«Shlomo Angel, Stephen C. Sheppard, and Daniel L. Civco With Robert Buckley, Anna Chabaeva, Lucy Gitlin, Alison Kraley, Jason Parent, and Micah Perlin ...»
The Dynamics of
Global Urban Expansion
Shlomo Angel, Stephen C. Sheppard, and Daniel L. Civco
Robert Buckley, Anna Chabaeva, Lucy Gitlin, Alison Kraley,
Jason Parent, and Micah Perlin
Transport and Urban Development Department
The World Bank
The Urban Growth Management Initiative
Background: Digital Globe QuickBird Image of London, 28 July 2002
Inset: 28 May 1989 (red) and 19 June 2000 (yellow) Built-up Pixels
Superimposed on Landsat Data
The Dynamics of
Global Urban Expansion Shlomo Angel, Stephen C. Sheppard and Daniel L. Civco With Robert Buckley, Anna Chabaeva, Lucy Gitlin, Alison Kraley, Jason Parent, and Micah Perlin Transport and Urban Development Department The World Bank Washington D.C., September 2005 © 2005 Department of Transport and Urban Development, The World Bank Disclaimer: The opinions and the information presented in this report are the sole responsibility of the authors of the report, and in no way reflect the policies of the World Bank. Care has been taken to avoid mistakes, and the remaining mistakes are also the sole responsibility of the authors of the report.
Background⎯Digital Globe Quickbird image of London, 28 July 2002.
Inset⎯28 May 1989 (red) and 19 June 2000 (yellow) built-p pixels of London superimposed on Landsat data.
An electronic version of this report is available at http://www.williams.edu/Economics/UrbanGrowth/DataEntry.htm.
Please address all inquiries to Dr. Shlomo Angel at email@example.com, Dr. Stephen C.
Sheppard at firstname.lastname@example.org; or to Dr. Daniel L. Civco at email@example.com.
The Dynamics of Global Urban Expansion ii
ABSTRACTThis study examined the dynamics of global urban expansion by defining a new universe of 3,943 cities with population in excess of 100,000 and drawing a stratified global sample of 120 cities from this universe. Population data and satellite images for two time periods⎯a decade apart⎯were obtained and analyzed, and several measures of urban extent and expansion⎯among them the built-up area of cities and the average density of the built-up area⎯were calculated. Data for 90 cities out of the global sample of 120 is presented and analyzed in this report. Weighted averages of the built-up area and the average density, as well as compactness and contiguity measures⎯and their change over time⎯are presented for nine regions, four income groups and four city size groups covering the entire globe. Densities in developing-country cities were found to be some three times higher than densities in cities in industrialized countries, and densities in all regions were found to be decreasing over time. If average densities continue to decline at the annual rate of 1.7%⎯as they have during the past decade⎯the built-up area of developing-country cities will increase from 200,000 km2 in 2000 to more than 600,000 km2 by 2030, while their population doubles. Ten econometric models that sought to explain the variation in urban extent and expansion in the universe of cities were constructed, and several hypotheses postulated by neoclassical theories of urban spatial structure were tested. All tests yielded R2 values in excess of 0.80. The policy implications of the analysis are presented and discussed. The Central message of this study is quite clear: Developing country cities should be making realistic⎯yet minimal⎯plans for urban expansion, designating adequate areas for accommodating the projected expansion, investing wisely in basic trunk infrastructure to serve this expansion, and protecting sensitive land from incursion by new urban development.
The Dynamics of Global Urban Expansion iii
1. The magnitude of global urban expansion The population in developing-country cities is expected to double in the next thirty years: from some 2 billion in 2000 to almost 4 billion in 2030.1 According to our own preliminary estimates, cities with populations in excess of 100,000 contained 1.7 billion people in 2000, and their total built-up area ⎯at average densities of some 8,000 persons per square kilometer2⎯was of the order of 200,000 square kilometers at that time. If average densities continue to decline at the annual rate of 1.7%⎯as they have during the past decade⎯the built-up area of developing-country cities will increase to more than 600,000 square kilometers by 2030. In other words, by 2030 these cities can be expected to triple their land area, with every new resident converting, on average, some 160 square meters of non-urban to urban land during the coming years.
In parallel, the urban population of industrialized countries is now expected to grow by 11% in the next thirty years: from some 0.9 billion to 1 billion.3 According to our own provisional estimates, cities with population in excess of 100,000 contained some 600 million people in 2000, and their total built-up area ⎯at average densities of almost 3,000 persons per square kilometer4⎯was of the order of 200,000 square kilometers at that time. If average densities continue to decline at the annual rate of 2.2%⎯as they have during the past decade⎯the built-up area of industrialized-country cities will increase to some 500,000 square kilometers by 2030. In other words, by 2030 these cities can be expected to increase their populations by 20% and their land areas by 2.5 times, with every new resident converting, on average, some 500 square meters of non-urban to urban land.
In total, urban built-up areas in the world consumed some 400,000 square kilometers in 2000, or 0.3% of the total land area of countries, estimated at some 130 million square kilometers.5 The land taken up by cities amounted to some 3% of arable land, estimated See United Nations, 2004, World Urbanization Prospects⎯The 2003 Revision, New York: United Nations, table 1, 14. The urban population in developing countries is expected to grow from
1.93 billion in 2000 to 3.97 billion in 2030.
The weighted average built-up area density for developing-country cities in our provisional sample of 90 in 2000 was found to be 8,049 persons per square kilometer (see table IV-2, Chapter 4).
United Nations, 2004, table 1, 14. The urban population in industrialized countries is expected to grow from 0.88 billion in 2000 to 1.01 billion in 2030.
