Density by pengxuebo

VIEWS: 2 PAGES: 25

									     CHAPTER 1   Density




                 M             ostly off the world’s radar, on
                               a dusty plain in West Africa,
                               is a city of 1.6 million people.
                 Bisected by the River Niger, its two halves—
                 with about 800,000 people each—are linked
                                                                      Despite its industriousness, Bamako is
                                                                  one of the sleepier cities in West Africa.
                                                                  Many of the manufactured staples come
                                                                  1,184 kilometers by road from one of the
                                                                  region’s metropolises, Abidjan, which has
                 by only two bridges. The pressure of move-       more than twice Bamako’s population.
                 ment is so strong that every morning one of      Abidjan seems small beside Lagos, where
                 these bridges is dedicated to incoming traf-     activity is so concentrated that its residents
                 fic: minibuses, bicycles, motorbikes, pedes-      speak of living in a pressure-cooker. Some
                 trians, and occasionally private cars. In the    families rent rooms to sleep for six hours
                 evenings, to leave the center means joining      and then turn them over to another fam-
                 an exodus of people toward the minibus           ily that takes their place. Shopping does
                 depots. Green vans loaded with passengers        not necessarily require travel: goods are
                 file out to residential neighborhoods as far      brought on foot and cart to drivers stuck in
                 as 20 kilometers away. This is Bamako, Mali.     Lagos’s interminable traffic jams. To some,
                 It contracts into its center every morning       like the authors of Lagos’s 1980 master plan
                 and breathes out again in the evening.           written when the city had just 2.5 million
                     With each breath Bamako grows bigger.        residents, the continuing growth of the city
                 It happens to be one of the fastest-growing      is “undisciplined.”2 What can possibly be so
                 cities in the world. Natural demographic         attractive about living in Lagos that, despite
                 growth is supplemented by migration from         its congestion and crime, it continues to
                 the countryside and other Malian cities. Its     draw migrants?
                 population in 2008 is 50 percent larger than         The short answer: economic density.
                 10 years ago, making it the same size as         Lagos is not the most economically dense
                 Budapest, Dubai, or Warsaw. It has 10 times      city in the world, nor even the most densely
                 more inhabitants than the next biggest           populated. Those distinctions belong to
                 Malian city and accommodates 70 percent          Central London and Mumbai, respec-
                 of the country’s industrial establishments.1     tively. Even so, Nigeria’s economic future
                 New neighborhoods—quartiers—formerly             and Lagos’s growth are as inextricably
                 villages, become consolidated with the rest      tied as Britain’s economy is with London’s
                 of the city, toward the south, east, and west.   growth. No country has developed with-
                 Some of Bamako’s people are now moving           out the growth of its cities. As countries
                 out into surrounding neighborhoods in            become richer, economic activity becomes
                 search of cheaper land and some tranquil-        more densely packed into towns, cities,
                 ity, but they remain within reach of the city    and metropolises. This geographic trans-
                 because it provides their livelihoods.           formation of economies seems so natural

48
                                                                                         Density   49



that—at an impersonal aggregate level—it              nomic density continues to increase in
is taken for granted. But moving to eco-              a postindustrial economy because ser-
nomic density is a pathway out of poverty             vices are even more densely packed than
both for those who travel on it and, ulti-            industry.
mately, for those left behind. Jane Jacobs,       •   Rural-urban and within-urban dispar-
the noted urbanist, did not have Bamako               ities in welfare narrow with develop-
and Lagos in mind when she wrote, “A met-             ment. In the early stages of development,
ropolitan economy, if it’s working well, is           geographic disparities in welfare are
constantly transforming many poor people              large. With development, these gaps
into middle-class people, many illiterates            may increase initially. Rural-urban gaps
into skilled people, many greenhorns into             in income, poverty, and living standards
competent citizens. Cities don’t lure the             begin to converge as economies grow,
middle class. They create it.”3 She might             faster for access to social services, and
as well have written: as Lagos and Bamako             faster in areas of more vibrant growth.
grow, they will fi ll in West Africa’s missing         Within-city gaps in welfare and hous-
middle.                                               ing—most obvious in informal settle-
    This chapter introduces density, the first         ments or slums—persist for much
of the geographic dimensions of develop-              longer, and narrow only at later stages of
ment, defined as the economic mass or out-             development.
put generated on a unit of land. Surveying
the evolution of density with development,
                                                  •   Neither the pace of urbanization nor
                                                      its association with economic growth
the chapter presents stylized facts about             is unprecedented. Today’s developing
how density in a country rises with urban-            countries are sailing in waters charted by
ization, rapidly at fi rst, and then more              developed nations, which experienced a
slowly. These changes are associated ini-             similar rush to towns and cities. The
tially with a divergence of living standards          speed is similar, and the routes are the
between places with economic density and              same. What is different today is the size
those without, later with a convergence.              of the ship: the absolute numbers of peo-
Living standards thus eventually converge             ple being added every year to the urban
between areas of different density, such as           populations of today’s developing coun-
urban and rural. Even within cities, densely          tries are much larger than for even the
populated slums amid formal settlements,              most recent industrializers such as the
the differences slowly disappear with devel-          Republic of Korea and Taiwan, China.
opment. But this convergence does not hap-            Later chapters of this report investigate
pen by itself. It requires the institutions to        the policy implications of these similari-
manage land markets, investments in infra-            ties and differences.
structure, and well-timed and executed
interventions.                                    Defining density
    The main findings:
                                                  Density refers to the economic mass per unit
•   The concentration of economic activ-          of land area, or the geographic compactness
    ity rises with development. The world’s       of economic activity. It is shorthand for the
    densest areas or settlements are in devel-    level of output produced—and thus the
    oped countries. But the path to these lev-    income generated—per unit of land area. It
    els, “urbanization” in this Report, is not    can, for example, be measured as the value
    linear. The share of a country’s popula-      added or gross domestic product (GDP)
    tion settled in towns and cities rises rap-   generated per square kilometer of land.
    idly during its transformation from an        Given that high density requires the geo-
    agrarian to an industrial economy, which      graphic concentration of labor and capital,
    generally coincides with its development      it is highly correlated with both employment
    from low to middle income. The pace of        and population density. Density is the defin-
    urbanization slows after that, but eco-       ing characteristic of urban settlements.
50   WO R L D D E V E LO P M E N T R E P O RT 2 0 0 9



