Subaltern Urbanization in India?

Movement of People
Transformation of Place

Partha Mukhopadhyay
Centre for Policy Research New Delhi

India China Institute

Sept. 12, 2012
The New School University
 References
• Basic references
– Denis, Eric and Kamala Marius-Gnanou (2011) “Toward a Better
Appraisal of Urbanisation in India”, Cybergeo: European Journal
of Geography,569.
– Denis, E., P. Mukhopadhyay and M.H. Zerah 2012. ‘Subaltern
urbanisation in India’ Economic and Political Weekly, XLVIII (30):
52-62.
– Pradhan, K. C. 2012. ‘Unacknowledged urbanization: The
census towns of India’ CPR Urban Working Paper 2, Centre for
Policy Research, New Delhi.
• Additional references
– Chandrasekhar, S (2011): “Workers Commuting between the
Rural and Urban: Estimates from NSSO Data”, Economic and
Political Weekly, 46 (46): 22-25
– Uchida, H and A Nelson (2010): “Agglomeration Index: Towards
a New Measure of Urban Concentration”, Working Paper
2010/29, United Nations University-World Institute for
Development Economics Research (UNU-WIDER), Helsinki

2
CPR CSH Urban Workshop Series, September 25, 2012
 “Subaltern Urbanisation”
• “If the ‘global city’ were • At one level, ‘subaltern
labelled as just another urbanisation’ perhaps
example of an best seen as a literary
‘industrial’ district device to:
(perhaps it should – focus attention on our
rather be called: new area of inquiry
industrial districts of – increase the possibility of
transnational discursive engagement
management and • Attempt to:
control), it might not – Interpret:“contribution
have attracted the made by the people on
attention it did.” their own, that is
– Jennifer Robinson (2002: independently of the elite”
536) – Read official urbanisation
data ‘against the grain’
• Not about the city, but
about the urban system
3
India China Institue Sept. 12 2012 The New School
 Plan of Presentation

• How urban is India?

• Diversity of urbanisation
– Large Cities
– Census Towns

• Economic Structure of Non-Metro Urban Areas

• Conclusion

4
India China Institue Sept. 12 2012 The New School
How Urban is India?
 Defining ‘Urban Areas’ over Time
1872: settlement with a population of 5000. 1951: “places with a smaller population
1881: “a collection of numerous dwellings [than 5000] with definite urban character
near each other within a limited area may be treated as separate towns”.
having shops which provide a continual 1961: current three-fold definition
open market for the supply of goods
especially of manufactured goods” ‘Census Towns’
– density
1921: 1040 settlements comprising 22% of the
nature of economic activity.
urban population which were “towns arbitrarily
1891: population size of 5000 reinstated as a classified as such for census purposes”
criterion for settlements not under 1931 increase in administratively designated towns
“Municipal or Chaukidari Act” by 699 and a decrease in the number of such
– place should not be “merely a large village “arbitrarily classified” towns to 600 with 11% of
but should have some distinctly urban urban population.
character as that of a market town” ------------------------------------------------

– to exclude “a considerable number of the 2001: 1362 Census Towns comprising 7.4% of the
large aggregates of homesteads on the urban population
Malabar coast which are merely revenue 2011: increase in administratively designated
units of a purely agrestic nature”. towns by 242 and a increase in the number of
• “undesirable to classify as towns such ‘census towns’ towns to 3894 with
overgrown villages which have no around 15% of urban population.
urban characteristics”.
Source: Asok Mitra Population and area of cities towns and urban agglomerations 1872-1971 Allied Bombay 1980
6
India China Institue Sept. 12 2012 The New School
 Only in India…
Population Economic No. of
Administrative Density Others# Notes
Size Activity Countries
X 50
X 22
X 9
X X 5
X X 1 China
X X 2
X X 7
OR X OR 1 Sudan
One of these criteria AND X 2
Either administrative OR all three 1 India
# Specific types of urban infrastructure e.g. street lights proximity of built up areas etc.
Source: United Nations Demographic Year Book 2005
• International definitions can be non-comparable
– Economic criteria is used only in six of hundred countries
• Indian urban definition is unique and the intersection of three criteria
– On economic activity India’s threshold of 75% is the highest
• Japan is 60% Lithuania is 2/3 “commercial importance” in Sudan Zambia is majority
Botswana does not use a labour force criterion but overall economic activity
7
India China Institue Sept. 12 2012 The New School
 Settlement Structure 2001

