MINISTRY OF FINANCE

GOVERNMENT OF INDIA

State of the States:

Today and Over Time
Module 8
Contemporary Themes in India’s Economic
Development and the Economic Survey

Arvind Subramanian
Chief Economic Advisor

1
Overview
• State of the States

• Is there an Aid or Resource Curse in India?

• India’s distinctive demographic dividend

2
 Section 1
State of the States

3
Overview

• Indian States: Getting to Know Them

• Models of Development among Indian States
• Regional Inequality Over Time: Convergence and
Divergence
• A Puzzle
4
 Models of Development among Indian States
“Please understand, Your Lower GSDP Higher GSDP

Excellency that India is two

countries: an India of Light,
and an India of Darkness.
The ocean brings light to my
country. Every place on the
map of India near the ocean
is well-off. But the river
brings darkness to India.”

“The White Tiger” by Aravind Adiga

5
How rich are the Indian states?
300
Per Capita Gross State Domestic Product in 2014

250
Thousands of INR (2011-12 prices)

Urban Poverty Line (Tendulkar; Updated)

200

150

100

50

-
GA

KL

GJ

JK

OR
JH
SK

KA
PB
AP

WB

AS
AR

BR
MZ

MG

MP
DL

MH

MN
UK

HP
HR

IND

CG
RJ
NA

UP
TN

TR
6
How do Indian states compare with China?
Thousands of PPP dollars (2011 prices)
20
Per Capita Income in 2014
18
16
14
Poverty Line (World Bank Definition)
12
10
8
6
4
2
0
GA

TN
GJ

WB

TR
OR
KA

MZ

JH

MP
MN
SK

MH
UK

KL
HP

PB

AR
IND

MG
JK

AS

BR
CHN

APC

CG
DL

RJ
NA

UP
HR

7
How many poor are there in the Indian states?

40
Head Count Ratio as of 2011
35
HCR (as % of population)

30

25
21.9
20

15

10

0
OR

KL
GA
AR
BR

IND
AS

KA

GJ

AP
PB
SK
CG

MN

MP

MZ
WB

MH

TR
MG
JH

UP

NL

RJ

TN
UK
HR
JK
DL

HP
Source: Rangarajan Committee Report, 2011, Tendulkar Methodology 8
What is the overall well-being of people in the
Indian states?
0.75
0.727 Human Development Index Scores as of 2015*

0.7
China HDI Ranking: 90/188
India HDI Ranking: 130/188
0.65
HDI Score

0.609
0.6

0.55

0.5 JK

GJ

OR
CHN

PB

KA

AP

WB

AS
KL

BR
MP
MH

UP
HP

HR

IND

RJ
TN

Source: “Why Kerala is like Maldives and Uttar Pradesh, Pakistan”, Livemint, 2015 9
How many infants die within 1 year of birth in
the Indian states?
Infant Mortality Rate (per 1,000 live births) as of 2014
60

50

40
IMR

30

20

10

0
MP AS OR UP RJ CG BR AP IN HR GJ TG JK JH UK HP KA WB PB MH DL TN KL
Source: Sample Registration Survey data 10
 % Children enrolled in Std V who can read Std II Level
Text

0
10
20
40
50
60
70

30
MN
HP
KL
PB
HR
UK
MH
CG
AP
RJ
GJ
OR
TR
NL
WB
MG
IND
47.8

TG

Source: ASER 2016 Report

MZ
TN
UP
KA
Reading Outcomes at Primary Level as of 2016

BR
MP
the Indian states?

