n PART ONE

Concepts
The chapters in Part 1 establish the concepts used throughout the book to assess
the existence and institutions of the welfare state, including relevant denitions
(Chapter 1), the historical backdrop (Chapter 2), political theory (Chapter 3), eco-
nomic theory (Chapters 4 and 5), and problems of dening and measuring poverty
and inequality (Chapter 6).
Readers with less technical background (or less time) can read the non-technical
summaries at the end of Chapters 36.
EOTC01 14/4/04 17:12 Page 1
EOTC01 14/4/04 17:12 Page 2
Introduction
[The duties of the state are] . . . rst . . . that of protecting the society from the violence and
invasion of other independent societies . . . second . . . that of protecting, as far as possible, every
member of the society from the injustice or oppression of every other member of it . . . third . . .
that of erecting and maintaining those publick institutions and those publick works which, though
they may be in the highest degree advantageous to a great society, are of such a nature, that the
prot could never repay the expence to any individual or small number of individuals.
(Adam Smith, 1776)
1. The approach
1.1. The central argument
One of the wellsprings of this book was the exuberant insistence of some of my students
and colleagues that economics appeared largely irrelevant to the central concerns of
social policy. They had a point, and this book, like previous editions, is an attempt to
remedy their grievances and to assert the importance of economics. To address the
concern about relevance, discussion relates economic theory to different notions of social
justice and to the historical development of the welfare state. In stressing the importance
of economics, two results stand out. First, the welfare state is not a subject apart, but ts
naturally into the framework of economic analysis. Secondly, the theoretical arguments
support the existence of the welfare state not only for well-known equity reasons but
alsoand powerfullyin ebciency terms. This is an area in which economic theory
is capable of strong results that can justify the general idea of the welfare state and, to a
surprising extent, can do so without resort to ideology.
To keep the subject manageable, the book is explicitly about economic theory and
its application to the welfare state. It is not a book about comparative systems, since the
welfare state, though existing in all countries in the wider Europe and, more generally, in
all advanced industrial countries, takes very different forms (Esping-Andersen et al. 2002;
Neil Gilbert 2002). The book has a very brief institutional description, mainly of the UK,
to motivate the theory, plus references to more detailed information and, where possible,
wider European and US examples, but with no attempt at systematic coverage. A second
restriction is that, though the book discusses in some detail how welfare-state services can
be nanced, there is less detailed discussion of how they should be delivered. Economic
theory has much to say about funding; in contrast, though economic fundamentals (for
1
EOTC01 14/4/04 17:12 Page 3
4 CONCEPTS
example, a sensible incentive structure) are important on the delivery side, institutional
organization and institutional change are things about which economic theory has less
to say.
The book addresses two broad questions: what theoretical arguments justify the various
parts of the welfare state in an industrial economy; and, given these arguments of principle,
how sensible (or otherwise) are arrangements in the UK
1
and other countries?
The approach is best illustrated by two questions that permeate throughout.
1. What are the aims of policy?
2. By what methods are those aims best achieved?
Question 1 is broad ranging. There is general agreement that the major aims of policy in
Western societies include eBciency in the use of resources; their distribution in accordance
with equity or justice; and the preservation of individual freedom. These aims, however, can
be dened in different ways, and may be accorded different weights. To a utilitarian,
2
the
aim of policy is to maximize total welfare; to Rawls the aim is social justice, dened in a
particular way; libertarians make their main aim individual freedom, and socialists their
prime concern equality. The answer to question 1 is explicitly normative and largely
ideological. The objectives of the welfare state are discussed in more detail in Section 2.2.
In contrast, once question 1 has been answered, question 2 should be treated not as
ideological but as technicalthat is, it raises a positive issue. Whether a given aim should be
pursued by market allocation or by public provision depends on which of these methods
more nearly achieves the chosen aim. Market allocation is neither good nor badit is
useful in some instancesfor example, private markets for food are generally effective
in achieving the aim that people should not starve; in others, however (it is argued in
Chapter 12 that health care is one), the market mechanism works less well, and substantial
state intervention can contribute to ebciency and to justice. Similarly, public provision
is neither good nor bad, but useful in some cases, less so in others. One of the questions
throughout is which method is more useful in different areas of the welfare state.
The distinction between aims and methods is fundamental, and bears reinforcement.
Consider two central questions that all societies face.