All Villages All Towns

593,616 Population > 5,000 5,161
742.5m 18,760
161.5m 286.1m

13,884 2,659
125.8m 30.8m
28.1 million
2,375 people in 2375
28.1m
Density > 1,000 Male Non-Ag settlements
per sq. mile Workforce > 75% met the urban
228,717 15,699 28,102 test in 2001
416.5m 48.8m 58.1m but were not
classified as
urban by the
census

8
India China Institue Sept. 12 2012 The New School
India may be more urban than it seems
Uchida and Nelson (2008) Denis and Marius-Gnanou (2011)
• Extensive inter-country • Global e-geopolis exercise
exercise using GRUMP data • Built-up definition
• Proximity definition – Locations of contiguous (defined
as less than 200 metres apart)
– The road network and built up areas from satellite
settlement data are used to imagery are matched geo-spatially
determine travel times with settlements from the Census
between settlements
• India
• India (based on 2001 census) – 37.1% in agglomerations of more
– 42.9% live within an hour of at than 10000 compared to 26.6% in
least a Class I town towns of more than 10000
– 52% live within an hour of at least – Built-up settlements may or may
a Class II town not meet the economic activity
• China criteria but are likely to meet the
density and population criteria
– Stays at official number of 36%
– Bihar goes from 10.4% to 31.2%
• Is India more urban than • Villages do not meet the non-farm
China? criterion
9
India China Institue Sept. 12 2012 The New School
 Official and….

Source: Denis and Marius-Gnanou (2011)

10
India China Institue Sept. 12 2012 The New School
 Bihar: from 10.4% to 31.2%

Source: Denis and Marius-Gnanou (2011)

11
India China Institue Sept. 12 2012 The New School
 India may be more urban than it seems
Chandrasekhar (2011) Growing Connectedness
• NSS 2009-10
• Commuting
– People who live in rural areas
and work in urban areas
• Results
– 8.05 million rural non-agricultural
workers commute to urban areas
– 9.1% of the total urban non-
agricultural workforce • 300,000 km of rural roads constructed
under the Prime Minister’s Rural
• Attractions of village life or Roads Program
urban exclusion?
– Fifteen years ago, India had
– Greif and Tabellini (2012) find that
urbanization was slower and more almost no four lane highways.
sparse in European regions where – As of April 2011, 15,000 km is
family traditions denote stronger four-laned and 10,000 km is
kin-based obligations
under implementation
12
India China Institue Sept. 12 2012 The New School
 Takeaway
• Urban may be more important than appears from
the Census
– Many people live in close proximity to a large town
– Rural non-farm economic activity may be linked to urban
proximity
• Himanshu et. al. (2011)
• 2009-10: 27.1% of labour force is rural non-farm vis-à-vis 25.5%
in urban non-farm
– Even in areas where farm activity dominates the
population is living in large built-up agglomerations
• Some ‘urban areas’ are not classified as urban
– 28.1 million people in 2375 settlements met the urban test
in 2001 but not classified as such
• Of these, 18.7 million people in 1625 settlements are new
census towns in 2011 (Pradhan 2012)

13
India China Institue Sept. 12 2012 The New School
 Diversity of Urbanisation

draws liberally on
Denis, Mukhopadhyay and Zerah (2012)
Diversity of Urbanisation

Larger Cities
 Political Influence on Urban Growth?
Single Growth Centre Multiple Growth Centres

District(s) of/including Chhattisgarh Madhya Pradesh

Karnataka
State Capital Orissa

District(s) Near State Andhra Pradesh Arunachal Maharashtra Tamil Nadu Jammu
or National Capital Pradesh Haryana Punjab and Kashmir Uttar Pradesh

