Transcript The Index of Economic Well

The Index of Economic Well-Being
- 1984 - 2006
Lars Osberg
Department of Economics,
Dalhousie University
Halifax, Nova Scotia
Conference: “Les Indicateurs Locaux de
Progrès Sociétal"
Rennes, France, November 17, 2006.
Origin – a seemingly simple question:
"Are you better off today than you were
four years ago?"

1980 Ronald Reagan



1976-80 actual increase in per capita disposable
income in USA = 8.8%
Audiences answered “NO!” – WHY ?
1984 - Osberg Paper for MacDonald
Commission emphasized:


Widespread dissatisfaction with GDP as a measure
of Economic Well-Being and:
Alternative aggregate measures also sum to a
single index, burying value judgments
Components of:
Index of Economic Well-Being (IEWB)
Consumption flows
Economic
Well-Being
Economic security
Economic equality
Index of Economic Well-being, Equal Weighting, OECD, 2004
0.8000
0.7081
0.7000
0.6565
0.6576
Germany
Netherlands
0.6716
0.6748
0.6786
France
Belgium
Denmark
0.6305
0.6039
0.6000
0.5709
0.5429
0.4934
0.5000
0.4405
0.4477
0.4610
0.4000
0.3000
0.2000
0.1000
0.0000
United
Kingdom
Source: Table 1
Australia
Spain
United States
Canada
Italy
Sweden
Finland
Norway
2006:
‘Are “you” better off ?’

Who is “you” ?

Individual or Citizen ?


Personal well-being – no statistics needed
Statistics on ‘well-being’ are only
needed if the issue is social
decision-making

“Well-being” as “citizen” requires
information on collectivity

“Indicateurs Locaux de Progres Societal”
2006: Real Issue in Social Indicators
‘Is the community “better off” ?’

As voters or bureaucrats, individuals make
decisions re: collectivity
 Voting example: I will vote for policy X if
Ix = 1 (own utilityx) + 2 (society’s wellbeingx)
> other alternatives

Indicators of “Society’s Well-being”
 Needed for individual policy & voting decisions


Statistics = feedback loop of public policy
Economic Well-Being - multi-dimensional

Index should respect heterogeneity
 Values / Preferences
What is the point of Index construction?

Policy choices must be made



With multiple outcomes of differing dimensionality
Affecting many dissimilar individuals
Objective of index construction:


To assist democratic discourse by disentangling
 When values differ
 When factual judgments differ
To enable individuals to make better summative
subjective judgments on social choices
Dimensions of Economic Well
Being
Concept
Present
Representative
Agent /
“Typical Citizen”
Average Flow of
Current Income
Per Capita GDP
or “Adjusted”
Average Income
Flow
Issues: Market
transactions only,
heterogeneity,
stocks
Dimensions of Economic Well Being
Concept
Present
Representative
Average Flow of
Agent /
Current Income
“Typical Citizen”
Diversity of
Population
Experiences
Distribution of
Current Income
- Poverty and
Inequality
Social Welfare
Function literature
SWF = f ( ,  )
Dimensions of Economic Well Being
Concept
Present
Future
Representative
Agent /
“Typical Citizen”
Average Flow of Aggregate
Effective Current Accumulation of
Consumption
Productive
Stocks (broadly
defined)
Issues:
Average Income
does not reveal
savings rate
- assets include
environment,
Human Capital,
R&D, etc.
Aggregate Savings
– not automatically
optimal, sustainable
–preferences for
social saving differ
among individuals
Dimensions of Economic Well
Being
Concept
Present
Future
Representative
Agent/
“Typical Citizen”
Average Flow of Aggregate
Effective Current Accumulation of
Consumption
Productive
Stocks
Diversity of
Population
Experiences
Distribution of
Insecurity of
Current Income: Future Income
- Poverty and
Inequality
Heterogeneity in Values


ECONOMIC WELL-BEING=
1 CONSUMPTION
+ 2 SUSTAINABILITY / INTERGENERATIONAL
BEQUEST
+ 3 INCOME DISTRIBUTION / POVERTY
+ 4 SECURITY
DIFFERENT VALUES WILL IMPLY DIFFERENT WEIGHTS


Useful to know whether (& how much) perceived trend in
aggregate well-being depends on weighting
 = 0 is a (strong) value choice

GDP per capita


sets 3 = 4 = 0
assumes 1 AND 2 optimal always
What is Well-Being ?
What is Economic Well-Being?