AS
JH
AR

% Children enrolled in Std V who can do division

0
10
20
30
50

40
60

HP
MN
HR
PB
KL
AP
UK
BR
TG
WB
RJ
MZ
OR
IND
25.9

JH
CG
UP
TN
NL
How well do children in schools learn in

MH
TR
Arithmetic Outcomes at Primary Level as of 2016

KA
11

MP
AR
GJ
AS
MG
 Overall, human development and income are
highly correlated: J&K and Kerala are outliers
0.75
Human Development Index (in 2015)

KL
0.7

PB TN HP MH HR
0.65 JK

KA GJ
0.6 WB

RJ

0.55 MP
AS OR
BR UP

0.5
22 42 62 82 102 122 142
Per Capita Income in 2014 in Thousands of INR (2011-12 prices)
12
Data available only for 16 states
Headcount Ratio and Incomes

45
Head Count Ratio(% of population) 2011

40 CG
MN JH
35 AR
BR OR
AS
MP
30 UP
25
KA CH
20 MZ
WB
MH
GJ
15 NA
TR RJ
MG TN HR
10 JK AP UK DL
PB HP SK
KL
5 GA

0
10 10.5 11 11.5 12 12.5
Per Capita GSDP (log) 2014
13
Are IMR and TFR correlated?
60
y = 14.4x + 3.1
R² = 0.47
MP
50 OR AS UP
Infant Mortality Rate (2014)

RJ
CG BR
40 AP HR
JK UK GJ JH
HP
30 KA
WB
PB
MH
20 TN
DL
KL
10

0
1.5 1.7 1.9 2.1 2.3 2.5 2.7 2.9 3.1 3.3 3.5
Total Fertility Rate (2014)
14
TFR or IMR data not available for few states. They have not been kept in the figure
 Models of Development among Indian States
• Peninsular/Coastal models:
• Manufacturing (Gujarat, Tamil Nadu)
• Remittance-Reliance and Human Development (Kerala)
• Skilled Services (Karnataka, Andhra Pradesh, Tamil Nadu)
• Hinterland models:
• Agriculture (Madhya Pradesh, Punjab, Haryana)
• Natural resources* (Jharkhand, Chhattisgarh, Odisha & Rajasthan)
• “Aid” or Special Status (North-Eastern States, J&K)
• Policy-induced manufacturing (Himachal Pradesh, Uttarakhand)

15
* Coal, Onshore Oil and Natural Gas, Major and Minor minerals
Manufacturing Model of Development: Higher share of
manufacturing associated with greater income
40%
UK
35%

Manufacturing as % of NSDP (2014-15)

30%

25% MG GJ
JH
HP
20% TN PU
MH
HR
15% OR CG
KA
TG
PB
10% UP AS JK RJ KL
AP
MP DL
5%
BR CH
NL MZ AR ANI
0%
10 10.5 11 11.5 12 12.5
Log of Per Capita Income (2014-15)
16
Natural Resources Model of Underdevelopment:
Higher share of natural resources associated with lower
income
16%
ANI
14%

Mining as % NSDP (2014-15)

12%
JH CG
10%
OR
8%
AS
RJ
6%

4% TG
MH
MP MG AP GJ
AR UK
2% DL
BR UP PU
JK KA
0%
10 10.5 11 11.5 12 12.5
Log of Per Capita Income (2014-15)

States with < 1% mining share and Goa have been dropped 17
Indian “Aid” Model of Underdevelopment:
Higher share of “aid” associated with lower income
40

Per Capita Redistributive Resource Transfers in

2015 in Thousands of INR (Current Prices)
35 MZ AR
30 NL

25 MN
MG
20 TR
15
JK
10
AS HP
5 UK
BR JH OR CG
UP MP
0 RJWB
TN KL
-5 AP KA
PB GJ
MH
HR
-10
0 50 100 150 200
Per Capita Income in 2014 in Thousands of INR (Current prices)

18
International Remittances Model: Higher share of
remittances associated with greater income
States with High International Remittances 60

Total remittances (in Thousands of INR Cr.)

2007-08
50

40

30

20

10

0
KL MH TN PB AP DL GJ KA UP RJ GA WB

Source: Chinmay Tumbe: Remittances in India: Facts & Issues

19
Regional Inequality Over Time:
Convergence and Divergence

20
Growth in most recent decade much higher
than growth in previous two decades
8 GJTN
Compounded Annual Average Growth Rate of

NA GA
7 MH
APC TR
Per Capita Income (2004-2014 period)

DL HR KL
HP
KA
6 MPC
BRC OR RJ WB
UPC PB
5
AR
MG
4
AS
JK MN
3

0
0 1 2 3 4 5 6 7 8
Compounded Annual Average Growth Rate of Per Capita Income (1984-2004 period)

