Inclusive growth in
Asia and the Pacific
Findings of the ESCAP Survey 2015
OECD/ESCAP/ADB REGIONAL CONSULTATION
Inclusive Growth in Southeast Asia
Oliver Paddison
United Nations Economic and Social Commission for Asia
and the Pacific
Has growth been inclusive?
• Inclusiveness is typically measured using income-related
indicators.
Country
Azerbaijan
Bangladesh
Cambodia
China
India
Indonesia
Kazakhstan
Malaysia
Pakistan
Philippines
Russian Federation
Sri Lanka
Thailand
Turkey
Growth of absolute
poverty (1990-2013)
-97.4%
-38.4%
-58.2%
-80.4%
-33.8%
-70.1%
-97.4%
-100.0%
-67.5%
-40.0%
-100.0%
-72.6%
-96.7%
-36.2%
2
Growth real pc
income (1990-2013)
95%
132%
193%
642%
189%
115%
77%
122%
54%
58%
22%
182%
119%
74%
Change in Gini
(1990-2013)
-1.3
4.5
-2.3
9.6
3.1
9.0
-3.6
-1.4
-3.2
-0.8
-8.3
3.9
-5.9
-1.5
Realizing inclusive growth
• Inclusiveness is a multidimensional concept.
• It should capture social and environment dimensions of
development (Rio+20).
• Inclusiveness is broadly defined in terms of:
(a) increasing the average standard of living of the
population;
(b) reducing income inequality;
(c) reducing levels of extreme poverty; and
(d) expanding and broadening equality in opportunities
(social and environment related).
3
Methodology
• Create composite indices for economic, social and
environmental inclusiveness.
– Select relevant indicators per index (5), using only outcome
indicators.
– Compute average for relevant time period (1990s and 20002012) and linearly re-scale in interval [0,1], with one
indicating best score in Asia-Pacific region.
– Compute arithmetic averages of indicators per index,
assigning equal weights.
4
Economic inclusiveness
Significant differences in poverty rates
between urban and rural sectors.
Income inequality has increased in many
countries.
Lack of productive employment
employment
5
40
Poverty headcount ratio, in percentage of population
Measured by:
1.Rate of poverty at $1.25 per day in 2005
PPP
2.Income inequality: Gini coefficient
3.Ratio of incomes of the highest quintile
to the incomes of the lowest quintile;
4.Unemployment rate; and
5.Ratio of the female-to-male
labour-force participation rate.
Rural
Urban
35
30
25
20
15
10
5
0
China
India
Indonesia
Economic inclusiveness
6
Social inclusiveness
Secondary school attendance
100
90
80
70
60
Percentage
Measured by:
1. Gender parity at the secondary
school level;
2. Gross secondary school
enrolment;
3. Average years of schooling;
4. Percentage of live births attended
by skilled health staff; and
5. Mortality rate of children under
age 5
50
40
30
Significant progress has been made.
20
10
Males Urban
Females Urban
7
Males Rural
Females Rural
Viet Nam
Pakistan
Nepal
Indonesia
India
Cambodia
Bangladesh
0
Azerbaijan
Yet, large disparities in education and
health remain within countries.
Social inclusiveness
8
Environmental inclusiveness
Measured by:
1. Access to improved sanitation
2. Access to water sources;
3. Annual change in total greenhouse
gas (GHG) emissions
4. Annual change in forest area;
5. Annual change in share of fossilfuel energy consumption in total
consumption of energy.
Access to electricity, 2012
Sri Lanka
Mongolia
Nepal
Lao PDR
Indonesia
Pakistan
Philippines
Bangladesh
The poor are particularly affected by
environmental degradation.
Myanmar
Cambodia
DPRK
Environmental degradation can also be
an outcome of economic inequality.
9
0%
20%
40%
Rural electrification rate (%)
60%
80%
100%
Urban electrification rate (%)
Environmental inclusiveness
10
ESCAP Inclusiveness Index
11
Policy recommendations
1.
Address the neglect of the rural sector.
• Increase agricultural productivity by focusing on quality and standards,
investments in R&D.
• Develop non-farm sector through rural industrialization.
2.
Strengthen financial development, foster financial inclusion.
3.
Foster creation of small and medium-sized enterprises.
4.
Make existing expenditure more development-oriented:
•
•
•
•
Reduce non-development expenditure (defence, energy subsidies).
Increase access to and the affordability of health systems.
Strengthen social protection programmes.
Expand investment in education.
12
Thank You!
www.unescap.org/publications/
economic-and-social-survey-asia-pacific
twitter.com/unescap
facebook.com/unescap
youtube.com/unescap
13
Has growth been inclusive?
• Inclusiveness is typically measured using income-related
indicators.