     The economic world is not flat                                    This density contrasts markedly with
     The geographic distribution of economic                       the agricultural areas of Belgium. In the
     activity, at any resolution, is uneven. No                    Flemish Flanders (Vlaams Gewest) area,
     matter the geographic scale examined, be                      6,323 square kilometers of land are used for
     it the country or a subnational area such as                  agriculture. Its area is almost 40 times that
     a province or district, there is a hierarchy                  of Brussels, but its employment is just 13
     of density. At the top is the primary city,                   percent of Brussels and its GDP a mere 4.5
     and at the bottom are agricultural lands or                   percent, translating into employment and
     rural areas. Between them is a continuum                      GDP densities of only seven workers and
     of settlements of varying density.                            €330,000 per square kilometer. The ratio of
         The geographic unevenness of economic                     output density between Brussels and Flan-
     mass, or bumpiness, tends to increase                         ders is 1,000 to 1. In between metropolitan
     with a country’s land area. But even the                      Brussels and rural Flanders is a range of set-
     economic geography of small countries is                      tlements, each with a different density (see
     bumpy. The Belgian city of Brussels has                       map 1.1). The cities of Antwerp, Brugge,
     a land area of 161 square kilometers, of                      Gent, and Leuven have an average output
     which 159 square kilometers are used for                      of €22 million and employment density of
     nonagricultural purposes. On this small                       342 workers per square kilometer.5
     area, a GDP of €55 billion is generated by                       In both developed and developing
     about 350,000 workers—that is, the aver-                      countries, then, the economic landscape is
     age square kilometer of land has more than                    bumpy. But the topography does not corre-
     2,000 workers annually producing almost                       spond to a simple urban-rural dichotomy.
     €350 million of services and goods. Brus-                     A continuum of density gives rise to a port-
     sels not only has high densities of GDP and                   folio of places. At the head is a country’s
     employment; it also has the highest popu-                     leading, primary, or largest city. Below the
     lation density of any European (EU27)                         primary city is a spectrum of settlements—
     area classified as NUTS1 (Nomenclature                         secondary cities, small urban centers,
     of Territorial Units for Statistics)—more                     towns, and villages (see figure 1.1). In some
     than 6,000 people per square kilometer,                       countries, such as France and Mexico, the
     18 times the average for Belgium.4 For the                    size difference between the top two cities is
     sake of comparison, the population den-                       phenomenal. With a population of 10 mil-
     sity of London and Madrid is about 5,000                      lion, Paris dwarfs second-ranked Marseille
     people per square kilometer.                                  with just 1.5 million. And with a population



     Map 1.1    The landscape of economic mass is bumpy, even in a small country like Belgium



                                                              Brussels

                    BELGIUM
                                            Antwerp
                                                                     FLANDERS


                                            Gent                                    WALLONIA
                                                                            Leuven

                   Brugge




     Source: WDR 2009 team and World Bank Development Research Group, based on subnational GDP estimates for 2005. See also
     Nordhaus 2006.
                                                                                                                           Density                                                 51



of 22 million, Mexico City is more than four                        Figure 1.1         From dichotomy to continuum: a portfolio of places
times as populous as Guadalajara, Mexico’s                           The simplified area economy                             and a more realistic representation
second city. Conversely, in India and the
United States, the size difference between                                                                                                                    Urban
the two biggest cities is relatively small. With                                                                                                                 Metropolis
populations of more than 22 million people,                           Rural                     Urban              Rural                           Towns         Large city
Mumbai and New Delhi stand shoulder to                                                                                                                        Secondary cities
                                                                                                                                     Villages
shoulder. New York has a population of 22
million, Los Angeles 18 million.6, 7
                                                                    Source: WDR 2009 team.
An evolving portfolio of places
Although the growth of cities appears
chaotic, the underlying patterns have a                             primacy” notwithstanding, the “portfolio of
remarkable order (see figure 1.2). A coun-                           places” is an enduring feature of economic
try’s urban hierarchy is characterized by                           development.
two robust regularities:                                                Settlements of different sizes complement
                                                                    one another. Metropolises, secondary cit-
•       The “rank-size rule”—the rank of a city in
                                                                    ies, market towns, and villages are all linked
        the hierarchy and its population are lin-
        early related.                                              through their complementary functions (see
                                                                    box 1.2). The primary city is often but not
•       Gibrat’s law—a city’s rate of population
                                                                    always the national administrative center and
        growth tends to be independent of its
                                                                    the seat of political power: Cambodia’s Phnom
        size.
                                                                    Penh, Cameroon’s Yaounde, and Colombia’s
    According to a special case of the rank-                        Bogotá. A country’s leading city also tends to
size rule, known as Zipf’s law, the popula-                         be its most diversified, both in the provision of
tion of any city is equal to the population of                      goods and services and in cultural and other
the largest city, divided by the rank of the                        amenities. For the cultural amenities, think of
city in question within the country’s urban                         Broadway in New York City, the Opera House
hierarchy (see box 1.1).8 As early as 1682,                         in Sydney, and the Louvre in Paris. But think
Alexandre Le Maître observed a systematic                           also of Trinidad and Tobago’s Port of Spain,
pattern in the size of cities in France.9 For                       famous for the annual carnival that attracts
all classes of country, the relative size dis-                      large numbers of visitors.
tribution has remained stable over time,                                Just as a primary city forms the core of
even as incomes and populations grew                                a country’s metropolitan area with other
(see figure 1.2). Concerns about “urban                              adjacent cities, other large urban centers or


Figure 1.2       Almost a law: relative size distributions of settlements remain stable over time

                     Low-income countries                                       Middle-income countries                                         High-income countries
Log of rank                                                 Log of rank                                                      Log of rank
6                                                           6                                                                6

5                                                           5                                                                5