How much redistribution (of income, wealth, power, etc.) should there be?
How should economic activity be organized, through markets, or central planning,
or as a mixed economy?
The rst question is clearly ideological and normative; it is an aims question and so prop-
erly the subject of political debate. But, once that question has been answered, the second
question is largely one of method (i.e. a positive issue) and better treated as technical than
as political. The approach is explained in detail in Chapters 3 and 4, and summarized in
the concluding section of Chapter 4.
1
The United Kingdom (UK) is Great Britain and Northern Ireland (Act of Union with Ireland 1800;
Government of Ireland Act 1920). Britain (or Great Britain) consists of England, with Wales and Scotland (Act of
Union with Scotland 1706).
2
Utilitarianism and other theories of society, including those of Rawls and libertarian and socialist writers, are
discussed in Chapter 3.
EOTC01 14/4/04 17:12 Page 4
1. I NTRODUCTI ON 5
1.2. Organization of the book
Part 1 sets the scene, starting in Chapter 2 with the historical development of the welfare
state in the UK, including some comparison with other countries, particularly the USA.
The three chapters that follow are the theoretical heart of the book: Chapter 3 discusses
denitions of social justice and their different implications for the welfare state; Chap-
ter 4 sets out the economic theory of state intervention and Chapter 5 the theory of
insurance. Chapter 6 discusses problems of denition and measurement, particularly of
poverty and inequality. For readers who are dibdent about their theoretical background,
each of the conceptual chapters (3, 4, 5, and 6) has a non-technical appendix summariz-
ing the essential material; and technical terms are explained in the Glossary.
Three major threads developed in Part 1 run through the rest of the book: the social-
welfare-maximization problem; alternative denitions of social justice; and measure-
ment problems. The social-welfare-maximization problem (set out in Chapter 4) is the
conventional starting point for economic theory. An important theorem states that
under appropriate assumptions a competitive market equilibrium will allocate resources
ebciently. It is argued that, where these conditions hold, the role of the state, if any, is
limited to redistribution; conversely, where these conditions fail, there may be ebciency
grounds for intervention in a variety of forms. The second major theme is social justice.
The denition chosen will determine the weights assigned to different individuals, with
major implications for the form and extent of interventionfor example, whether
people with no income should be supported at subsistence or at a higher level. The third
thread, discussed in Chapter 6, concerns problems of denition and measurement. Many
variables are hard to dene and, once dened, hard to measure. A crucial and recurrent
dibculty is that utility
3
is not measurable. This makes it hard both to measure living
standards and to compare them. Costs or benetsof health care or education, for
examplemay also be hard to measure.
As far as possible, each chapter in Parts 2 and 3 has a similar layout to clarify the
structure of the argument. Each chapter discusses in turn: the aims of policy; the methods
by which they might be achievedthat is, the theoretical arguments about intervention
for reasons of ebciency and social justice; assessment in the light of this theoretical
discussion of the appropriateness (or otherwise) of the UK and other systems, including
discussion of the empirical literature; and reform.
Part 2 analyses cash transfers. Chapter 7 briey describes the nances of the welfare
state. Chapter 8 looks at unemployment and sickness benets, whose primary purpose is
insurance; Chapter 9 discusses retirement pensions, whose main purpose is consumption
smoothing;
4
and Chapter 10 reviews non-contributory benets, whose main purpose is
poverty relief. Each chapter starts with the theory and then assesses the practice. Chapter 11
considers a variety of reform strategies. Part 3 discusses provision in kind. Chapter 12, on
health, analyses the theoretical arguments for public production and allocation, assesses
the effectiveness of the UK National Health Service in comparison with systems in other
countries, and discusses alternative ways in which health care might be organized.
Chapters 13 and 14 cover similar ground for school education and higher education.
3
See the Glossary.
4
See the Glossary.
EOTC01 14/4/04 17:12 Page 5
6 CONCEPTS
The conclusions of the book are summarized in Chapter 15, which picks up some of the
questions asked at the end of this chapter. Readers in a hurry can get an idea of the books
approach and its main conclusions by reading Chapter 15 and the concluding sections of
Chapters 4 (economic and political theory), 11 (income support), and 12, 13, and 14
(health care, school education, and higher education, respectively).
2. The welfare state and its objectives
2.1. Dening the welfare state
This section denes the welfare state in three ways: in principle, for the purposes of this
book, and in practice.
the welfare state in principle. We shall see in Chapter 6 that important concepts such
as poverty and equality of opportunity are hard, if not impossible, to dene in principle,
and even harder to measure. The concept of the welfare state similarly dees precise
denition, and I make no serious attempt to offer one. Even Richard Titmuss (1958)