Country
Growth of absolute Growth real pc
poverty (1990-2013) income (1990-2013)
Change in Gini
(1990-2013)
Rank of
inclusiveness
Azerbaijan
-97.4%
95%
-1.3
Bangladesh
Cambodia
China
India
Indonesia
Kazakhstan
Malaysia
Pakistan
Philippines
Russian
Federation
Sri Lanka
Thailand
Turkey
-38.4%
-58.2%
-80.4%
-33.8%
-70.1%
-97.4%
-100.0%
-67.5%
-40.0%
-100.0%
132%
193%
642%
189%
115%
77%
122%
54%
58%
22%
4.5
-2.3
9.6
3.1
9.0
-3.6
-1.4
-3.2
-0.8
-8.3
12=>12
14=>15
10=>7
13=>14
11=>11
1=>1
4=>4
16=>16
7=>10
2=>2
-72.6%
-96.7%
-36.2%
182%
119%
74%
3.9
-5.9
-1.5
6=>5
3=>3
14
THE OECD INCLUSIVE GROWTH
FRAMEWORK
Paul Schreyer
Deputy Director OECD Statistics Directorate
OECD/ESCAP/ADB Regional Consultation on Inclusive
Growth in Southeast Asia
Bangkok, 9 June 2015
The issue
 Ensure that growth goes hand-in hand with
improvements in people’s living conditions
 Policies need to target multiple objectives
simultaneously, not just GDP
 Need new metrics
• Aspects beyond income
• Distribution
 Need to revisit our models
• Integrate multidimensionality and interactions
16
Defining Inclusive Growth
• A 3-pronged approach:
– Which growth? -> Multidimensional
– Whose growth? -> Distributions
– What drivers? -> Policy relevance
Which growth? OECD How’s Life?
framework
Housing
Income
Work-Life Balance
Jobs
Education and skills
Social Connections
Civil Engagement and Governance
Environmental Quality
Personal Security
Subjective Well-being
Health
Whose growth?
• Measuring evolution of income, health,
employment of particular parts of the
population:
• Median households
• Bottom 10%
• Being able to assess the net effect of
policies on these variables
• Drawing conclusions for governance,
institutions and policy design
19
How do we measure?
• For assessment of net effects of policies, we need common
units
• Translate changes in health or jobs into income
equivalents
• Econometric techniques, well-researched
• Life assessment = f(income, health, jobs)
• Valuation in money terms
1 year of life expectancy = 5% of income
1 point of unemployment = 2% of income
• Weights are conservative and standard and representative
of peoples (implicit) preferences
• Measure of multi-dimensional living standards
20
Average growth in MLS 1995 and 2012
Inequality
Unemployment
%
Longevity
Income
Inclusive growth
Economic growth
USA – AUS: similar GDP/cap and real HH
income growth
But unemployment declines in AUS, life
expectancy rises and inequality effects are
small
 Growth in living standards AUS>USA
10
8
6
Average OECD MLS
4
2
0
Note: OECD calculations based on OECD National
Accounts, Health and Income Distribution databases.
CHN
CHN urb
CHN rur
FIN
NOR
AUS
NZL
HUN
CAN
CZE
FRA
SWE
BEL
NLD
DEU
DNK
AUT
USA
PRT
ITA
-2
Time profiles of MDLS
$, PPP
adjusted
… and 2012 MLS levels
Inequality
Unemployment
Longevity
Income
Living standards
USD per capita
USA higher
income levels
35000
than
AUS
But overcompensated by
25000
differences
in standard
LE and
Living
of the median household (OECD average)
inequality
15000
5000
-5000
-15000
Note: OECD calculations based on OECD National
Accounts, Health and Income Distribution databases.
LUX
NOR
CHE
AUS
AUT
CAN
DEU
SWE
USA
FRA
NLD
BEL
JPN
GBR
FIN
DNK
NZL
ITA
IRL
KOR
SVN
ESP
CZE
PRT
GRC
POL
SVK
EST
CHN urb
HUN
CHL
MEX
CHN
CHN rur
-25000
Quantifying policy transmission:
example GDP and household income
 Long experience about policy effects on GDP per
capita
 But much less on HH income
 GDP growth has trickled down less since the mid1980s.
 The gap may reflect differential impact of pro-growth
policies on household disposable incomes
 Different effects for different social groups along the
distribution of income.
Gains in GDP have not fully trickled down to
household incomes
(on average since the mid-1980s)
A. Household disposable incomes elasticities to GDP
1
Average income
0.95
0.9
0.85
0.8
0.75
0.7
0.65
0.6
0.55
0.5
-12 -11 -10 -9
-8
-7
-6
-5
-4
-3
-2
-1
0
1
2
3
4
5
6
7
8
9
10 11 12
Bottom to top-sensitive income standards
The elasticity of household disposable income to GDP per capita has
been even lower at the bottom end of the distribution pointing to growing
inequality.