4                                                           4                                                                4

3                                                           3                                                                3

2                                                           2                                                                2

1                                                           1                                                                1
          1950       1980   2005                                     1950       1980     2005                                        1950       1980   2005
0                                                           0                                                                0
    2            4          6         8       10                2           4             6             8     10                 2          4           6           8         10
                       Log of population                                           Log of population                                               Log of population
Source: United Nations 2006c.
Note: Each data point represents an agglomeration area of population size of 750,000 or more.
52                                              WO R L D D E V E LO P M E N T R E P O RT 2 0 0 9




     B OX 1.1       Two laws and a rule: the empirical regularities of a country’s city-size distribution
     The rank-size rule, discovered in 1913, can              U.S. city sizes is more even—and that the               path can have important long-term
     be expressed as the rank r associated with               rule fails to hold at the extremes of the U.S.          repercussions for the welfare of a city’s
     a city of size S is proportional to S to some            city-size distribution, a common finding                inhabitants. On whether the power in the
     negative power. The special case in which                for many countries.b Moreover, the rank-                rank-size rule equals –1, so that Zipf’s law
     the estimated power equals –1 is known as                size rule also holds for countries as diverse           holds, many researchers seem to agree
     Zipf’s law, named after a linguist, George               as Kazakhstan and Morocco, providing                    that, in general, it does not.
     Zipf. Evidence on the pervasiveness of the               further evidence of its universality (see the              The robust message from the rank-size
     rank-size rule comes not only from large                 figure below).                                          rule is that, for a given country or area, a
     cities belonging to countries of different                  Whether the rank-size rule is really a               wide range of city sizes coexists. Even the
     income classes, but also from the experi-                rule with underlying theoretical structure              most developed countries have a portfolio
     ence of individual countries. The remark-                is still under debate. It can be shown to               of settlements of different sizes, ranging
     able westward and southward expansion                    follow from Gibrat’s law, which implies                 from the small to the large, as opposed to
     of the U.S. urban hierarchy notwithstand-                that cities grow in parallel.c This is consis-          a single megacity or a collection of cities,
     ing, the rule provides a good description                tent with the absence of any systematic                 all of similar size. Agglomeration is a bal-
     of the size distribution of U.S. cities for              growth differences between cities. But                  ancing act between centripetal and cen-
     every decade between 1790 and 1950.a                     this does not imply that policy is inca-                trifugal forces. The balancing point differs
     Indeed, even today, the rank-size rule con-              pable of influencing a city’s size and                  depending on the sector, the economic
     tinues to describe well the size distribution            economic performance. Cities can and                    activities, and the type of industries.
     of U.S. cities (see figure below). This is so            do move up and down their national
                                                                                                                      Contributed by Mark Roberts.
     despite evidence that the shape of the rule              urban hierarchies as a result of good                   a. Madden 1956, cited in Kim and Margo 2004.
     has changed over time, becoming slightly                 and bad policy choices. And even transi-                b. Gabaix and Ioannides 2004, p. 14.
     flatter so that the overall distribution of              tory departures from a parallel growth                  c. Gabaix and Ioannides 2004, pp. 16–17.

     The rank-size rule, for nations as diverse as the United States, Morocco, and Kazakhstan

                    United States 2000                                            Morocco 1993                                             Kazakhstan 1993
     Log of rank                                             Log of rank                                                 Log of rank
     6                                                        3.0                                                        4.0
     4                                                                                                                   3.0
                                                             2.0
                                                                                                                         2.0
     2                                                       1.0                                                         1.0
     0                                                       0.0                                                         0.0
              13               15               17             11.5        12.5      13.5         14.5    15.5             11.5           12.5          13.5           14.5
                   Log of population                                          Log of population                                          Log of population
     Sources: The graph for the United States is from Rose (2005); the graphs for Kazakhstan and Morocco are based on data for cities and urban agglomerations from Brak-
     man, Garretson, and Marrewijk (2001).



                                                secondary cities act as regional foci for both                    private medical colleges, is a seat of learning
                                                the economy and society. For example, they                        in southern India.
                                                are the local centers for the financial sector,                        These large regional cities are connected
                                                which serve the areas around them. Düs-                           to smaller cities or major towns. The Ruhr
                                                seldorf, Hamburg, Hanover, and Munich                             area of Germany, the Randstadt area of the
                                                are all home to regional stock exchanges, as                      Netherlands, and the Padang-Medan hub in
                                                well as local concentrations of venture capi-                     Indonesia’s Sumatra represent alliances of cit-
                                                tal firms.10 Dallas and Atlanta emerged as                         ies. Smaller cities within these areas consti-
                                                regional centers of commerce and fi nance                          tute more specialized urban centers, typically
                                                in the lower South of the United States,                          focusing on manufacturing and the produc-
                                                and both host regional offices of the Fed-                         tion of traditional and standardized items.
                                                eral Reserve Bank.11 Large urban centers                          Symbiosis is the ruling order: just as the larger
                                                and secondary cities also act as local politi-                    cities help to serve the smaller cities, so the
                                                cal centers, and provide advanced public                          reverse is true. For instance, the larger cities
                                                health, education, and cultural facilities.                       depend on the smaller ones for the daily pro-
                                                Hyderabad, the state capital of Andhra                            vision of workers through commuting.12
                                                Pradesh, with numerous universities, lead-                            Just as there are mutually beneficial links
                                                ing institutes for technical education, and                       between larger and smaller cities, the same is
                                                                                                              Density                                                   53




