in the Real World
feature, “Institutions and Development: The Institutional
Legacy of Colonialism,” on pages 492–493.
The Attributes of Effective Institutions
In general terms, effective institutions are laws, organizations,
and accepted modes of behavior that allow a society to make
the best use of its resources. Specifically, effective institutions
have certain attributes:
n Authority. Let’s assume you have access to safe clean water.
If you do, it will because some institution will have the
authority to assure that access. That institution may be
a state water authority that sets standards for local
water treatment plants. That’s a simple example, but it
illustrates the main idea here. A policy can exist only if
some institution has the authority to administer the policy.
In that context, there’s a downside to the geopolitical
organization of our planet-based nation state sovereignty.
Assume that global climate change must be addressed in
the interest of humanity-at-large. How can that happen

absent an internationally recognized authority enabled
to administer ecology-friendly polices worldwide?
n Legitimacy. An institution can be thought of as legitimate
if its policies and practices are in accord with the preferences
of people whose lives are affected by that institution. In
some cases, legitimacy is based in custom and tradition.
Norms and practices surrounding the institution of marriage
offer one example. At the level of the state, monarchy may
be granted legitimacy even though actual governance is
implemented by a parliament composed of democratically
elected representatives. On the other hand, people gener-
ally feel that autocratic governance backed by force and
fear is illegitimate.
n Mission. An institution must have a clearly defined purpose
if it is to effectively improve people’s lives. For example,
the Rural Electrification Administration (REA) is a federal
agency under the Department of Agriculture. It was
established under the presidency of Franklin Roosevelt
in the 1930s, during the Great Depression. Its goal was
extending electric power to rural areas to improve both the
lives and productivity of farmers and ranchers located far
from urban centers. The program was justified because
private sector energy providers refused to extend the
Lesson 4 121
range of substations and transmission lines beyond a
profitable distance from power plants. The REA program
worked. All of America was “electrified, and REA co-ops
continue to be part of people’s lives across the prairie
and mountain states to this day.

n Money, people, and expertise. A budget, personnel, and
expertise are absolute requisites if programs like the REA
are to be initiated, implemented, and sustained. One of
the pressing problems in less developed countries is the
absence of these fundamental resources. In your text-
book discussion of this topic, comparisons are made
between the effectiveness of private and public sector
approaches to utilizing these resources. In either case,
attracting funding is a major concern.
n Internal incentives. The key idea here is that institutions
are more likely to be effective if the incentives offered to
the people carrying out a policy are likely to align their
self-interest with the goals of the program or policy. An
infamous example of misaligned incentives has, historically,
been illustrated by the New Orleans Police Department.
Officers of the NOPD are notoriously underpaid and all
but universally tempted to accept bribes from vested
interests that don’t serve the public interest. Another
example has to do with tenure policies for college teachers.
If the route to tenure is basically measured in journal
articles published, academics have weak incentives for
devoting time and energy to teaching.
n Enforcement. The basic idea here is simple: A law that
can’t be enforced is not really a law at all. Similarly, if a
law can be only weakly enforced, it’s a lousy law. America’s
experiment with prohibition simply created a massive
illegal market for alcoholic beverages.
n Rules. Rules have profound effects on how power and
resources are allocated within an institution. Recall, if
you will, the insights you may have gained from consid-
ering how bills may or may not become laws depending
on the organization and procedures of the House of
Representatives and the Senate. Rules are necessary,

but some rules are better than others when it comes to
initiating and implementing public policy.
As you wrap up your study of this chapter, think about these
three observations:
n The same set of preferences filtered through different
institutions will produce different outcomes.
n Often, the quality of an institution will dictate the quality
of policy outcomes.
n Because power structures are strongly inclined to sustain
the status quo, institutions are inherently difficult to change.
You’ll also want to spend some serious time pondering the
For Discussion feature, “Global Governance: Reform at the
United Nations,” on pages 506–509. Pay special attention
the five criteria for the legitimate use of force on page 508.
Lesson 4 123
Self-Check 14
1. The fact that public institutions will take on socially
desirable tasks that the private
sector won’t or can’t accomplish is one reason why government
institutions can’t

operate like a/an _______.
2. Among other things, institutional _______ can help us reform
defective institutions and make
sense of observed public policy outcomes.
3. An institution is most likely to be effective if the _______
incentives for employees, like
adequate pay and a good job fit, are aligned with and
complement the institution’s mission.
4. Thinking about the attributes of effective institutions, like
authority and enforcement,
legislative procedures, like the committee system and the
filibuster, exemplify the importance
of _______ that may either facilitate or obstruct the formulation
of effective public policies.
5. In developed countries, _______ is the most common
mechanism for attributing legitimacy to
an institution.
6. Two important roles of institutions include aggregating
people’s _______ and protecting
minorities and people’s right to _______.
ASSIGNMENT 15
on