* Combined numbers for Bihar, Madhya Pradesh, Uttar Pradesh, Andhra Pradesh have been used 21
 Income divergence “Big Time” in India (increase in
regional inequality) for the last 3 decades
6
9
TR GA

Average Growth Rate of Per Capita GSDP

5 8 GJ
Average Growth Rate of Per Capita GSDP

TN
HP
KL NA GA
7 TR MH
APC KL
APC R MH
ATN HPHR DL

between 2004 - 2014 (%)

4 GJKA HR KA
WB
between 1984-2004 (%)

6 BRC MPC RJ
OR WB
RJ DL UPC PB
MG 5
3 PB AR
OR MG
MPC 4
NAMN AS
2 UPC MN JK
3

AS JK
BRC 2
1
1

0 0
9.25 9.75 10.25 10.75 11.25 9.5 10 10.5 11 11.5 12
Log of Per Capita Income in 1984 (2011-12 prices) Log of Per Capita Income in 2004 (2011-12 prices)

* Combined States 22
Income Convergence Observed in World and
within China, But Divergence within India
18%

16%
Average Growth Rate of Real GDP Per

14% World China India

12%

10%
Capita

8%

6%

4%

2%

0%
6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0
Real GDP Per Capita in PPP (log) in 2004 23
India and most Indian states have higher
Infant Mortality Rate(IMR) for their levels of income

(2014)

(2014) 24
India and Indian states have lower
Total Fertility Rate (TFR) for their level of income

(2014)

(2014) 25
A Puzzle
• Evidence of high internal integration and porous borders: internal trade
and migration are higher than ever before

• Income and consumption divergence despite these equalizing forces

• Is it because governance or institutional traps keep real returns low even

where capital is low?

• Is it because of Indian development patterns that favour skill-intensive

sectors over low skill sectors?

• Still puzzling given dynamic of competitive federalism and competition

between states
26
 Section 2
Is there an Aid or Resource Curse in India?

27
Overview
• Aid and Natural Resources: Manna from Heaven?
• Redistributive Resource Transfers (RRTs): Aid in India
• The RRT curse?
• How to make RRT’s more effective?

28
Aid and Natural Resources:
Manna from Heaven
• Old view: Developing countries poor as they lacked
capital. People too poor to save.
• Possible solutions?
• Foreign Aid

• Natural Resources

29
 But is it really Manna from Heaven?!
• No robust positive relationship found between aid/natural resources
and economic growth*
• Why?
• Aid perpetuates resource dependency. As revenues flow in,
recipient countries may fail to develop internal tax base or
strengthen institutions which are key to long-run development
• Dutch Disease: As spending increases, prices rise, economy
becomes uncompetitive and manufacturing suffers

*(Easterly (2000), Rajan & Subramanian (2007), Sachs and Warner (1995, 1999), Sala-i-Martin and Subramanian
(2003) and Ross (2014)) 30
The Indian Experience of “Aid”: Gross Devolutions

• “Aid” within India exists: Centre provides transfers to certain poorer

states
• Transferred funds referred to as “Gross Devolution” in the form of:
1. a share of central taxes, as stipulated by Finance Commissions;
2. plan and non plan grants; and
3. plan and non-plan loans and advances

31
Measurement of Redistributive Resource Transfers
(RRTs)
• Redistributive Resource Transfers defined as gross devolution to the state adjusted for
the respective state’s share in aggregate Gross Domestic Product*

• RRTi=GDi-Si*GD where GD=ΣGDi

• Redistributed resources from the Centre differ from traditional “aid” in two important
aspects.
a) These are intra-country transfers and do not augment overall national disposable income like
foreign aid does;
b) The donor-recipient relationship is very different because states benefiting from transfers are part
of national governance structures that have determined them.