   B OX 1.2     The Republic of Korea’s portfolio of places
   Illustrating a well-developed portfolio          manufacturing, especially standardized                      At the bottom of the hierarchy,
   of places are seven settlements in the           manufacturing, than cities farther up the                Jeongeup and Sunchang, both in the
   Republic of Korea’s urban hierarchy:             hierarchy. Although both cities serve as                 Jeonbuk province, are close to the inter-
   Seoul, Pusan, Daegu, Ansan, Gumi, Jeon-          manufacturing centers, they differ in their              face between rural and urban. So while
   geup, and Sunchang.                              specializations. Gumi is heavily specialized             Jeongeup has a relatively large popula-
       Seoul is at the pinnacle of the hierarchy.   in the radio, television, and communica-                 tion (129,050), one in four of its inhabit-
   Located 50 kilometers from the Republic          tion equipment industry, which by itself                 ants is a farmer. Likewise, Sunchang is a
   of Korea’s border with the Democratic            accounts for more than 50 percent of local               rural town: half of the 32,012 residents are
   Republic of Korea in the Han River basin,        manufacturing employment. Ansan is                       farmers. To the extent that they exhibit
   it is the country’s capital and home to          specialized in such high-tech industries as              any specialization in manufacturing, it
   a quarter of its population (that is, 9.76       electrical machinery and computers and                   is either in traditional resource-related
   million people). It serves as the nation’s       office machinery. It also has agglomera-                 industries, as in Jeongeup, or in the man-
   political center and cultural heart. Also        tions in several heavy industries: almost                ufacture of food and beverage products,
   typical is its specialization in business        14,000 workers, or 14.7 percent of the local             as in Sunchang.
   services, finance, insurance, real estate,       manufacturing workforce, are employed in
   and wholesaling and retailing. Overall,          the fabricated metal products industry.                  Contributed by Park Sam Ock.
   services account for 60 percent of the local
   economy. Seoul is also highly specialized        Seoul heads the hierarchary of settlements in the Republic of Korea
   in publishing and printing and in fashion
   design and high-end apparel, with the two
   industries employing more than half the
   city’s 465,000 manufacturing workforce.
       Next in the urban hierarchy are Pusan
   and Daegu. With a population of 3.7                                                      Seoul                                                       REPUBLIC
   million, Pusan is the Republic of Korea’s                                                                                                            OF KOREA
   second largest city. In the southeastern                                           Ansan
   corner of the Korean Peninsula, its sea-
   port, one of the world’s largest, handles
   more than 6.5 million container ships a
   year. Daegu is a metropolitan area of 2.5
   million, dominated by textile and cloth-                                                                  Gumi
   ing manufacturing and automotive parts
   manufacturing and assembly. Since 1970,                                                                          Daegu
   the Gyeongbu Expressway has connected                                             Jeongeup
   Pusan to Seoul through Daegu. About 20                                               Sunchang
   flights operate daily between Seoul and
                                                                                                                            Pusan
   Daegu, and since 2001, the two cities have
   been linked by a high-speed train.
                                                                                                                                                Population, 2007
       Much farther down the hierarchy, Ansan                                                                                                   (thousands)
   and Gumi are secondary cities, with popu-                                                                                                              > 4,000
                                                                                                                                                          1,000–4,000
   lations of around 679,000 and 375,000,                                                                                                                 500–1,000
                                                                                                                                                          150–500
   respectively. In Gyunngi province, Ansan                                                                                                               < 150
   belongs to the Seoul National Capital Area,
   as part of Seoul’s suburban area. Gumi is in
   Gyungbok province, in the southeast. As
   tends to be the case with secondary cities,
   Ansan and Gumi are more specialized in           Sources: WDR 2009 team, using data from the National Statistical Office of the Republic of Korea.




true for smaller cities and towns, and towns             scale in postsecondary education and health
and rural areas. Towns are the connective tis-           care services. Symbiosis is again the rule.
sue between rural and urban areas. They act              Towns draw sustenance from the agricultural
as market centers for agricultural and rural             activity of rural areas, but their prosperity
output, as stimulators of rural nonfarm activ-           also spills over to villages by providing non-
ity, as places for seasonal job opportunities for        farm employment opportunities. Farmers in
farmers, and as facilitators of economies of             Vietnam migrate seasonally to work in urban
54   WO R L D D E V E LO P M E N T R E P O RT 2 0 0 9



     construction, returning to invest the money        interchangeably, agglomeration, density,
     earned in their farms.13 Farmers in Makueni,       or geographic concentration of economic
     Kenya, use nonfarm income to invest in ter-        activity—across countries.
     racing, planting trees, clearing bush, building        The index identifies an area of 1 square
     houses, and educating their children. Farm-        kilometer as urban, agglomerated, or dense
     ers in the semiarid Diourbel region of Senegal     if it satisfies the following three conditions:
     have responded to growing urban demand for
     meat by diversifying away from groundnut           •   Its population density exceeds a thresh-
                                                            old (150 persons per square kilometer).
     production into animal husbandry.14
                                                        •   It has access to a sizable settlement within
                                                            some reasonable travel time (60 minutes
     Measuring density
                                                            by road).
     Measures of gross product at a refined spa-
     tial scale, such as a district or a city, are      •   The settlement it has access to is large
                                                            in that it meets a population threshold
     difficult to come by. Even for developed
                                                            (more than 50,000 inhabitants).
     countries, output estimates tend to be
     available only for rather broadly defi ned          Box 1.3 summarizes the rationale and
     subnational areas (fi rst level and adminis-        methodology underpinning the index.
     trative units, such as provinces or states). At        One advantage of the agglomeration
     this level, important variations in economic       index is that it incorporates both density
     density are likely to average out. Fortunately     and the local distance to density. Based on
     though, as illustrated earlier for Belgium,        the criteria of population density and acces-
     output and population density are closely          sibility to a sizable market, the index also
     correlated. Reliable population estimates          comes closer to providing an economic defi-
     are more easily available, even for villages       nition of an area that can both benefit from
     or townships, because in most countries, a         and contribute to agglomeration economies.
     population census is taken every decade.           Although economic density is both a cause
         The strong correlation between popula-         and a consequence of agglomeration econo-
     tion density and economic mass is consistent       mies, accessibility to this economic mass
     with urban areas being a conglomeration of         from the outer parts of the city facilitates the
     consumers and producers, of buyers and sell-       exploitation of such benefits to proximity.
     ers, and of firms and workers. For a typical        This is especially true in the service sector
     metropolitan area, the gradient of popula-         in which face-to-face interactions are often
     tion density for distance from the city center     necessary. By reducing the need to allocate
     is similar to the corresponding gradient for       valuable land area to residential uses in and
     employment density.15 As implied above, the        near urban centers, transport infrastructure
     extent to which a country’s population lives       facilitates economic density.
     in urban areas bears a strong relationship to          Going to work by car or by high-speed
     how “bumpy” its economic geography is.             public transportation is a luxury that devel-
     Density goes from smoothly spread out to           oped country commuters do not always
     quite uneven as a country develops. Urban-         share with their counterparts in developing
     ization is thus synonymous with a tendency         countries. For any given geographic dis-
     toward greater agglomeration within a coun-        tance, therefore, accessibility to a city tends
     try. A country’s urban share is a good proxy       to be lower in developing countries because
     for the proportion of its population living in     of the need to rely on alternative, more time-
     areas of high density and, therefore, for the      intensive modes of transportation, such as
     “bumpiness” in its economic geography.             walking, cycling, or inefficient public trans-
         This Report proposes the use of an             portation operating on poor-quality roads.
     agglomeration index computed using geo-            In Mumbai, India, 44 percent of people walk
     graphic information systems as a measure           to work,16 and in Hefei City, China, more
     of density. Measures of urbanization are           than 70 percent either walk or cycle.17
     nonuniform across countries, which makes               Such variations in accessibility deter-
     comparability and aggregation a challenge.         mine both the shape and form of a city.
     The index allows for a more consistent com-        When most people walk to work, a city is
     parison of the level of urbanization—or,           more likely to be monocentric and densely
                                                                                                                    Density                                             55