The weighted average built-up area density for industrialized-country cities in our provisional sample of 90 in 2000 was found to be 2,824 persons per square kilometer (see table IV-2, Chapter 4).
This estimate is considerably lower than previously published estimates. The Earth Institute
at 14 million square kilometers in 2000.6 Cities are now expected to grow 2.5 times in area by 2030, consuming some 1 million square kilometers, or 1.1% of the total land area of countries. They may possibly consume as much as 5−7% of total arable land, depending on the future rate of expansion of arable land, which is currently 2% per annum.
2. The implications of urban expansion What are the implications of the accelerated rate of global urban expansion and what can or should be done about it?
The basic dimensions of the policy debate on the expansion of cities are certainly not new. The age-old question underlying this debate is still whether expansion should be resisted, accepted, or welcomed. At one extreme, there have been those who fought to limit the growth of cities by any and all means. At the other, there were those who welcomed it and actively prepared cities for absorbing the oncoming waves of new migrants. Two historical examples⎯one from London and one from New York⎯can serve to frame this debate.
In 1580, under pressure from the influential guilds, which were fearful of competition from recently arrived craftsmen, Queen Elisabeth issued a proclamation restricting development near and within the city. Enacted by Parliament in 1592, her decree had three major provisions: to prohibit “any new building of any house or tenement within three miles of any of the gates of the said city of London; to restrict the construction of habitations ‘where no former house has been known to have been’; and to forbid in any house “any more families than one only to be placed”….
[B]etween 1602 and 1630, no fewer than fourteen such proclamations were enacted in attempts to limit London’s growth.7 In contrast, in 1811, when New York City had only 100,000 people crowded into the southern tip of the island of Manhattan, three Commissioners⎯Morris, de Witt Clinton and Rutherford⎯ drafted a plan to expand its street grid so as to prepare for more than a tenfold increase in the city’s population. In presenting their now-famous plan, the
urban areas, an increase of at least 50% over previous estimates that urban areas occupied 1of the Earth’s total land area”; see Earth Institute News, posted on 8 March 2005 at www.earthinstitute.columbia.edu/news/2005/story03-07-05.html. The GRUMP estimates are based mostly on night light data, as against the Landsat data used in our estimates. These and other differences in defining and measuring urban built-up areas will be discussed at length in Chapter IV.
World Bank, World Development Indicators⎯2005, Washington DC: World Bank.
Lai, Richard Tseng-Yu, 1988, Law in Urban Design and Planning, New York: Von Nostrand,
To some it may be a matter of surprise that the whole island has not been laid out as a city. To others it may be a subject of merriment that the commissioners have provided space for any population that is collected at any spot on this side of China.8 Four hundred years have passed since the Queen’s proclamation and two hundred years since the Commissioners’ plan. Still, the fundamental question of whether urban expansion should be resisted, accepted or welcomed is still with us today and is still largely unresolved. While many will readily agree that urban expansion is an issue of serious concern, there is no consensus among scholars, policy makers or urban residents themselves about whether further development should be restricted or encouraged. In the US, for example, respondents to a survey in 2000 by the Pew Center for Civic Journalism “were almost evenly split between those wanting local government to limit further development to the infilling of already built-up areas and those wanting local government to also plan for and encourage new development on previously undeveloped areas”.9 In industrialized countries, where rural-urban migration is now minimal and where most population movements are now inter-urban or intra-urban, there have been recent attempts to provide answers to this question that are particularly relevant to their present level of urbanization and development. Concerns for unwieldy urban expansion⎯ typically castigated as “sprawl”⎯have recaptured the attention of both policy makers, academics and, more recently, voters during the last decade. In contrast, we note, virtually no attention has been paid to this issue in developing countries, where levels of urbanization and development are typically lower, where rural-urban migration has by no means ebbed, and where most urban population growth is about to take place.
The central objective of the Urban Growth Management Initiative is to examine the available policy options for confronting the projected urban expansion in the cities of developing countries. In other words, it seeks an answer to the question of what can and should be done about it. This demands gaining a better understanding of the key dimensions of this expansion as well as of the forces that are driving it globally, regionally and locally, so as to be able to consider carefully the kinds of policies that are likely to be effective, efficient, equitable and sustainable, while keeping in mind that such policies may be quite different from those available or of interest in industrializedcountry cities.
Why should we concern ourselves with the projected spatial expansion of developing-country cities? Does urban expansion take place in substantially different forms, or it is essentially identical everywhere? Does it really matter in what form it takes place? What are the forces that are now shaping urban expansion? How can we Morris, de Witt Clinton and Rutherford, quote in Mackay, Donald A., 1987, The Building of Manhattan, New York: Harper and Row, 20.
Burchfield, Marcy, Henry G. Overman, Diego Puga and Mathhew E. Turner, 2004, “The
measure urban expansion in meaningful ways that address our concerns? What are the key policy areas that have a bearing on shaping urban expansion? At this early stage, the Urban Growth Management Initiative seeks to begin to provide meaningful answers to these questions and to lay the foundations for fruitful research on and effective action to manage urban expansion in developing-country cities.
3. Concerns about urban expansion Why should we concern ourselves with the projected spatial expansion of developingcountry cities, and why now?
Considering that research and policy interests are often subject to fashion and that such fashions originate in the metropolitan centers of industrialized countries, we should suspect that the recent concerns with “sprawl” would be diffused globally, and sooner rather than later. These concerns have now become paramount, especially in the
In 1998, New Jersey voters approved a plan to buy one million acres of undeveloped land (20% of the state’s total land area) using state funding, to ensure that this land is never developed. Between 1998 and 2002, another 620 ballot measures allocating $25 billion in public funds for land conservation measures were approved by voters across the United States.