32
 Which are the highest RRT recipients?
70

60
Per Capita RRT Transfer in 2014-15

50
(in thousands of INR)

40
Rural Poverty Line
Urban Poverty
30 Line (2015-2016)
(2015-2016)

20

10

-10
SK AR MZ NL MN MG TR JK HP AS UK OR BR MP CG JH UP RJ WB AP KA PB TN KL GJ MH HR GA

33
 Is there an Aid and Natural resource curse in India?
Figure 4b: Manufacturing
Figure
Manufacturing share
4c: Fiscal share
effort &&
and RRT
RRT
RRT Figure 4d: Governance
Governance and RRT
& RRT

100
40

GA KR
TN UK

Governance index (higher the better)

8

KE GA AP HP
GA KE
GJ
PB
HRPBAP CH
MH MH
30

80
GJ KR UK
MP TN
GJ RJCH
Manuf. share (%)

RJUP
UK HP MPAS MZ
JH HP HR WB UP
OD JK
MG MGTR
20

60
TN
MH JH
WB BHAS JH
OD
HR CH
4

MZ OD
KR BH
AR PB RJ MG
TR
APUP JK
KE MPAS
10

40
JK MN
2

MN NG AR
BH TR NG ARMZ
MN
NG ARMZ

20
0

000 -3000-3000 30003000 9000

9000 15000
15000 -3000 3000 9000 15000
RRTRRT
per per
capita (Rs.(Rs.
average) RRT per capita (Rs. average)
capita average)

34
Is there an Aid and Natural resource curse in India?
• Analysis shows that higher the Aid:
• Slower is the growth
• Smaller is the share of manufacturing in GSDP
• Lower is own tax revenues
• Poorer performance in overall governance (ATC losses)
• This is not necessarily true for natural resources:
• Mineral wealth neither a boon nor a bane
• Post bifurcation, new states have managed natural resources less efficiently
• Neither the bounty of aid or natural resources seems to increase
consumption
35
How to obviate the Aid and Natural Resource curse?
• Redirecting flows directly to households: Channel a quantum of
RRTs/gains from resources directly to households as part of a
Universal Basic Income (UBI) scheme
• Performance-contingent transfers: Offset the fiscal bias uncovered by
the above analysis, in which states to relax their own tax effort.
Perhaps revert to conditioning transfers on revenue performance of
states (12th FC)
• Governance-contingent transfers: Fund transfer mechanism could
explicitly include a few easily monitored institutional indicators.

36
 Section 3
India’s Distinctive Demographic Dividend

37
What is the demographic dividend?
• The demographic dividend refers to the increase in a country’s
growth potential due to having a young population: a large
share of WA population appear brings economic dynamism

• Younger populations are more productive and save more 

larger fiscal base because of higher economic growth and fewer
dependents

• However, the real demographic sweet spot arises from not only
from a large WA/NWA ratio but also a growing one 38
Overview

• A Unique Demographic Profile

• A Rapidly Closing Demographic Sweet Spot

• Demographic Heterogeneity: An Equalizing Force

39
India’s demographic profile is distinctive:
Peaks later, at lower levels, but lasts longer
Ratio of Working age to Non-Working Age Population

40
India’s demographic sweet spot is rapidly
closing

Additional average annual growth

due to the demographic dividend
(in percentage points)
Decade Additional growth
2001-10 1.44
2011-20 2.62
2021-30 1.81
2031-40 1.92
2041-50 1.37
41
India’s demographic profile is highly
heterogeneous: Soon-to-age South vs. Young
Hinterland India
Ratio of Working age to Non-Working Age Population

42
Two demographic India’s offering possibilities
of mutual help through migration

43
Demographic heterogeneity is an equalizing force:
Dividend greater in currently poorer states (with exceptions)
Per Capita Income in 2011 and the Demographic
Dividend (2011-31)
2

Additional annual per cpaita income growth

BIH J&K
1.5

UP
relative to 2011
1 JHA
ASM
MP RAJ KAR HAR
0.5 WB MAH
ODI CHH
0 ANP PNJ GUJ
TNA

-0.5
KL

-1
0 20 40 60 80
Real per capita income in 2011 (Thousands of INR) 44
Instrumental Variables

45
How do we determine causality? Instrumental
Variables

https://xkcd.com/552/
46
Instrumental Variables
• Separate out the part of RRTs that is unrelated to economic
outcomes (growth, manufacturing share, fiscal effort) or
governance.