   B OX 1.3         Computing the agglomeration index
   The United Nations maintains the World                            adjusted for purchasing power differ-            settlement center is calculated based
   Urbanization Prospects database, a trea-                          ences between countries—is that it               on the maximum travel time to the
   sure trove of information. It provides                            allows international comparisons and             center.
   urban shares and population data for                              calculations that aggregate poverty for        • Create population density grids. These
   229 countries stretching back to 1950.                            regions and the world. The agglomera-            are created at a 1-kilometer spatial
   But these data are based on country                               tion index allows the same comparisons           resolution using two global grid-based
   definitions, which can be quite different.                        and aggregation.                                 population data sources, GRUMP and
   This Report proposes a new measure of                                The methodology underlying the cal-           LandScan.b
   agglomeration, based on a uniform defi-                           culation of the agglomeration index can        • Identify the areas. Identify the grid cells
   nition of what constitutes an “urban” or                          be summarized as follows:                        that satisfy thresholds for all three criteria.
   agglomerated area, using the technique
                                                                     • Specify thresholds. To be classified as      • Aggregate grid cell populations. The
   outlined in Chomitz and others (2007) and
                                                                       “urban” using the agglomeration index,         result is analogous to urban popula-
   elaborated in Uchida and Nelson (2008).
                                                                       an area must satisfy three criteria based      tion. The proportion of this number to
      This should not be read as implying
                                                                       on (1) minimum population size used            that country’s total population is the
   that World Urbanization Prospects data
                                                                       to define a sizable settlement, (2) mini-      agglomeration index, a summary mea-
   are flawed. A better interpretation is to
                                                                       mum population density, and (3) maxi-          sure of the proportion of the popula-
   see the challenge of measuring urbaniza-
                                                                       mum travel time, by road, to the sizable       tion living in areas of high density.
   tion as analogous to the measurement
                                                                       settlement.
   of poverty. Each country has its own                                                                             In calculating the index, this Report uses
   poverty line and criteria to track changes                        • Locate the centers of sizable settlements.   a base case set of thresholds of 50,000 for
   in national poverty rates. But these mea-                           This mapping is done for cities that         minimum population size of a settlement,
   sures do not allow reliable comparisons                             meet the minimum population size             150 people per square kilometer for pop-
   of poverty between countries, and they                              criterion using data from the Global         ulation density, and 60 minutes for travel
   cannot be used to aggregate poverty                                 Rural-Urban Mapping Project (GRUMP)          time to the nearest large city.
   for groups of countries. The merit of                               human settlements database.a                    The density and travel time thresh-
   a uniform poverty measure—such as                                 • Determine the sizable settlement’s bor-      olds are those employed in Chomitz,
   those living below US$1 or US$2 a day,                              der. The border surrounding a sizable        Buys, and Thomas (2005). The density
                                                                                                                    threshold is the same as the one used
                                                                                                                    by the Organisation for Economic Co-
   The internationally comparable agglomeration index can yield different urban shares than those                   operation and Development (OECD). The
   from country-specific definitions
                                                                                                                    threshold of 50,000 for a sizable settle-
   Urban share (%)                                                                                                  ment is reasonable for developing and
   100                                                                                                              developed countries. Many developing
                           Country definition                                                                       nations have more than 10 percent of
    90
    80                                                                                                              their total population in urban centers
                              Agglomeration Index                                                                   of between 50,000 and 200,000. Some
    70
                                                                                                                    examples include Chile in 2002, Brazil
    60
                                                                                                                    in 2000, and Malaysia in 2000, all with
    50                                                                                                              around 17 percent of their national
    40                                                                                                              population living in urban centers of
    30                                                                                                              50,000–200,000 inhabitants. Of India’s
    20                                                                                                              urban population in 2001, 20 percent
    10                                                                                                              lived in settlements of this size.
     0                                                                                                                 According to the World Urbanization
                                                                                                                    Prospects database, the worldwide urban
                a

                      il

                              a

                                         a

                                               sh


                                                            a

                                                                 a

                                                                                                 a

                                                                                                m

                                                                                                of

                                                                                              sia


                                                                                                 n

                                                                                                          y
                    az




                                                                                                         ke
             tin




                             ol

                                    nd




                                                         nk


                                                                di

                                                                                              in




                                                                                             tio
                                                                                            na
                                             de




                                                                                             p.