ducked the problemthat book is called Essays on The Welfare State (his quotes). As he
later put it, I am no more enamoured today of the indenable abstraction The Welfare
State than I was some twenty years ago when . . . the term acquired an international as
well as a national popularity (Titmuss 1968: 124). Three areas of complication stand out
(for fuller discussion, see Glennerster 2003a: ch. 1).
1. Welfare derives from many sources in addition to state activity. Individual welfare derives
from at least four sources.
The labour market is arguably the most important, rst through wage income. Full
employment is a major component of welfare broadly dened. High levels of
employment and rising labour productivity over the 1950s and 1960s were at least
as much an equalizing force as redistribution.
5
In addition to wage income, rms
(individually or on an industry-wide basis, voluntarily or under legal compulsion)
provide occupational welfare in the face of sickness, injury, and retirement.
Private provision includes voluntary private insurance and individual saving.
Voluntary welfare arises both within the family and outside, where people give time
free or at a below-market price, or make voluntary charitable donations in other forms.
The state intervenes by providing cash benets and benets in kind. It also contri-
butes through tax concessions to the nance of occupational and private provision.
2. Modes of delivery are also diverse. A service may be funded by the state, but it does
not follow that it must necessarily be publicly produced. The state can produce a service
itself and supply it to recipients at no charge (e.g. health care under the National Health
Service); or it can pay for goods produced in the private sector (e.g. free pharmaceutical
5
As discussed in Chapter 2, Section 5.1, full employment was one of Beveridges central assumptions.
EOTC01 14/4/04 17:12 Page 6
1. I NTRODUCTI ON 7
drugs under the National Health Service); or it can give individuals money (either explicitly
or in the form of tax relief) to make their own purchases (e.g. tax relief in some countries
for private medical insurance premiums). The issue of privatization, as we shall see in
Chapter 4, Section 6, is more complex than is often recognized in public discussion.
3. The boundaries of the welfare state are not well dened. Though the states role should
not be exaggerated, neither should it be understated. Some typically excluded activities
such as public health and environmental policies are very similar in purpose to ones that
are included.
Welfare is thus a mosaic, with diversity both in its source and in the manner of its
delivery. Nevertheless the state, through various levels of government, is much the most
important single agency involved in most industrialized countries. Throughout the book
the term welfare state is used as a shorthand for the states activities in four broad areas:
cash benets; health care; education; and food, housing, and other welfare services.
In broad terms the modern welfare state comprises cash benets and benets in kind.
The latter embrace a wide range of activities that can include education, medical care, and
more general forms of care for the inrm, the mentally and physically handicapped, and
children in need of protection. Cash benets have two major components.
1. Social insurance is awarded without an income or wealth test, generally on the basis
of (a) previous contributions and (b) the occurrence of a specied contingency, such
as becoming unemployed or reaching a specied age.
2. Non-contributory benets are of two sorts. So-called universal benets are awarded on
the basis of a specied contingency, without either contributions or an income
test. Major examples in the UK are child benet and the National Health Service (dis-
cussed in Chapters 10 and 12, respectively). Social assistance is awarded on the basis
of an income test. It is generally a benet of last resort, designed to help individuals
and families who are in poverty, whether as an exceptional emergency, or because
they are not covered by social insurance, or as a supplement to social insurance.
the welfare state for the purposes of this book. The welfare state exists to enhance the
welfare of people who (a) are weak and vulnerable, largely by providing social care, (b) are
poor, largely through redistributive income transfers, or (c) are neither vulnerable nor
poor, by organizing cash benets to provide insurance and consumption smoothing, and
by providing medical insurance and school education.
This book is mainly about (b) and (c), covering cash benets, health care, and education.
Two omissions are deliberate. There is little discussion of element (a)for example, pro-
tection of vulnerable children, the frail elderly and people with emotional and physical
disabilities. This is not because they are unimportant (absolutely not the case), but be-
cause they are topics about which economics has relatively little to say. Secondly, welfare
depends also on access to food and housing. These topics do not gure prominently for
different reasons: each raises economic issues that are largely settled territory: markets, it
will be argued, work well for these two vitally important goods. Both sets of markets need
regulation; but in both cases, the poor are best protected by income transfers, allowing
them to buy food and shelter. Thus food is discussed in the following chapters mainly
in terms of regulation to ensure it is safe, and housing largely in terms of income transfers
EOTC01 14/4/04 17:12 Page 7
G
D
P