Source: Causa, de Serres and Ruiz (2014)
Reforms can have a differential impact on wage
dispersion and employment
Effect of change on:
A pro-growth change in:
Wage dispersion
Employment
Overall earnings
inequality
Technical progress (Higher MFP)
+
=
+
Higher R&D intensity
+
=
+
Deeper trade integration
=
=
=
Higher FDI openness
=
=
=
-
+
-
+
+
=
Innovation and Technology
Globalisation
Education / Human capital
Higher share of skilled workers
Product market competition
Lowering regulatory barriers to
entry
Source: Going for Growth 2015, Chapter 2
Work ahead
• Further quantifying policy effects on health
and jobs and computing net effects
• Health and Unemployment inequalities:
monetisation allows combining with income
inequalities
• Adding education -> ‘welfare return to
education’ as opposed to income return to
education
27
Thank you!
Website: http://www.oecd.org/inclusive-growth/
Email: paul.schreyer@oecd.org
ADB Approach to Inclusive Growth
Juzhong Zhuang
Deputy Chief Economist
Economic Research and Regional Cooperation Department
Asian Development Bank
Presentation at OECD/ESCAP/ADB Regional Consultation on Inclusive Growth in Southeast Asia, Bangkok
9 June 2015
29
Asia’s high growth has led to large
reductions in poverty …
GDP growth and poverty reduction
35
30
25
20
15
10
5
0
32.4
7.0
Developing Asia
9.7
3.7
3.4
5.7
Sub-Saharan Africa Latin America and
Caribbean
2.4
1.5
Middle East and
North Africa
Annual GDP growth (1990-2010), %
Cumulative reduction in poverty rate (1990s-2000s), percentage point
30
…but has been accompanied by rising
inequality in many countries
Gini Coefficients, Selected Economies, 1990s and 2000s
Singapore
Georgia
PRC
Indonesia
India
Lao PDR
Mongolia
Sri Lanka
Taipei,China
Bangladesh
Tajikistan
Korea, Rep. of
37.1
32.4
30.4
33.2
32.5
31.2
27.6
24.5
0
10
20
1990s
42.1
42.1
29.2
32.5
44.2
46.3
38.9
37.0
36.7
36.5
36.4
33.8
32.1
29.0
30.8
28.9
30
40
50
2000s
31
Why does inequality matter?
• Rising inequality slows down the pace of poverty reduction
– If inequality had been stable, additional 240 million Asians
(6.5% of Asia’s population) would have been lifted out of
poverty
• Inequality can weaken the basis of growth by affecting human
capital, social cohesion, middle class, and quality of
governance
– Empirical studies show lower inequality is associated with
longer growth duration. A 10-percentile decrease in
inequality increases the expected length of a growth spell
by 50% (IMF 2011)
32
Why has inequality risen?
• Technological progress, globalization, and marketoriented reform have led to rapid growth in Asia, but
working together they have favored:
– capital over labor
– skilled over unskilled workers
– cities/coastal regions over rural/inland areas.
• These have been compounded by unequal access to
opportunity due to social exclusion.
• Rising income inequality increases wealth inequality,
which in turn contributes to rising income inequality.
33
Share of labor income declined while share of
capital income increased
Labor Income Share
Manufacturing
70
60
50
40
30
20
10
0
Early 1990s
Mid-1990s
Early 2000s
Mid-2000s
34
Skill premium has risen; education inequality
accounts for 25–35% of total inequality
35.7
30.8
25.0
23.2
24.7
2002
2008
29.8
2010
29.9
26.5
30
20.3
20
10
46.2
2005
40
44.2
1995
50
8.1
PRC
India
Indonesia
Pakistan
2009
1994
1990
2009-10
1993
2007
0
1995
Share of between-group inequality, %
Income inequality decomposition by educational attainment of
household head
Philippines
Thailand
35
Spatial inequality—urban-rural and inter-province
combined—accounts for a large share of total inequality
Share of spatial inequality (%)
60
54
50
40
32
30
35
38
26
21
22
20
13
10
0
Sri Lanka Philippines Pakistan
(2009)
(2009)
(2008)
Indonesia
(2009)
India
(2008)
Viet Nam
(2008)
Bhutan
(2007)
PRC
(2007)
36
How to respond to rising inequality?
• The three drivers of growth should be promoted.
• Governments can address rising inequality through
– Growth that is more employment friendly to increase
the labor income share
– Efficient fiscal measures to reduce inequality in human
capital, supported by effective and fair tax systems
– Interventions to reduce spatial inequality, including
both urban-rural income gaps and regional disparity
– Governance reform to equalize opportunities.
37
More broadly, move toward inclusive
growth
• Inclusive growth means everyone can participate in and
benefit from the growth process.
• Inclusive growth makes a distinction between
– Inequality due to differences in individual efforts, and
– Inequality due to differences in individual circumstance (ethnic
background, gender, parental education, location, etc.), that is,
inequality in opportunity.
• Reducing or eliminating inequality in opportunity is at the
heart of an inclusive growth strategy:
– Inclusive growth is “growth coupled with equality of
opportunities”.
38
39
21
Thank you!
For more details, contact
jzhuang@adb.org
40