                                                                                           ni
                              g




                                                                In

                                                                        Ch
                    Br




                                                                                                                    share in 2000 was 47 percent. Using the
                                                                                                          r
                                                    La
           en




                                     a




                                                                                          ra
                           An




                                                                                          et

                                                                                        Re




                                                                                                       Tu
                                                                                        Tu
                                             la
                                  Ug




                                                                                      de
            g




                                                                                       Vi
                                                     i
                                       ng

                                                  Sr
         Ar




                                                                                     ab




                                                                                    Fe




                                                                                                                    base case criteria, this ratio is 52 percent,
                                     Ba




                                                                                  Ar




                                                                                  n
                                                                               ia
                                                                                t,




                                                                                                                    but using 100,000 as the minimal settle-
                                                                              yp




                                                                            ss
                                                                             Eg




                                                                         Ru




                                                                                                                    ment size, it is 44 percent, according to
   Sources: Chomitz, Buys, and Thomas 2005; Nelson 2008; Satterthwaite 2007; United Nations 2006c.                  the agglomeration index. But country
   a. The GRUMP human settlements database was developed by the Center for International Earth Science
   Information Network (CIESIN) at Columbia University (http://sedac.ciesin. columbia.edu/gpw/index.jsp).           level estimates can be further apart (see
   b. LandScan was developed by Oak Ridge National Laboratory (http://www.ornl.gov/sci/landscan/).                  figure at left).


populated at its core. In Mumbai, half of all                            agglomeration in industrial districts, work-
workers commute less than 2 kilometers,                                  ers in nineteenth-century Britain had to live
implying that they live close to their places of                         nearby. The centers of industrial towns were
work. Similarly, to obtain the advantages of                             densely populated, and overcrowded housing
56   WO R L D D E V E LO P M E N T R E P O RT 2 0 0 9



     was common. Not until the electric tram was            redefi ned as 15,000 (Nigeria and Syria,
     introduced did this change.                            for example, have cutoffs of 20,000), that
         In determining accessibility, and thus the         share would drop to 67 percent.
     shape and form of cities, features of physical     •   Mauritius. In 2000 about a quarter of
     geography can also be important. Manhattan             Mauritius’s population lived in settle-
     Island in New York City is difficult to get to,         ments with between 5,000 and 20,000
     simply because of geography, so it has sky-            inhabitants. Some of these settlements
     scrapers and a classic monocentric structure,          are district capitals, but none of them
     with half its employment within a three-mile           are classified as urban. If they were, the
     radius of Wall Street. By contrast, in Los             urban share would have been more than
     Angeles, one has to widen the area to a radius         two-thirds rather than less than half.
     of 11 miles from the center to find as large
     a share of employment.18 The implication:              At a regional level, according to World
     economic density in New York City is $1.44         Urbanization Prospects data, South Asia
     billion of gross product per square kilometer,     poses the paradox of being the least urban-
     in Los Angeles it is $0.49 billion.19              ized region (27 percent urban) in the world
         In the United Kingdom, Stevenage, Basil-       while also the most densely populated. Using
     don, and Crawley are commuter towns that           the agglomeration index, South Asia’s urban
     serve London. About 11 percent of Lon-             share in 2000 was 42 percent, making it more
     don’s GDP is generated by commuters from           urbanized than both Sub-Saharan Africa
     suburban areas.20 Similarly, in the United         and East Asia and the Pacific (figure 1.3).
     States, a daily tide of workers commute into       The World Urbanization Prospects also pose
     Washington, D.C., from the neighboring             a puzzle for Latin America and the Carib-
     states of Maryland and Virginia. In 2005 the       bean. The urban share in this region in 2000
     net contribution of commuters from these           was greater than that in Eastern Europe and
     two states to Washington, D.C.’s output            Central Asia and almost on par with the
     was $36.4 billion. Maryland’s Montgomery           OECD’s. The OECD has an average GDP per
     County—within easy commutable distance             capita more than six times that of the aver-
     of the district—alone contributed $6.4 bil-        age Latin American country. More reason-
     lion to Washington’s gross product.21              ably, the agglomeration index indicates that
         The biggest advantage of the agglomera-        Latin America and the Caribbean’s urban
     tion index is its comparability across coun-       share in 2000 was similar to that of Eastern
     tries. Here the index has an advantage over        Europe and Central Asia, and 15 percentage
     the United Nations’ World Urbanization             points lower than that of the OECD.
     Prospects database, which contains the “de             Despite these drawbacks, the World
     facto population living in areas classified         Urbanization Prospects data are the only avail-
     as urban according to the criteria used by         able information for comparisons over time.
     each area or country.”22 The heterogeneity         The agglomeration index is available only for
     across countries can makes cross-country           2000, because time-series data on road net-
     comparisons misleading. A few examples:            works, necessary to estimate travel time, are
                                                        not readily available. So, the agglomeration
     •   India. With the criterion for an urban         index and World Urbanization Prospects data-
         area used by Zambia or Saudi Arabia,           base should be considered as complementary
         defined as settlements with populations         data sources for examining urbanization
         of 5,000 or more, the share of India’s         and density, and this Report uses both the
         population in urban areas in 1991 would        agglomeration index and the World Urbaniza-
         be 39 percent instead of the official figure     tion Prospects data.23 Calculating comparable
         of 26 percent. This is because 113 mil-        urban share measures for at least some coun-
         lion inhabitants of 13,376 villages would      tries in the past is possible; going forward, it
         be reclassified as urban.                       should be a priority for all countries.
     •   Mexico. Based on Mexico’s official cri-
         terion of settlements of 2,500 or more         Economic concentration—
         as urban, the country’s urban share in         the richer, the denser
         2000 was 74.4 percent. But if the settle-      In the early stages of development, when an
         ment population threshold were to be           economy is primarily agrarian, people live
                                                                                                    Density                                        57



spread out on farmland. Even the largest         Figure 1.3   The agglomeration index helps to compare urbanization across regions
towns and cities are small. Urban settlements    Urban share (%)
are likely to be small port cities and market    100
towns, serving the rural needs and trading        90
surpluses of agriculture. Industrialization       80
brings with it a rapid process of urbaniza-       70
tion—new cities are born, and existing cities           Agglomeration Index
                                                  60
expand. As people crowd into these cities at
                                                  50
a faster rate than their boundaries expand,              United Nations
                                                  40
population and economic density increase.
Quite early in a country’s development, this      30