1
,
1
0
8

b
n
.
T
O
T
A
L

P
U
B
L
I
C

S
P
E
N
D
I
N
G

4
5
5
.
7

b
n
.
W
E
L
F
A
R
E

S
T
A
T
E

2
6
0
.
6

b
n
.
[
1
0
0
%
]
C
A
S
H

B
E
N
E
F
I
T
S
a

1
1
4
.
7

b
n
.
[
4
4
%
]
B
E
N
E
F
I
T
S

I
N

K
I
N
D

1
4
5
.
9

b
n
.
[
5
6
.
0
%
]
C
O
N
T
R
I
B
U
T
O
R
Y

5
6

b
n
.
[
2
1
.
5
%
]
N
O
N
-
C
O
N
T
R
I
B
U
T
O
R
Y

5
8
.
7

b
n
.
[
2
2
.
5
%
]
N
H
S

7
4
.
8

b
n
.
[
2
8
.
7
%
]
P
E
R
S
O
N
A
L

S
O
C
I
A
L
S
E
R
V
I
C
E
S

1
2
.
5

b
n
.
[
4
.
8
%
]
E
D
U
C
A
T
I
O
N

5
8
.
6

b
n
.
[
2
2
.
5
%
]
S
i
c
k
n
e
s
s

a
n
d
i
n
c
a
p
a
c
i
t
y

6
.
9

b
n
.
R
e
t
i
r
e
m
e
n
t

4
6
.
4

b
n
.
O
t
h
e
r

2
.
6

b
n
.
J
o
b
s
e
e
k
e
r

s
a
l
l
o
w
a
n
c
e

2
.
3

b
n
.
I
n
c
o
m
e
s
u
p
p
o
r
t

9
.
9

b
n
.
M
i
n
i
m
u
m
i
n
c
o
m
e
g
u
a
r
a
n
t
e
e

5
.
1

b
n
.
H
o
u
s
i
n
g
b
e
n
e

1
5
.
9

b
n
.
O
t
h
e
r

3

b
n
.
I
n
v
a
l
i
d
i
t
y
b
e
n
e

t
s

1
3
.
4

b
n
.
C
h
i
l
d
b
e
n
e

9
.
1

b
n
.
F
i
g
.

1
.
1
O
v
e
r
v
i
e
w

o
f

t
h
e

w
e
l
f
a
r
e

s
t
a
t
e
,

U
K
,

2
0
0
3
/
4

(
e
s
t
.
)

(

b
n
.

a
n
d

%
)
a
T
h
e

g
u
r
e

e
x
c
l
u
d
e
s

a
b
o
u
t

1
2

b
i
l
l
i
o
n

o
f

t
a
x

c
r
e
d
i
t
s

w
h
o
s
e

f
u
n
c
t
i
o
n

i
s

i
d
e
n
t
i
c
a
l

t
o

c
a
s
h

b
e
n
e

t
s
.
S
o
u
r
c
e
s
:

U
K

T
r
e
a
s
u
r
y

(
2
0
0
2
a
:

t
a
b
l
e

A
3
;