Economic Well-Being < Well-Being
Economic Well-Being > GDP
Economic output > Marketed $ output


GDP omits many sources utility
 value household labor
 value of leisure
 length of life, etc.
GDP includes “regrettable expenditures”
 Costs of pollution, crime, commuting, etc
Human Well-being
- includes well-being from much more than economics
(e.g. personal freedoms, relationships, spiritual & intellectual
discovery)
Economic Well-being < Well-being
- but some aspects of well-being depend on
tradeoffs in scarce resources – ‘economic’
Economic
Well-being
Chart 2: Growth in the Index of Economic Well-being, OECD, 1980-2004 (percentage points)
0.7000
Level of the Index of Economic Well-being, Equal
Weighting, 1980
Percentage Point Change, 1980-2004
0.6000
0.5000
0.4000
0.3000
0.2000
0.1000
Ne
th
er
lan
ds
iu
m
Be
lg
ed
en
Sw
an
y
Ge
rm
ay
No
rw
nl
an
d
Fi
an
ce
Fr
ar
k
De
nm
ly
Ita
ng
do
m
Ki
a
da
ain
Sp
Au
str
ali
Un
ite
d
Source: Table 1.
Ca
na
Un
ite
d
St
ate
s
0.0000
Economic Well-being and GDP
marketed $ output < total goods & services
GDP
GDP
Economic
Well-being
“Social regrettables”
– part of GDP, but not well-being
“Social
regrettables”
GDP
GDP
Economic
Well-being
- Costs of crime,
pollution, commuting
GDP per capita

GDP rigorously standardized across countries
(SNA) – the clear point of comparison


But - Strong Implicit assumptions when
used as measure of economic well-being




Can one do better? Does it make any difference ?
aggregate share of income devoted to accumulation
(including value of unpriced environmental assets)
automatically optimal
poverty, inequality & economic insecurity do not matter
changes in leisure time, length of life, family size, costs
of commuting, pollution & crime - all irrelevant
+ poor match to popular perceptions of trends in
economic well-being
Payoff to per capita GDP growth in selfreported happiness ≈ nil
Chart 3: Average Annual Growth of the Overall Index of Economic Well-being and GDP per Capita,
OECD, 1980-2004 (per cent)
3.50
Overall Index of Economic Well-being
GDP per Capita (2000 US dollars)
3.00
2.50
2.00
1.50
1.00
0.50
s
St
ate
Un
ite
d
Ki
ng
do
m
ed
en
Sw
ain
Sp
ay
No
rw
lan
ds
ly
Ita
an
y
rm
Ge
an
ce
Fr
an
d
Fi
nl
ar
k
ad
a
nm
De
Ne
th
er
Un
ite
d
Source: Tables 1 and 2.
Ca
n
iu
m
Be
lg
Au
str
ali
a
0.00
Average Consumption Flows $

Marketed real consumption per capita


Adjustments
 value of increased longevity of life
 reduced economies of scale in household
consumption
 changes in working hours – leisure
Government services

provision of non-marketed or heavily subsidized
services
 includes defense and capital consumption
allowances
 excludes debt service charges and transfer
payments
Chart 4: Total Consumption Flows per Capita, OECD, 2004
(2000 constant US dollars)
35,000
31,971
30,000
25,000
25,175
25,397
25,568
25,632
Belgium
Australia
Italy
Netherlands
United
Kingdom
26,287
26,422
France
Norway
23,707
21,291
20,000
25,033
22,154
22,481
22,709
Sweden
Germany
Denmark
19,362
15,000
10,000
5,000
0
Finland
Source: Table 3.
Spain
Canada
United States
Wealth Stocks, Sustainability and
Intergenerational Bequest $