Other District(s) Jharkhand Kerala Gujarat Rajasthan West Bengal

Note: Growth centre districts are population growth outliers (those with population growth rate 1.65 standard
deviations more than the state population growth rate). Includes only states with more than ten districts. Assam
Bihar Himachal Pradesh Nagaland and Uttarakhand do not have any outlier districts

• Bengaluru is the only district including the state capital which has the
highest growth rate in the state
• In other states districts including the state capital or districts near the
state capital are high-growth districts (relative to the state growth rate)
• Is the political primacy of state capitals influencing the spatial
character of urban growth in India?
16
India China Institue Sept. 12 2012 The New School
 Major Indian Cities (above 5 mn.)
B: Ahmedabad
C: Mumbai
I : Pune
D: Bengaluru
E: Chennai
F: Hyderabad
G: Kolkata
H: Delhi

• Each one of these,

except Pune, is the I
administrative
capital of the state
where it is located
• Well distributed but
with a peripheral
character
17
India China Institue Sept. 12 2012 The New School
Growth is happening around large cities
Population Core Growth Peripheral Peripheral
City
2011 (UA) mn. (% p.a) Growth (% p.a) District

Mumbai* 18.4 0.4 3.6 Thane

Delhi 16.3 2.1 7.4 Gurgaon

Kolkata 14.1 1.3 1.8 South 24 Parganas

Chennai 8.7 0.8 3.9 Kancheepuram

Bengaluru 8.5 4.7 1.6 Bengaluru Rural

Hyderabad 7.7 1.6 4.8 Rangareddi

Ahmedabad 6.4 2.1 1.2 Gandhinagar

Pune 5.0 3.4 3.0 Pune

*Mumbai includes Mumbai and Mumbai (Suburban)

18
India China Institue Sept. 12 2012 The New School
 Urban Growth around Delhi and Mumbai
Urban Growth Growth
2001 2011 Urban Population 2001 2011
Population p.a. p.a.
NCT of Delhi 12906065 16333916 2.4% Mumbai City 3337895 3145966 -0.6%
Ghaziabad 1816413 3144574 5.6% Mumbai Suburban 8640386 9332481 0.8%
Faridabad 1062286 1429093 3.0% Mumbai 11978281 12478447 0.4%
Gurgaon 309704 1042000 12.9% Thane 5902467 8503094 3.7%
G. B. Nagar (NOIDA) 442271 997410 8.5% Raigarh 534834 972809 6.2%
NCT
3630674 6613077 6.2% Neighbourhood 6437301 9475903 3.9%
Neighbourhood
Meerut 1451992 1762573 2.0% Total 18415582 21954350 1.8%
Bulandshahar 681583 867791 2.4%
Panipat 392076 552945 3.5% • While population growth in the core of
Sonipat 321371 451687 3.5% both Delhi and Mumbai have slowed down
Rohtak 329593 444819 3.0% Delhi’s immediate neighbourhood is
growing much more rapidly than Mumbai.
Baghpat 229440 274135 1.8%
• Delhi’s neighbourhood is smaller in
Jhajjar 195097 242974 2.2%
proportion to the core city compared to
Palwal 159038 235663 4.0%
Mumbai
Rewari 136172 231411 5.4%
• Is this due to concavity or the
Mewat 59301 124017 7.7% presence of other strong urban
Rest NCR 3955664 5188015 2.7% centres nearby, e.g., Pune and
NCR 20492403 28135008 3.2% Nashik? 19
India China Institue Sept. 12 2012 The New School
 Changing Size Distribution 2001-11
40% 2001 2011_New

35% 2011_Old

30%
5.9%
2.6%
25%

20%
34.1% 32.9%
15% 28.4%
25.6% 24.8% 25.1%
10% 7.6%

5%
7.4%
5.6%
0%
Million Plus 100,000 to 1 million Census Towns Other Urban Areas

20
India China Institue Sept. 12 2012 The New School
 Urban population is in larger cities
40.0%
37.6%
35.0%
32.0%
30.0%
27.0% 26.8%
25.0% 24.7%
22.5%
20.0%
18.5%