How can public policy change people’s behavior in ways that
lead to better social outcomes? That central question has
been addressed throughout this course. You’ve explored basic
concepts in economics and been introduced to the analytical
tools used by public policy analysts. In that context you’ve
pondered game theory, particularly as illustrated by the pris-
oner’s dilemma. You’ve explored a host of principles that
should inform effective public policies and, with an overview
of the all these things, pondered a wide variety of illustrative
“snap shots” of public policy issues in the real world.
A famous poem written by T. S. Eliot includes this passage:
“We will return to where we began to know the place for the
first time.” That idea characterizes this final chapter of your
textbook. Beginning with a case study focused on poverty
in Brazil, you’ll be afforded the opportunity to pull together
all you’ve learned to this point. You’ll notice, for example,
that you’ve come full circle in considering the steps in the
policy-making process. The content of this chapter is both a
summary and a review of your textbook and of this course.
Note: Use this chapter to reflect on both what you’ve learned
and what you may have missed. To do that, go back to sections
of your textbook that will help refresh your understanding of
key terms and basic concepts.
Lesson 4 125
Self-Check 15

1. In the context of evaluating policy options, the downside of
cost-benefit analysis is that it fails
to take _______ effects into consideration.
2. _______ is using language in such a way as to broaden the
appeal of a policy proposal.
3. Who will be affected? How will people respond? Both of
these questions are important focuses
of _______ analysis.
4. Step 1 in the policy process is identifying the potential
_______.
5. A/An _______ tax, like a tax on alcoholic beverages, is
meant to offset the social cost of
inadvisable or addictive behaviors related to the use of alcohol.
6. Step 2 in the policy process is figuring out why the _______
isn’t delivering a desirable outcome.
NOTES
127
G

ra
d
e
d
P
ro
je
c
t
G
ra
d
e
d
P
ro
je
c
t
R
e
s
e
a

rc
h
P
ro
je
c
t
R
e
s
e
a
rc
h
P
ro
je
c
t
BACKGROUND AND
PREPARATION
For this research project, you’ll write a two-part essay identi-
fying an ineffective public policy and suggesting how it can be
improved. In the first part, you’ll choose a topic and discuss
a related the current policy that you feel is ineffective. In the

second part, you’ll research your point of view and write about
changes you would make to the policy to make it more
effective.
You must use at least two outside sources in addition to the
textbook and you must document the sources using a standard
format, such as MLA or APA.
OBJECTIVE
This research project will help you apply what you’ve learned
about public policy. More precisely, you’ll have a chance to
make some public policy suggestions of your own on a topic
of your choice.
PROCEDURE
Write an essay of at least 500 words, typed and double spaced
in a standard, 12-point font, such as Times New Roman. You
may use both Internet and print sources for your research. The
format of your essay must include an introductory paragraph
that summarizes what the paper is about and a concluding
paragraph that summarizes your observations and conclusions.
Part 1
Select a topic related to one of the following issues:
n Immigration
n Education
n Entitlement programs
n Foreign policy
n Economy

Once you’ve selected your topic, identify a current policy
within
the topic that you feel isn’t effective. Read related journal arti-
cles or books and use the most relevant material to explore
the issue. Discuss relevant ethical, ideological, practical, and
any other relevant problems you find with respect to the
policy issue you’ve selected. Analyze every aspect of the
policy;
determine which goals are being met, which aren’t being met,
and the reasons for the success or failure of the policy.
Part 2
Once you’ve completed assessment of the policy, introduce
the changes you would make to remedy the problems you
identified in Part 1. Refer to your research sources as needed,
using proper citation for any quotes or paraphrased ideas
that aren’t your own. Discuss how your changes would
address the problems you’ve identified:
n Compare the merits of your proposed policy to the
existing policy
n Analyze both policies with respect to the empirical and
normative dimensions, the cost-benefit analysis, short-
and long-term goals, and indirect benefits as discussed
in your textbook
n Discuss how your policy would be implemented in light
of groups (stakeholders) that would favor or oppose your
policy proposals