2
0
0
3
a
:

t
a
b
l
e
s

C
3
,

C
1
1
)
,

U
K

D
W
P

(
2
0
0
2
a
:

t
a
b
l
e

7
)
.
EOTC01 14/4/04 17:12 Page 8
1. I NTRODUCTI ON 9
to poor people to support their housing needs. Other, equally important aspects of
housingsuch as unsafe neighbourhoodsraise issues more usefully tackled by discip-
lines other than economics.
the welfare state in practice. The UK welfare state can be taken to comprise, at a minimum,
the publicly provided benets (representing about 23.5 per cent of gross domestic product)
shown in Figure 1.1, together with the contributions that pay for them. Cash benets
follow the pattern described above. National insurance is payable to people with an
adequate contributions record; benets cover, inter alia, unemployment, sickness (short-
and long-term), and retirement, of which the last (about 18 per cent of social spending)
is much the largest. Non-contributory benets include child benet (a weekly cash pay-
ment to the parent or guardian of every child), and income support (i.e. social assistance
for people with little or no other income). The major benets in kind are the National
Health Service (29 per cent of total social spending) and education (22.5 per cent).
As Figure 1.2 shows, the UK is by no means unusual. The data show 1998 gures for
public social spending, which the OECD dene to include all cash benets, health care, and
social services, but to exclude spending on education. Spending was highest in the Nordic
20 20 18 16 14 12 10 8 6 4 2 0 0 2 4 6 8 10 12 14 16 18
Sweden (31.0)
Denmark (29.8)
France (28.8)
Switzerland (28.1)
Norway (27.0)
Austria (26.8)
Finland (26.5)
Germany (26)
Italy (25.1)
Belgium (24.5)
Netherlands (24.5)
EU (24.2)
a
Poland (22.8)
Greece (22.7)
Luxembourg (22.1)
New Zealand (21.0)
OECD (20.8)
a
United Kingdom (20.8)
Spain (19.7)
Czech Republic (19.4)
Iceland (18.4)
Portugal (18.2)
Canada (18.0)
Australia (17.8)
Ireland (15.8)
Japan (14.7)
United States (14.6)
Slovak Republic (13.6)
b
Turkey (11.6)
Mexico (8.2)
Korea (5.9)
Cash benets Services
Income support to the working age population
Pensions (old age and survivors)
Health
Other social services
Fig. 1.2. Public social expenditure by broad social policy area, OECD 1998 (% of GDP)
Note: Countries are ranked by decreasing order of total public social expenditure as a percentage of GDP.
a
OECD and EU are unweighted averages.
b
Slovak Republic: data for total are underestimated because data about health are not available yet.
Source: OECD (2003c: 55).
EOTC01 14/4/04 17:12 Page 9
10 CONCEPTS
countries and, among the richer countries, lowest in Japan and the USA. UK spending was
equal to the OECD average, somewhat below the average for the EU.
2.2. The objectives of the welfare state
The objectives of social institutions, as in any other area of economic policy, are
ebciency, equity, and administrative feasibility. In this context, however, it is useful to
adopt a more rened categorization.
efficiency has at least three aspects.
1. Macro-eBciency. The ebcient fraction of GDP should be devoted to the totality of
welfare-state institutionsfor example, policy should avoid distortions leading to
cost explosions.
2. Micro-eBciency. Policy should ensure the ebcient division of total welfare-state
resources between the different cash benets, different types of medical treatment,
and different kinds of educational activity.
3. Incentives. Where institutions are publicly funded, their nance and the structure
of benets should minimize adverse effects on labour supply, employment,
and saving.
Objectives 13 are different aspects of allocative eBciency, sometimesparticularly in
the context of health care and educationreferred to as external eBciency. As an example,
if the objective of health policy is to maximize the health of the population, external
ebciency is concerned with producing the quantity, quality, and mix of health inter-
ventions (including preventive care and education about diet and lifestyle) that bring
about the greatest improvement in health.
supporting living standards, the second strategic aim, has at least three components.
4. Poverty relief. No individual or household should fall below a minimum standard
of living. The aim could be to eliminate poverty or to alleviate it. As discussed in
Chapter 6, there is no analytically satisfactory way of dening a poverty line, so
the denition of the minimum standard is largely normative. Once the poverty line
has been decided, the effectiveness of the system is measured by statistics relating to
how many people are below the poverty line (headcount measures), by how much
(poverty-gap measures), and for how long (life-cycle and intergenerational matters).
5. Insurance. No one should face an unexpected and unacceptably large drop in her
living standard. This is a major objective of unemployment benets and most
health-related benets. Its success is measured by the replacement ratio, which
shows a persons income when on benet in comparison with her previous income.