15.0%

10.9%
10.0%

5.0%

0.0%
1951 1961 1971 1981 1991 2001

Half-Million Plus 100,000-500,000 20,000-100,000 Less than 20,000

Source: Town Directory, Census of India 2001

In 1951 41% of urban population was in settlements of more than 100,000

which rose to 62% in 2001 and 70% in 2011, if we consider urban
agglomerations. The share living in cities over 500,000 rose from 18.5% in
1951 to 50.5% in 2011 21
India China Institue Sept. 12 2012 The New School
But, more large cities are visible…

UA existing in 2001 by size New UA in 2011

Less than 500,000 to More than Less than 500,000 to
500,000 1 million 1 million 500,000 1 million
Growth of UA (2001-2011) 3.0% 2.9% 2.3% .. ..
Growth of Core City (2001-2011) 2.0% 2.0% 2.5% 1.3% 2.5%
Share of Core City in UA Pop. 76% 79% 72% 86% 84%
Share of UA in Urban Pop. 10.2% 7.0% 34.0% 2.1% 2.0%
Share of UA in Urban Pop. (2001) 10.0% 6.9% 35.7% .. ..
Share of Core in Urban Pop. 7.8% 5.5% 24.5% 1.8% 1.7%
Share of Core in Urban Pop. (2001) 8.4% 5.9% 25.3% 1.6% 1.3%
Number of UAs 139 28 31 38 8

• Growth in share of population in large cities is not driven by migration to big

cities. It is also the growth of existing mid-level cities
• The share of population of half-million plus UAs that existed in 2001 has
declined in 2011 from 42.6% to 41% but 8 new half-million plus UAs have
emerged in 2001, that add 2% of urban population 22
India China Institue Sept. 12 2012 The New School
and, large cities have small beginnings
Size
2011 500 to More than 1
100 to 200 200 to 300 300 to 400 400 to 500
Size 1 million million
1961
N.A. 6% 13% 5% 14% 7% 6%
Less than 50 67% 34% 22% 10% 10% 3%
50 to 100 25% 41% 42% 25% 12% 6%
100 to 200 1% 10% 31% 45% 43% 7%
200 to 300 1% 5% 23% 10%
300 to 400 5% 10%
400 to 500 7%
500 to 1 mn. 11%
More than 1 mn 41%
100% 100% 100% 100% 100% 100%

• Diverse origin of today’s class I cities

– Over 50 years, a four-fold increase means 2.8% growth p.a.
– While a number of the super-fast (10X) growing towns are administrative or near
large metros, many are not
• Nashik, Bhiwandi and Aurangabad (Maharashtra) Surat and Vapi (Gujarat),
Rudrapur (Uttarakhand), Akbarpur (UP), Saharsa (Bihar), Miryalaguda (AP)
• What do we know of such towns? 23
India China Institue Sept. 12 2012 The New School
 China’s in-situ urbanization

China India
• In-situ urbanization • Similar process leading to
(Zhu 2002) dispersed urbanization and
– Cities like Jinjiang and fast growing Indian cities
Quanzhou in Fujian grew – Surat (Gujarat) is an old
from inside-out trading town,
• Historical contingency – Aurangabad, (Maharashtra)
is a textile centre that now
• Good road and telecom has many industries,
connectivity
– Miryalaguda (Andhra
• Population density
Pradesh) is a rice market
• Local access to capital town that has diversified
– Independently connected – Tiruppur (Tamil Nadu) is a
to global markets global knitwear production
centre
24
India China Institue Sept. 12 2012 The New School
Diversity of Urbanisation

