What made Indian Cities and Towns Grow in the 2000’s? Stylized Facts and Determinants Rana Hasan*, Yi Jiang*, and Debolina Kundu** India Policy Forum July 11-12, 2017 *Asian Development Bank and **NIUA The views expressed in this presentation are those of the authors and do not necessarily reflect the views and policies of the ADB or its Board of Governors or the governments they represent.
Outline 1. Background and motivation 2. Data 3. Results: – City size – City growth 4. Discussion
1. Background and motivation • Cities are widely believed to be “engines of growth” … and (better) jobs • But, are there certain characteristics that cities should possess to enable them to play this role? • Some skepticism about urbanizations underway in the developing world (e.g., the “self - organized” cities of Hendersen, 2014; “consumption cities” of Gollin et al, 2016)
The Indian context • Though slower than others, India Urban share, 1960-2015 60 is urbanizing. Will this be good for PRC growth? Will it generate more 55 INO productive and better paying THA 50 jobs? Urban population (% of total) 45 • Considerable concern that India 40 may be losing an important 35 VIE IND opportunity (Ahluwalia et al, 30 2014; Kundu and Samanta, 2011; 25 and GoI 2011) 20 – Urban infrastructure 15 – Economic and spatial planning 10 – Governance frameworks Source: World Development Indicators Database.
Empirical literature • The empirical literature on urban issues in India is small but growing. • Much of the work so far has been at the district level – Chauvin et al (2016) on the importance of urban agglomeration economies. They find 7-8% elasticity of nominal wages to district urban density and a larger elasticity w.r.t. population – Lall et al. (2004) find little benefit of urban density at district level on firm output in India – Ghani et al. (2014) on whether the structure of economic activity influences employment growth. They find more diverse industrial structures associated with higher employment growth — but, this is driven by rural India
Empirical literature (2) Recently, more studies at the city level are emerging • Transport infrastructure and city/urban growth (e.g., Alder, Roberts, and Tewari, 2017) • Size of urban agglomeration economies (Hasan, Jiang and Rafols, 2017) • Urban form and shape of cities and economic growth (Harari, 2016 and Tewari et al, 2017)
• Some takeaways: – Unit of analysis matters. Example: Agglomeration economies seem to be smaller when switching from district to cities as the unit of analysis – Use of satellite image based data is growing • Better connected and more compact cities experience faster economic growth (as proxied by nightlights data) • In this paper: – Take a step back and use a traditional but rarely used combination of data: Population and economic censuses – Examine urbanization over 2001 and 2011 to see how city-level economic structure, infrastructure/connectivity, and human capital are related to city size and growth
Main results • Notwithstanding the rapid growth of small towns between 2001 and 2011, urbanization has been driven by larger cities and higher income states • Larger cities score higher on measures of human capital, infrastructure access, more diverse economic structures, larger share of employment in formal firms, etc. • However, not all of these factors seem to matter for city growth. Measures of city level human capital and infrastructure are surprisingly poor correlates of urban growth • Connectivity to other locations and an economic environment conducive to manufacturing, and, especially, new firm formation seem to be better predictors of urban growth • Results are consistent with the idea that Indian cities are “centers of production”; urban policy should help cities work as labor markets.
2. Data • Population census: Town directory and primary census abstract, 2001 and 2011 – Urban centers classified as statutory towns and census towns. Urban agglomerations (and outgrowths) have been identified from the Primary Census Abstract. – Terminology: Urban centers with population 100,000 or above termed as cities; below that are towns. – City/town-level information on population, infrastructure (roads, electricity), literacy, educational and other social infrastructure, etc. • Economic Census 1998 – Covers establishments in all economic activities except crop production and plantation; location ID variables available at state, district, and city/town levels – Provides information on industry of activity (4-digit NIC 1987), total and hired employees, years of operation (a proxy for age), registration status, ownership type – Use data for 22 manufacturing and 22 service sectors to create various measures, including share of employment in manufacturing, in firms with 10+ workers, share of young firms, and measures of diversity and specialization of employment structure.
Data (2) • Road connectivity and market access – We compute the straight- line distance from a city’s center to the nearest state highway and to the nearest national highway or expressway by 2001 as two measures of a city’s road connectivity. – For market access, we consider each city’s access to the largest 74 cities with population above 500,000 in 2001 and calculate: where d ci is the distance form city c to one of the 74 large cities, city i travelled through available road network. d ci is set to 1 if c=i.
Matching of town directories across 2001 and 2011 • Limit attention to urban centers with population of 10,000+ in 2001 • Use town IDs to merge the town directories, checking town names, district names and state names to verify and match. • Information on composition of urban agglomerations (UA) obtained from primary census abstract; UAs are treated as as an integrated city/town • Trim sample to towns/cities with population growth greater than -50% and less than 500%; and area growth greater than -10% and less than 500%. • Final dataset: 2,427 cities/towns belonging to 502 districts across 21 states and 4 union territories
City size and growth OFFICIAL NUMBERS OUR SAMPLE Pop. Cont. Pop. Cont. Population Size 2001 2011 Growth to 2001 2011 Growth to (%) Growth (%) Growth Below 100k 79,181,305 98,025,832 23.8 22.6 63,975,854 74,617,788 16.6 18.6 100k and above 200,098,105 264,745,519 32.3 77.4 182,180,228 228,831,647 25.6 81.4 Total 279,279,410 362,771,351 29.9 100.0 246,156,082 303,449,435 23.3 100.0 OFFICIAL NUMBERS OUR SAMPLE Population Size Pop. Pop. Cont. to Cont. to 2001 2011 Growth 2001 2011 Growth Growth Growth (%) (%) 77.4 Class I: 100k and above 264,745,519 32.3 25.6 81.4 200,098,105 182,180,228 228,831,647 Class II: 50K to 99,999 18.3 6.0 16.2 6.7 27,192,982 32,179,677 23,592,327 27,410,437 Class III and IV: 10K to 16.6 26.7 16.9 11.9 49,999 51,988,323 65,846,155 40,383,527 47,207,351 Total 279,279,410 362,771,351 29.9 100.0 246,156,082 303,449,435 23.3 100.0
City-size distributions: 2001 vs 2011 .6 .4 .2 0 8 10 12 14 16 Log Population 2001 Low Income 2001 High Income 2011 Low Income 2011 High Income
Initial population versus population growth 400 300 200 100 Surat Kochi Bangalore Coimbatore Indore Ahmadabad Hyderabad Agra Lucknow Jaipur Chennai Bhopal Madurai Vadodara Patna Varanasi Delhi Ludhiana Nagpur Greater Mumbai Kanpur Kolkata 0 Visakhapatnam -100 10 12 14 16 18 Log Population, 2001
Matching of Town Directory 2001 and EC-4 (1998) by urban centers • EC-4 follows spatial boundaries as of August 1997 • Used secondary sources such as the administrative atlas of India, pin code database, digital maps and other online resources to track changes in boundaries and match locations with population census 2001 • Nature of city/town level activity captured by various measures: – Share of manufacturing employment (22 industries) in “total” employment (22 mfg + 22 services industries) – Share of employment/firms in 10+ worker firms (to capture formality); also 100+ worker – Share of employment/firms in young firms (<=5 years of operations) – Indexes of diversity and specialization
Recommend
More recommend