6. Consumption smoothing. Institutions should enable individuals to reallocate con-
sumption over their lifetime. As discussed in Chapter 9, individuals can redistribute
from their younger to their older selves (an actuarial private pension scheme); or
such redistribution could be notional (an unfunded state pension (Samuelson 1958)).
Alternatively, there could be tax-funded provision, with no pretence of individual
EOTC01 14/4/04 17:12 Page 10
1. I NTRODUCTI ON 11
contributions, to groups whose stage in the life cycle suggests that they are likely to
be nancially constrained (e.g. benets for families with young children). Analog-
ously, student loans, discussed in Chapter 14, allow people to redistribute from their
middle years to their younger self.
These three objectives largely shape the rest of the book. The simple Fisher model dis-
cussed in Chapter 4, Section 3.2 and illustrated in Figure 4.5 illustrates a persons choices
between consuming now (period 1) or later (period 2)that is, his or her options for
consumption smoothing. A person can consume less in period 1 and save, so as to be able
to consume more in period 2; similarly, someone who wants to consume more earlier can
borrow, thus consuming more in period 1.
The simple model assumes rational behaviour in a world of certainty and competitive
markets. In that world, the welfare state is largely unnecessary: there is no risk, and hence
no need for insurance; people provide for their old age through voluntary saving; and
temporary poverty is dealt with by borrowing or saving. Thus the only role for the welfare
state is to provide poverty relief for someone who is lifetime poor.
The rest of the book is largely about the effects of relaxing the certainty assumption. A
central argument is that imperfect information, risk, and uncertainty give the state (or
parastatal agencies) a key role, not only in providing poverty relief, but also in ensuring
that people have access to insurance and consumption smoothing that the private
market would provide ineffectively or not at all.
Figure 1.3 shows the broad division of cash benets into spending on the three object-
ives. Just over one-third of spending on cash benets is directly redistributive from rich to
poor, the remaining two-thirds, involving insurance and consumption smoothing, being
life-cycle redistribution. Though the task of poverty relief is central, it is clear that the
welfare state exists also for much wider reasons.
the reduction of inequality, in contrast, is almost entirely an equity issue.
7. Vertical equity. The system should redistribute towards individuals or families with
lower incomes. This aim is contentious. All income-tested benets contribute to it;
Anti-poverty
benets
36%
Insurance
benets
21%
Life-cycle
benets
43%
Fig. 1.3. Spending on cash benets by type of benet
Source: UK DWP (2002a: table 7).
EOTC01 14/4/04 17:12 Page 11
12 CONCEPTS
so, secondly, do non-means-tested benets whose recipients disproportionately
have lower incomes (e.g. the UK at-rate pension). A third form of redistribution
arises where the benet formula favours lower-income individuals. Free provision
of a tax-funded service (e.g. health care in the UK) is also generally redistributive.
The success or otherwise of benets in reducing inequality is assessed by inspection
over time of aggregate inequality measures, though with all the caveats noted in
Chapter 6.
8. Horizontal equity. Differences in benets should take account of age, family size, etc.,
and differences in medical treatment should reect only factors that are regarded as
relevant (e.g. whether or not the patient has dependent relatives), but not irrelevant
factors like ethnic background.
social inclusion. So far the objectives have been conventional economic ones. Some
commentators also include broader goals.
9. Dignity. Cash benets and health care should be delivered so as to preserve indi-
vidual dignity and without unnecessary stigma. Beveridge emphasized the im-
portance of contributions in this context: The popularity of compulsory social
insurance today is established, and for good reason; by compulsory insurance . . .
the individual can feel assured that [his] needs will be met . . . by paying . . . a
contribution, he can feel that he is getting security not as a charity but as a right
(Beveridge Report 1942: para. 296).
10. Social solidarity. Cash benets and health care should foster social solidarity
a frequently stated goal in mainland Europe. So far as possible, benets should
depend on criteria that are unrelated to socio-economic status. Retirement pensions
are an example; so is medical care in many countries. Additionally, benets should
be high enough and health care and education good enough to allow recipients to
participate fully in the life of the society in which they livea broader aim than
that of relieving income poverty.
administrative feasibility has two aspects.
11. Intelligibility. The system should be simple, easy to understand, and as cheap to
administer as possible.