Physical capital stock from SNA
State of environment and national heritage
(degradation -)


Value of natural resource stocks


cost of CO2 emissions @ $ 85 per tonne
price + quantity change
Stocks of human capital

Evaluated at cost of schooling

Research and development capital stock
Net foreign indebtedness (-)

NOTE: Real productive assets only

Chart 5: Total per Capita Wealth, OECD, 2004 (2000 constant US dollars)
140,000
127,524
122,909
120,000
114,307
113,762
115,484
118,598
107,528
101,945
99,006
100,000
101,997
95,238
87,723
79,524
82,324
80,000
60,000
40,000
20,000
an
y
Ge
rm
ar
k
De
nm
ay
iu
m
No
rw
th
er
Be
lg
lan
ds
S.
an
ce
Fr
Ita
ly
an
d
nl
Fi
ad
a
Ca
n
Sw
ed
en
K.
U.
Ne
Source: Table 4
U.
a
Au
str
ali
Sp
ain
0
Income Distribution Index

How to summarize “Distribution”?



Inequality


Gini coefficient
 After-tax & transfer household income
familysize
 Equivalence scale =
Poverty


Simplicity desirable if index to be used
Poverty & Inequality differ, but both matter
Sen-Shorrocks-Thon measure
 Rate
 Average poverty gap ratio
 Intensity = rate x gap
Index = 0.75*Poverty + 0.25*Inequality
Chart 6: Index of Equality, OECD, 2004
0.9000
0.8287
0.8000
0.7000
0.6665
0.7186
0.7027
0.6894
0.7231
0.6344
0.6153
0.6000
0.4903
0.5000
0.3965
0.4000
0.3616
0.3000
0.2000
0.4056
0.2636
0.1783
0.1000
an
d
nl
Fi
ay
No
rw
iu
m
Be
lg
an
ce
Fr
Sw
ed
en
an
y
Ge
rm
ar
k
De
nm
lan
ds
th
er
Ca
n
Ita
ly
ain
Sp
K.
U.
a
ad
a
Ne
Source: Table 5.
Au
str
ali
U.
S.
0.0000
Universal Declaration of Human
Rights - 1948

[25] “Everyone has the right to a standard of
living adequate for the health and well
being of himself and of his family, including
food, clothing, housing and medical care
and necessary social services, and the right
to security in the event of unemployment,
sickness, disability, widowhood, old age or
other lack of livelihood in circumstances
beyond his control.”
“Economic Security”

Risk of income loss due to unemployment


Risk of financial loss due to illness


Uninsured medical expenses as % disposable income
Risk of single parent poverty



changes in employment rate x UI coverage x UI
replacement rate
poverty rate & gap for single women with children
divorce rate of legally married couples
Risk of poverty in old age

chance x depth of elderly poverty
“Economic Security”

Risk of loss due to unemployment


Financial Risk of Illness


Divorce rate x poverty rate x poverty gap of single parents
Risk of poverty in old age


Unreimbursed private medical expenses as share of
disposable income
Risk of single parent poverty


Risk of Unemployment + E(financial loss|unemployment)
chance x depth of elderly (>65) poverty
Security risks weighted by relevant population size
Security from Unemployment

Original method – financial loss implied by
compound probability
=P(U)*P(B|U)*(E(B/W)
Assumes components matter equally
 Decline UI/EI coverage has big impact on trends


New literature on self-reported happiness

Di Tella, MacCulloch, Oswald (2003) “The Macro Economics of Happiness” RESTAT



Ordered Probit life satisfaction – n= 271,224
Recover Implicit weights on Unemployment
Rate and Unemployment Benefits
This paper: Unemployment rate = 4x UIBen