leads to a hierarchy of places.                   20
   So, two transitions characterize eco-          10
nomic development. The first involves              0
                                                       Sub-Saharan   South       East Asia Middle East   Latin         Europe &    Other      OECD
the movement from a primarily agrarian                    Africa      Asia       & Pacific  & North America &           Central high-income countries
economy to a much more manufacturing-                                                        Africa    Caribbean         Asia    economies
oriented economy. The second transition,         Sources: Chomitz, Buys, and Thomas 2005; Nelson 2008; Satterthwaite 2007; United Nations 2006c.
taking place at a much higher level of devel-
opment, involves the transformation to a         for disproportionate shares of their national
service-oriented economy. The first phase         GDP. In 2005, Mexico City contributed 30
of urbanization, which occurs at a faster        percent of Mexico’s GDP despite occupying
rate, coincides with the transition from         only 0.1 percent of its land. Luanda contrib-
a rural to an urban economy. The second          uted a similar share of Angola’s GDP, while
phase of urbanization, at a slower rate and a    occupying 0.2 percent of its land. Like-
much higher level of development, is linked      wise, the largest cities in Hungary, Kenya,
to a within-urban evolution. In most coun-       Morocco, Nigeria, and Saudi Arabia—
tries, these transformations happen at the       Budapest, Nairobi, Casablanca, Lagos, and
same time but in different areas.                Riyadh—contributed about 20 percent of
   To measure concentration, we have to          their country’s total GDP while taking up
defi ne an area. The policy debate often          less than 1 percent of land.25
involves a discussion of urban primacy,              Density, defined as GDP in purchasing
such as whether developing country cities        power parities per square kilometer, rises
are too big or too small. More academic          with the level of development, and the dens-
discussions use a purer geographic notion        est places in the world are in the richest coun-
of space. This chapter uses both spatial         tries. Dublin, London, Paris, Singapore, and
units—primary cities and the densest grid        Vienna ranked at the top, in 2005, with more
cell of 1° longitude by 1° latitude of a coun-   than $200 million in gross product per square
try—to measure concentration.                    kilometer. Likewise, Tokyo-Kanagawa, New
                                                 York–New Jersey, Oslo–Akershus-Vestfold,
Historically, rapidly rising                     and Vienna-Mödling were the densest grid
concentration, then a leveling off               cells of 1° longitude by 1° latitude, generating
By one definition, a city is a geographic area    more than $30 million of gross product per
characterized by a concentration of eco-         square kilometer (figure 1.4).
nomic actors.24 Globally, the top 30 cities,         A century of data on aggregate urban
ranked by GDP, generated around 16 per-          shares, and two centuries of population
cent of the world’s output in 2005, while the    estimates for primary cities, suggest that
top 100 generated almost 25 percent. The         urbanization is initially rapid before slowing.
urban agglomerations of Tokyo and New            Developing countries—especially those in
York have estimated GDPs (in purchasing          Africa and Asia—are at phases during which
power parity) broadly similar to those of        urban shares increase sharply. People in
Canada and Spain, respectively, whereas          Western Europe and North America, which
London has a higher estimated GDP than           went through the same phase a century ago,
either Sweden or Switzerland. Similarly, pri-    have understandably forgotten. Emerging
mary cities in developing countries account      economies such as the Republic of Korea that
58                                                   WO R L D D E V E LO P M E N T R E P O RT 2 0 0 9



Figure 1.4   The richer a country, the more concentrated its economic mass

                              a. Primary city                                                                            b. Area of 1° longitude by 1° latitude
Gross product (US$ millions) per km2                                                            Gross product (US$ millions) per km2
300
                                                                  London
                                          Singapore                                              90
250
                      Seoul        Madrid                          Dublin                        70                                                               Tokyo-Kanagawa
200                                                  Vienna

150                                                      Paris                                   50                       New York–New Jersey

                                                                                                                             Ontario                   Oslo–Akershus-Vestfold
100
                                                                                                 30                                                Vienna-Mödling

 50
                                                                                                 10
  0
      0       5         10          15           20           25               30
                                                                                                –10
              GDP per capita (constant US$, thousands)                                                0             10                 20          30               40             50
                                                                                                                     GDP per capita (constant PPP US$, thousands)
Sources: WDR team estimates based on World Bank (2007j), and databases from www.citymayor.com and www.gecon.yale.edu.



Figure 1.5 Developing countries have a pace of urbanization similar to that of early                                          developed rapidly provide the best case stud-
developers
                                                                                                                              ies for understanding the pace and pattern of
                                a. At magnified scale: GDP per capita < $10,000                                               geographic concentration. Their experience
Urban share (%)
100                                                                                                                           traces the initially rapid and the more grad-
 90                                                                                                                           ual growth of today’s wealthiest nations.
 80
                                                                                                                                  At the aggregate level, using the popula-
 70
                                                                                                                              tion shares in urban areas, the urbanization
 60
                                                                                                                              pattern of developing countries in Asia,
                                                                                                                              Africa, Middle East, and Latin America
 50
                                                                                                                              over the last 50 years closely tracks the first
 40
                                                                                                                              part of the historic path earlier traversed by
 30
                                                                                                                              OECD countries between 1900 and 2000
 20
                                                                                                                              (figure 1.5). The urbanization in Asia mir-
 10
                                                                                                                              rors the rapid phase of urbanization that
  0
      0       1         2           3           4             5            6             7        8         9        10       OECD countries experienced in the nine-
                             GDP per capita (1990 int’l Geary-Khamis $, thousands)                                            teenth century. Likewise, the geographic
                                                                                                                              transformations in Latin America and the
                                         b. Full range of GDP per capita
Urban share (%)                                                                                                               Caribbean, in Eastern Europe and Central
100                                                                                                                           Asia, and in the Middle East and North
 90                                                                                                                           Africa are qualitatively similar to those
 80                                                                                                                           experienced by the OECD in the first phase
 70                                                                                                                           of urbanization. Quantitatively, the urban
 60                                                                                                                           shares for Latin America and the Carib-
 50                                                                                                                           bean and for Eastern Europe and Central
 40                                                                                                                           Asia regions are higher than those for the
 30                                             OECD                                Middle East & North Africa
                                                                                                                              OECD at comparable incomes.
 20                                             East Asia & Pacific                 Latin America & Caribbean                     This may, however, be an artifact of the
                                                South Asia                          Eastern Europe & Central Asia             data. Data from the World Urbanization
 10                                             Sub-Saharan Africa
  0
                                                                                                                              Prospects database systematically overstate—
      0                 5                       10                     15                        20                  25       purely as a definitional matter—the urban
                         GDP per capita (constant int’l Geary-Khamis $, thousands)                                            shares of Latin America and the Caribbean,
Sources: Maddison 2006; United Nations 1969; United Nations 1949; United Nations 1952; Historical Database                    Eastern Europe and Central Asia, and Sub-
of the Global Environment; United Nations 2006c.                                                                              Saharan Africa. The safest conclusion may
                                                                                                                              be that the pattern of urbanization—the
                                                                                                                         Density                                                59