12. Absence of abuse. Benets should be as little open to abuse as possible.
problems of denition and measurement abound. Ebciency objectives 13 have precise
analytical denitions, but measurement problemsparticularly the incidence of taxes,
contributions, and benetsmake it dibcult to assess how far they are achieved. How
do we dene a poverty line in objective 4; and how large a drop in living standard is
unacceptable (objective 5)? The appropriate extent of vertical redistribution and a work-
able denition of horizontal equity (objectives 7 and 8) have occupied economists,
philosophers, and political theorists almost since the dawn of time, and have plagued
policy-makers at least since the British Poor Law Act of 1601. Even equality is dibcult to
dene unambiguously (Okun 1975: ch. 3; Le Grand 1982: ch. 2), especially in the context
of benets in kind like health care. Concepts such as dignity, stigma, and social solid-
arity (objectives 9 and 10) are hard to dene and raise major measurement problems.
EOTC01 14/4/04 17:12 Page 12
1. I NTRODUCTI ON 13
Writers like Hayek (1976) argue in addition that the term social solidarity is devoid of
meaning, and that its pursuit is both pointless and dangerous. These problems are dis-
cussed in some detail in Chapters 36.
Even were these problems assumed away, a second set of dibculties arises, in that some
objectives are inherently in conict and others may be. The trade-off between ebciency
and distributional objectives is no less intractable for its familiarity; the same is true of
the trade-off between horizontal equity and administrative simplicity. Other objectives
conict almost by denition. Consumption smoothing implies that an individual with
higher earnings should receive higher benets, which sits uneasily with the requirement
that benets should redistribute towards those with lower incomes, and with the objective
that benets should contribute to social solidarity. On one interpretation of equity every-
one should receive benets proportional to their past contributions, but that, again,
conicts both with redistribution towards lower incomes and with social solidarity. The
choice of objectives and of priorities between them is a fundamental normative issue.
3. A changing world: Challenges and responses
A number of longer-term trends have major implications for the design of the welfare
state and recur throughout the book.
demographic change. Life expectancy has increased in all industrial countries while birth
rates have declined, simultaneously increasing the number of older people and reducing
the number of younger workers. As a result, from about 2005 onwards, the ratio of people
over 60 to those of working age will increase sharply. If present policies continue
unchanged, spending on pensions and health care in some countries is set to double.
Policies to accommodate these changes are discussed in detail in Chapter 9.
globalization has at least two roots. First, since 1970 the international trade regime has
become increasingly open. Secondly, as a result of technological change, in Quahs terms,
economic activity is increasingly dematerializedthat is, it is in the form of encoded
binary bits (computer programs, music, videos) rather than solid, such as a Boeing 747.
One of many implications of this trend, according to Quah (1996: 7; see also Quah 2003),
is that international trade becomes not a matter of shipping wine and textiles . . . but
of bouncing bits off satellites. In these circumstances, national boundaries become
increasingly porous.
For both reasons, globalization reduces the ability of a country to act independently
in designing its institutions. Countries with expensive welfare states, it is argued, will
increasingly be at a competitive disadvantage relative to those with more parsimonious
ones. At the same time, however, demands on the welfare state are rising: there are more
old people, and in many countries rising numbers of unemployed; in addition, as dis-
cussed below, there are more lone-parent families, and increasing numbers of low-paid
and part-time workers.
changes in family structure have taken several forms. First, families have become more
uid. The institutions of the immediate post-war period assumed an archetypal nuclear
EOTC01 14/4/04 17:12 Page 13
14 CONCEPTS
family: the main (frequently the only) source of income was the wages of the husband;
and husband and wife stayed married, so that the husbands pension entitlement also
covered his wife. Though not wholly valid even then, the assumption was true enough
to form the basis of most social policy. Today, in contrast, many more marriages end
in divorce; and parenthood is less closely tied to marriage. These changes have major
implications for social policy, particularly so far as child support and pension arrange-
ments for women are concerned. A second set of changes arises from the increasing
number of women who have jobs outside the home.
changes in the structure of jobs. The nature of work is also changing. Post-industrial