• Identify an instrumental variable (IV) for the explanatory

variable (i.e. RRT), which is strongly correlated with RRT
but not with outcomes or governance.

47
Instrumental Variables
Two IVs are proposed:
(i) the distance of the state capital from New
Delhi.

(ii) the distance of the state capital from the

nearest international border.

48
 Is proximity of states to foreign countries or
proximity to New Delhi a good proxy for recovering
RRTs?
Figure A2:A2:
Figure RRT and
RRT distance
and distance Figure A3:A3:
Figure RRT and
RRT distance
and from
distance New
from Delhi
New Delhi
from RRT and distance from
international
from international international border
border
border RRT and distance from New Delhi
10000

10000

15000

15000
AR AR MZ MZ
MZ MZ

RRT per capita (Rs. average)

5000

10000
RRT per capita (Rs. average)
5000

10000
AR AR
NGJK NGJK
HP HP
GJ GJWB WB OD OD NG NG

5000

5000
MN MN UK UKAP AP MN MN
MH MH
0

MG MG UP RJUP RJ
TR TR BH TR TR
GABH
PB PB
GA MP TN
MP TN JK OD
JK OD KE KE MG MG
HR JH
HR JH TN TN
AP AP
CH CH WB WB

0
HP HP
-5000

GA
-5000

AS AS KE KE MH GA
MH
CH CH GJ
MP MP BHJHBH GJ
JH KR AS AS
UK KR
UPUKUP

-5000
RJ RJ

-5000
KR KR PB PB
HR HR
-10000

-10000

-500 -500 0 0 500 500 1000 1000 -1000-1000 -500 -500 0 0 500 500 1000 1000
Distance from international
Distance borderborder
from international controlling for being
controlling landlocked
for being landlocked Distance from New
Distance from Delhi controlling
New Delhi for being
controlling landlocked
for being landlocked

49
 Do the Effects change?
-5000 0
RRT per capita controlling for being landlocked
5000 -5000 0
RRT per capita controlling for being landlocked
5000

r- Manufacturing
Figure
Figure 5b: share
5c:Manufacturing
Fiscal (%)and
effort and per-capita
per-capita
share (%) and RRT
RTT
per- FigureGovernance
5d: Governance index
index and and per-capita
per-capita RRT RTT
controlling
capita controlling for landlocked
for landlocked
RRT controlling nature
nature nature
for landlocked controlling
controllingfor
forlandlocked nature
landlocked nature
30

R KR HP
4

20
PB

Governance index (higher the better)

UK UK KR
RJ CH

10
20

HRPB
CH GA AP GA
2

MP TN MP KE AS
RJ GJ MZ
UP KE HR
UP MH
UKHP

0
HP AP JH JKGAMH TN TR
010

GJ JH MG
JH
BH GJ MG
AS
HRCH

-10
KR BH WB
PB OD MG
0

RJ UP TN MH TR
-2

MP JK ODWB AS
JKOD

-20
MN MN
BH NG AR
MZ TR
-10

AP MN NG AR

-30
KE NG AR
-4

MZ

-5000
-5000 00 5000
5000 -5000 0 5000
RRT
RRTperper
capita controlling
capita forfor
controlling being landlocked
being landlocked RRT per capita controlling for being landlocked

Figure 5d: Governance index and per-capita RTT

controlling for landlocked nature 50
Recommended Readings
• Chapters 1, 12, 13, Economic Survey 2016-2017
• “Growth in the 2000s - Key Facts”, Arvind Subramanian, Business
Standard, June 22, 2011
• “Globalization and States’ Growth”, Arvind Subramanian, Business
Standard, November 23, 2011
• “Alms Dealers”, Philip Gourevitch, The New Yorker,October 2010
http://www.newyorker.com/magazine/2010/10/11/alms-dealers
• Natural Resource Abundance and Economic Growth, Sachs J.,
Warner A., NBER Working Paper No. 5398
• “The Demographic Dividend: Evidence from Indian States”, Mody A.,
Aiyar S., IMF Working Paper WP/11/38
51