Census Towns
Based on Pradhan (2012)
The drops in the Ocean…now 15%
• In-situ urbanisation?
– Almost a third of the growth in
urban population over 2001-11 is
reclassification of villages as
census towns
• Estimated by matching new census
India: towns to villages in 2001 and
High Estimate: 29.5% increasing village populations in
Low Estimate: 26.0% 2001 by the state population growth
rate
– Extent varies across states
• Estimate of migration is
22.2%
– 44% natural growth (Bhagat 2011)
– 29.5% villages to Census Towns
– 2.3% villages to Statutory Towns
– 2% boundary expansion (minimum)
– 22.2% is residual estimate of migration

26
India China Institue Sept. 12 2012 The New School
 Kerala Desakota

Almost all the urban growth in Kerala

over 2001-11 is due to the growth of
census towns
Source: Denis and Marius-Gnanou (2011)

27
India China Institue Sept. 12 2012 The New School
 Census Towns of 2011, c. 2001

• 64% (1625 of 2553) settlements of18.5 mn classified as new census towns

in 2011 already met the three-fold test in 2001
– Remember the 28.1 million ‘unrecognised’ urban people
• 90% (all except 258) of new census towns in 2011 met the density and
economic activity test in 2001 (urban under the 1951 definition) 28
India China Institue Sept. 12 2012 The New School
New Census Towns are not outgrowths
Case-I Case-II Case-III
Size Class of Class I Towns
(Base) (+25%) (-25%)
(base case radial distance)
No (Pop.) No (Pop.) No (Pop.)
100,000 to 500,000 (10 km) 45.1% (42.3%) 41.9% (41.1%) 51.7% (49.4%)
500,000 to 1,000,000 (15 km) 14.8% (18.6%) 14.9% (18.3%) 14.7% (17.5%)
1,000,000 to 4,000,000 (20 km) 18.4% (15.6%) 17.1% (14.3%) 19.5% (16.5%)
More than 4,000,000 (25 km) 21.7% (23.4%) 26.1% (26.3%) 14.1% (16.7%)
Proximate to Large Towns 926 (7.8) 1115 (9.5) 735 (6.2)
Distant from Large Towns 1563 (15.4) 1374 (13.7) 1754 (16.9)
Note: If a CT is near multiple classes of city proximity, then it is considered under the proximity of larger city class.

• Population of new CTs within a radial distance of Class I

towns is only 33% (41% in case II)
– Proportion near million plus is only 13% (17% in case II)
– Much less than proportion of urban population in larger towns
• Two broad types of relationship with larger towns?
– Proximate (dependent?)
– Distant (independent?)
29
India China Institue Sept. 12 2012 The New School
New Census
Towns are
Widely Spread

While there
appears to be a
concentration
around the large
cities of Delhi,
Mumbai, Pune,
Chennai,
Hyderabad and
Kolkata, there is
also a large
number that are
spread widely
around the
country

30
India China Institue Sept. 12 2012 The New School
 Widely spread but not isolated
Size of SA
(2001) Less 10,000 30,000 50,000 100,000 200,000 Greater
Total in Not
Size than to to to to to than
SA in SA
of new 10,000 30,000 50,000 100,000 200,000 500,000 500,000
CT(2001)
Less than
12 52 22 34 52 41 128 341 376
5,000
5,000 to
413 138 36 86 99 67 280 1119 35
10,000
10,000 to
158 23 30 23 28 189 451 11
20,000
20,000 to
19 11 8 3 6 162 209 4
50,000
More than
1 1 5 7
50,000
425 367 92 159 177 143 764 2127
Total 426
(20.0%) (17.3%) (4.3%) (7.5%) (8.3%) (6.7%) (36%) (100%)

• Of the 1836 census towns that had a population of more than 5,000
in 2001, all but 50 are part of a settlement agglomeration, as defined
by Denis and Marius-Gnanou (2011)
– 988 of them part of agglomeration of more than 50,000
31
India China Institue Sept. 12 2012 The New School
Economic Structure of Non-Metro
Urban Areas