employment tends to favour professional and highly skilled occupations. The demand for
unqualied workers is lower than in the past and, in consequence, their wages are low
and their employment often precarious and part-time. There are worries about increasing
polarization between a core of skilled workers and a peripheral workforce. Contributory
social insurance is of doubtful relevance to the latter group.
rising expenditure on the welfare state over past decades has been universal. Rising
income leads to demands for higher insurance benets and more extensive consumption
smoothing, and to the adoption of a poverty line that rises in real terms. A second cause
is technological advance: more advanced medical technology extends the range of what
is possible, fuelling the drive for higher spending on health care; and technological
advance is a major driver of mass tertiary education. Thirdly, population ageing will
increase spending on age-related benets, notably pensions, health care, and social care.
resulting challenges and responses. Two challenges stand out. A problemboth for eco-
nomic policy and for social policyis the possibility that the strategic design of the wel-
fare state is based, at least in part, on a past social order with stable, two-parent families,
with high levels of employment, and where most jobs were full-time and relatively stable.
Secondly, the conict between economic growth and equality has become sharper over
the years. The harmonious coexistence of full employment and income equalization that
dened the postwar epoch appears no longer possible (Esping-Andersen 1996a: 4). There
is a major debate, discussed in Chapter 15, Section 2.2, about why this is so.
Despite much public discussion of a crisis of the welfare state, change has been mainly
marginal. Esping-Andersen (1990, 1999, 2002 et al.) distinguishes three broad approaches
to economic and social change since the rst oil shock of the 1970s.
Under what Esping-Andersen calls the social-democratic approach (broadly that in the
Scandinavian countries, and particularly in Sweden), policy was aimed at increasing the
demand for labour through active labour-market policies and increased public-sector
employment. The problem with this approach was its cost. By the mid-1990s Sweden
faced major scal problems at a time of rising pressure on public jobs. Part of the response
was a move towards wage exibility; there was also reform to the state pension system.
The corporatist approachthat in the rest of mainland Europetried to reduce the
supply of labour, notably through early retirement. In many ways this is the Scandinavian
solution by a different route: instead of nding jobs, frequently in the public sector, for
people who would otherwise be unemployed, this approach tries to open up jobs by
offering early retirement, either explicitly or through the award of a disability pension.
EOTC01 14/4/04 17:12 Page 14
1. I NTRODUCTI ON 15
The cost in this case is not that of public employment but of public pensions. The
approach is coming under increasing scal pressure.
The neo-liberal approachbroadly the Anglo-Saxon model (the UK, the USA, Australia,
and New Zealand)sought to increase the demand for labour by liberalizing labour mar-
kets, not least through increased wage exibility. This approach has two advantages: it
avoids the heavy scal cost of the Scandinavian or mainland European arrangements;
and employment growth in the Anglo-Saxon countries over the 1980s was signicantly
higher than in the rest of the OECD. The besetting problem of the approach is rising
inequality and poverty, particularly among unskilled workers and single-parent house-
holds. As discussed in Chapter 2, Section 6, in the later 1990s reform in both the UK and
the USA rened this approach by combining labour-market exibility with policies of
activation based round benet structures that gave strong labour-market incentives,
together with policies aimed at improving peoples labour-market skills.
This book is primarily about what economic theory tells us about how to respond to
these challenges. The resulting debates are drawn together in the concluding part of the
nal chapter.
n QUESTIONS FOR FURTHER DISCUSSION
1. What is the welfare state?
2. What are the main purposes of the welfare state?
n FURTHER READING
On the welfare state generally, see Barr (2001a,b), the latter a collection of 100 articles about the
welfare state, and, for an introduction, see Glennerster (2003a).
Titmuss (1958) attempts to dene the welfare state; see also Esping-Anderson (1996b).
Esping-Andersen (1996b) also gives a wide-ranging overview of the challenges facing welfare
states across a broad range of countries, including the former Communist countries, Latin
America, and the newly industrializing countries of East Asia. For contrasting views on the debate
about the welfare state, see Lindbeck (1997), Atkinson (1999a,b), and Neil Gilbert (2002).
EOTC01 14/4/04 17:12 Page 15