= .8*(scaled Unemp) + .2*(scaled P(B|U)*(E(B/W))
Chart 7: Index of Economic Security, OECD, 2004
0.9000
0.7746
0.8000
0.7000
0.6523
0.7238
0.7080
0.6192
0.6146
0.6000
0.6809
0.6766
0.6642
0.6950
0.5642
0.5102
0.5000
0.4000
0.3134
0.3000
0.2030
0.2000
0.1000
ar
k
De
nm
an
d
nl
Fi
Sw
ed
en
ay
No
rw
an
ce
Fr
th
er
lan
ds
Ita
ly
iu
m
Be
lg
an
y
ain
Sp
ad
a
Ca
n
a
K.
Au
str
ali
Ge
rm
Ne
Source: Table 6.
U.
U.
S.
0.0000
Does it matter?
How different is trend in IEWB & GDP?

Trend in IEWB depends partly on
how heavily current consumption is
weighted compared to:




Sustainability / accumulation
Income Distribution
Security
Excel data sheet available for
experimentation @

http://www.csls.ca/iwb.asp
Figure 2a: The Index of Economic Well Being and its Components in the United
Kingdom, 1980-2001
19
80
19
81
19
82
19
83
19
84
19
85
19
86
19
87
19
88
19
89
19
90
19
91
19
92
19
93
19
94
19
95
19
96
19
97
19
98
19
99
20
00
20
01
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Consumption Flows
Wealth Stocks
Economic Security
Economic Well-being Index
Equality Measures
Policy Implications ?


Much less gain in economic well-being
than in real GDP per capita 1980-2004
Major reason has been growth in
inequality & insecurity



Reducing Inequality & Insecurity was the
major objective of the welfare state
BUT de-emphasized in recent years
Social Policy Design should aim at
increasing Well-Being
The role of the natural environment
Natural
Capital
EWB
GDP
Physical Investment
Produced
Capital
Natural
Capital
EWB
GDP
Human and Social Capabilities
Produced
capital
Natural Capital
Human and Social
Capabilities
GDP
EWB
The role of knowledge/skills
Produced
capital
Natural Capital
Human and Social
Capabilities
Human
capital
GDP
EWB
A DIGRESSION
Definitions


Human Capital
 “The knowledge, skills, competencies
and attributes embodied in individuals
which facilitate the creation of personal,
social and economic well-being
Social Capital
 “Networks together with shared norms,
values and understandings which
facilitate co-operation within or among
groups”
The role of networks/social norms
Produced
capital
Natural Capital
Human and Social
Capabilities
Human
capital
Social
capital
GDP
EWB
Close ties between human and social capital
Produced
capital
Natural Capital
Human and Social
Capabilities
Human
capital
Social
capital
EWB
GDP
The role of institutions
Produced
capital
Natural Capital
Human and Social
Capabilities
Human
capital
Social
capital
Political, institutional
and legal arrangements
GDP
EWB
Produced
capital
Natural Capital
Human and Social
Capabilities
Human
capital
Social
capital
Political, institutional
and legal arrangements
GDP
EWB
Produced
capital
Natural Capital
Human and Social
Capabilities
Human
capital
Social
capital
Political, institutional
and legal arrangements
GDP
EWB
In both 1984 & 2006
– why do we care if indicator goes ‘up’ ?

Standard Indicators have ambiguous
relation to Well-being


GDP per capita excludes leisure,
environment & more
Hourly wages ? Employment ?

Not valued directly – but indicate a more
fundamental objective


Wage = price of labour;
 potential consumption? Market ‘power’?
Unemployment = unused labour;
 insecurity? Social exclusion ?
Methodology

Variables now scaled linearly




Consistent with other indices (e.g. HDI)
Solves “Directionality Problem”
 (Max – value)/(Max – Min)
 OR (Value – Min)/(Max-Min)
Problems:
 Reporting trends as % change or % points
 Scaling removes base – sensitive to comparison
group
“Base Case” assigns equal weight to all
dimensions

Excel data sheet available for experimentation
 http://www.csls.ca/iwb.asp