Based on
Denis, Mukhopadhyay and Zerah (2012)
 Million-Plus cities are but one part
Sector 1993-94 2004-05 2007-08 2009-10 Notes
Mining 3.0% 8.3% 4.4% 8.4% Up
Food Mfg. 10.6% 16.3% 13.7% 12.9% Stable
Clothing Mfg. 22.9% 33.2% 36.6% 33.8% Up strongly
Machinery Mfg. 39.0% 44.5% 48.5% 41.2% Stable
Other Mfg. 27.6% 35.3% 35.6% 32.8% Up
Utilities 18.6% 28.6% 21.2% 17.7% Stable
Construction 24.4% 25.5% 22.5% 16.9% Down strongly
Govt. services 28.5% 27.2% 29.1% 28.3% Stable
Traditional services 23.6% 27.9% 30.4% 27.5% Up
Modern services 37.4% 45.0% 45.0% 42.5% Up
Social services 19.9% 25.4% 23.1% 22.2% Stable
Household service 34.3% 42.6% 43.1% 46.6% Up strongly
Total 22.7% 28.7% 29.7% 27.0% Up

• Share of urban employment in million plus cities by economic sectors

– Million-plus cities are not the main hub of urban economic activity
– Share is not growing, except in clothing manufacturing and modern services
• Construction is growing strongly in smaller urban areas
– Is this because these areas are being built or because the construction workers
stay in these areas and work in the larger cities?
33
India China Institue Sept. 12 2012 The New School
 Economic Structure of Urban Areas
1993-94 2004-05 2007-08 2009-10
Million Other Million Other Million Other Million Other
Plus Urban Plus Urban Plus Urban Plus Urban
Mining 0.2% 1.7% 0.2% 1.1% 0.1% 0.9% 0.2% 0.9%
Manufacturing 27.4% 27.9% 28.2% 24.9% 28.5% 24.3% 27.0% 23.1%
Food Mfg. 1.8% 5.1% 1.6% 3.7% 1.3% 3.7% 1.3% 3.5%
Clothing Mfg. 9.1% 10.1% 10.5% 9.3% 10.6% 8.5% 10.5% 8.2%
Machinery Mfg. 5.0% 2.6% 3.9% 2.1% 4.8% 2.4% 4.5% 2.6%
Other Mfg. 11.6% 10.1% 12.1% 9.7% 11.8% 9.8% 10.7% 8.8%
Utilities 1.2% 1.8% 0.9% 1.0% 0.7% 1.2% 0.7% 1.3%
Construction 7.2% 7.5% 7.6% 9.7% 6.7% 10.7% 6.6% 13.0%
Govt. services 13.1% 10.9% 6.3% 7.4% 6.4% 7.2% 6.8% 7.0%
Traditional svc 34.3% 36.9% 33.9% 38.6% 36.5% 38.6% 34.9% 36.9%
Modern svcs 8.7% 4.8% 12.0% 6.5% 13.1% 7.4% 14.1% 7.7%
Social services 5.0% 6.7% 6.3% 8.1% 5.2% 8.0% 6.1% 8.6%
Household svc. 3.0% 1.9% 4.6% 2.7% 2.8% 1.7% 3.5% 1.6%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%
• Modern services is growing in million-plus cities but still relatively small part
– Manufacturing is stable but may be moving from smaller cities
• Economic structure of smaller cities is similar to large cities
– Similar functions at a different scale?
34
India China Institue Sept. 12 2012 The New School
Conclusion
 Conclusion
• Urban may be more important than appears from the Census
• Indian urbanisation is a distinct (subaltern?) story; a
“contribution made by the people on their own”
– Many facets of urbanisation and metropolitan growth is but one of them
• Many economically vital small settlements
– Contrary to perceptions that India’s urbanisation is low, its smaller
settlements are stagnant and its cities are unproductive (Nijman 2012)
• Not just movement of people, also the transformation of places
A Classification Scheme for Urbanisation
Administrative
Recognition Invisible Denied Contesting (I) Contesting (II) Recognised
Spatial proximity

Non-peripheral

Peripheral

Note: Contesting (I) refers to a situation where the settlement wants to be urban but the administrative
classification is rural, while the reverse situation is Contesting (II)
36
India China Institue Sept. 12 2012 The New School
 Thank You

partha@cprindia.org