n Explain and discuss just how your policy would
be administered
Lesson 1 129
WRITING GUIDELINES
1. Type your submission, double-spaced, in a standard 12-
point font. Use a standard document format with one-inch
margins. (Don’t use any fancy or cursive fonts.)
2. Include the following information at the top of your paper:
a. Name and complete mailing address
b. Student number
c. Course title and number (Introduction to
Public Policy, SSC265)
d. Research project number (50178700)
3. Read the assignment carefully and complete both parts,
using proper citations according to either an APA or an
MLA style guide.
4. Be specific. Limit your submission to the assigned issues.
5. Include a reference page in either APA or MLA style. On
this page, list websites, journals, and all other references
used in preparing your submission.
6. Proofread your work carefully. Check for correct spelling,

grammar, punctuation, and capitalization.
GRADING CRITERIA
Your project will be evaluated according to the following
criteria.
Content 60 percent
Written communication 15 percent
Format 25 percent
Here’s a brief explanation of each of these points.
Content
The student
n Provides clear answers to the assigned issues
n Addresses the issues in complete sentences
n Supports his or her opinion by citing specific information
from references using correct APA or MLA guidelines for
citations and references
n Stays focused on the assigned issues
n Writes in his or her own words and uses quotation
marks to indicate direct quotations

Written Communication
The student
n Addresses the issue in complete paragraphs that include
an introductory sentence, at least four sentences of
explanation, and a concluding sentence
n Uses correct grammar, spelling, punctuation, and sen-
tence structure
n Provides clear organization (for example, uses transi-
tional words like first, however, on the other hand, and so
on, consequently, since, next, and when)
n Makes sure the paper contains no typographical errors
Format
The paper is double-spaced and typed in font size 12. It
includes the student’s
n Name and complete mailing address
n Student number
n Course title and number (Introduction to
Public Policy, SSC265)
n Research project number (50178700)
Lesson 1 131
SUBMITTING YOUR ASSIGNMENT

Follow this procedure to submit your assignment online:
1. On your computer, save a revised and corrected version
of your essay.
2. Go to http://www.pennfoster.edu and log in.
3. Go to My Courses.
4. Click on Take Exam next to the lesson you’re working on.
5. Enter your e-mail address in the box provided. (Note:
This information is required for online submission.)
6. Attach your file as follows:
a. Click on the Browse box.
b. Locate the file you wish to attach.
c. Double-click on the file.
d. Click on Upload File.
7. Click on Submit Files.
After you submit the assignment, you should receive a confir-
mation e-mail with a tracking number. If you don’t receive this
number within 24 hours, you must resubmit your assignment.
NOTES

Self-Check 1
1. True
2. True
3. False
4. False
5. True
Self-Check 2
1. False
2. True
3. True
4. True
5. True
Self-Check 3
1. True
2. False
3. True
4. True
5. False

6. True
Self-Check 4
1. e
2. g
3. d
4. b
5. h
6. a
133
A
n
s
w
e
r
s
A
n
s
w
e
r

s
Self-Check 5
1. communist
2. horizontal
3. index
4. initial endowment
5. Pareto
6. deadweight
Self-Check 6
1. rent seeker
2. median voter
3. Earmarks
4. Arrow’s
5. comparative
6. logrolling
7. Byrd
Self-Check 7
1. d

2. a
3. h
4. b
5. e
6. f
Lesson 1 135
Self-Check 8
1. Pigovian
2. transaction
3. paternalistic
4. monopoly, pricing
5. nonexclusivity
6. intellectual
Self-Check 9
1. relative
2. 5 + 6/2 = 5.5
3. selection
4. standard deviation

5. positive-findings
6. Longitudinal
Self-Check 10
1. five
2. decision tree
3. simple random
4. type I
5. central, sample
6. alternative
Self-Check 11
1. causality
2. sign
3. a
4. standard
5. Ordinary least
6. residual
Self-Check 12

1. opportunity
2. hedonic
3. Willingness
4. sensitivity
5. tradable
6. shadow
Self-Check 13
1. treatment
2. experiment
3. identification
4. omitted
5. time-series
6. causality
Lesson 1 137
Self-Check 14
1. business
2. analysis
3. internal

4. rules
5. democracy
6. preferences, dissent
Self-Check 15
1. distributional
2. Framing
3. stakeholder
4. benefit
5. Pigovian
6. market

Study GuideIntroduction toPublic PolicyByRobert G..docx

More Related Content

Similar to Study GuideIntroduction toPublic PolicyByRobert G..docx (20)

More from hanneloremccaffery (20)

Recently uploaded (20)

Study GuideIntroduction toPublic PolicyByRobert G..docx