     B OX 1.4      Africa’s urbanization reflects industrialization
     Between 1970 and 1995, the urban popula-                 rate population estimates, a population of at                in total GDP—a doubling of their
     tions in Sub-Saharan Africa were growing                 least 1 million in 1995, and data on sectoral                economies—also witnessed the fastest
     at 5.2 percent a year while their GDP per                value added for 1970 and 1995.                               growth in urban population—a four-
     capita was shrinking at 0.66 percent a year.                This whittles the sample down to just 10                  fold increase. The leaders in the sample
     Since the work by Fay and Opal (2000),                   countries: Benin, Botswana, Central Afri-                    were Benin and Zimbabwe.
     many have argued that urbanization does                  can Republic, Ghana, Mauritania, Niger,                    • The pace of urbanization was positively
     not necessarily accompany development,                   Rwanda, Senegal, Zambia, and Zimbabwe.                       correlated with growth in industries
     with Sub-Saharan Africa in mind (Com-                    Of these 10 countries, five experienced                      and services, activities predominant in
     mission for Africa 2005). But Satterthwaite              conflict at least once, and the other five                   urban areas.
     (2007) questions the validity of the urban               were peaceful throughout the period. The
     population numbers in most studies. Since                results do not appear to differ systemati-                 These patterns do not support the claim
     many were based on projections, some                     cally between these two sets of countries.                 of African urbanization without growth.
     may have been grossly overestimated.                     The main findings follow:                                  In contrast, countries with higher GDP
        The problem is the lack of regular popula-                                                                       growth experienced faster urbanization,
     tion censuses. For Chad and Eritrea the pop-             • Except for Botswana, the countries                       and rapid urbanization came hand-in-
     ulation projections spanning 1950 through                  experienced on average a doubling                        hand with higher growth in industries
     2030 were based on one population census.                  of population, but only 60 percent                       and services. A counterfactual of an Africa
     Those for the Democratic Republic of Congo                 cumulative growth in GDP. Population                     without urbanization is one with even
     were derived from two observations, the                    growth outpaced increases in gross                       slower economic growth, greater GDP per
     most recent for 1984. It is thus reasonable                value added, and GDP per capita fell.                    capita losses, and increases in poverty.
     to consider only countries with at least                 • Urban population growth and total
     two censuses during the period examined                    GDP growth are positively correlated.                    Sources: Fay and Opal 2000; Satterthwaite
     (1970–95), a census post-2000 for more accu-               Countries with the fastest growth                        2007; United Nations 2006c.



relationship between economic growth and                              At a disaggregated level, the primary
urbanization—is not unprecedented. Even                            city’s population share of a country dis-
in Sub-Saharan Africa, faster urbanization                         plays a similar, nonlinear pattern of initially
between 1970 and 1995, albeit with negative                        rapidly rising concentration, followed by a
GDP per capita growth, was associated with                         subsequent leveling (figure 1.6). This inten-
higher total GDP growth. Urbanization also                         sification of economic mass within a coun-
came hand-in-hand with rapid growth in                             try’s largest cities is seen for a wide range
industries and services (see box 1.4).                             of incomes, from Budapest, Cairo, Kuala


Figure 1.6    Density intensifies rapidly in the early phase of urbanization before leveling off

% city population to                                                                          % city population to
national population                                                                           national population
                                                Santiago, 1800–2000
35                                                                                            35
                                                   Athens, 1800–2000
30                                                                                            30        Vienna, 1800–2000
                                                      Lisbon, 1800–2000
25                                                                                            25                Dublin, 1800–2000
                                                          Seoul, 1800–2000
20                                                                                            20                                                              Sydney, 1800–2000
                                      Budapest, 1850–2000                                                                                 Toronto, 1800–2000
15                                                                                            15
                  Cairo, 1800–2000                                                                                                                            Zurich, 1800–2000
10                            São Paolo, 1850–2000                                            10
                                                                                                                                                           Brussels, 1800–2000
                                           Kuala Lumpur, 1900–2000
 5                                                                                              5
                                       Warsaw, 1850–2000
 0                                                                                              0
     0                    5                      10                       15                        0           5             10           15            20            25
                    GDP per capita (constant int’l $, thousands)                                                    GDP per capita (constant int’l $, thousands)
Sources: WDR 2009 team estimates, based on the Staff City Population Database, Human Settlements Group, International Institute for Environment and Development (IIED). Data
from 1950 to the present, primarily from United Nations (2006c); data before 1950, primarily from Chandler and Fox (1974), Chandler (1987), and Showers (1979). Latin America drew
on a review of 194 published censuses.
60                                              WO R L D D E V E LO P M E N T R E P O RT 2 0 0 9



                                                Lumpur, and Warsaw to Athens, Lisbon,                             urban share and development holds until
                                                Santiago, and Seoul. These evolutions have                        a GDP per capita of around $10,000. This
                                                also been observed in Brussels, Dublin, Syd-                      incipient urbanization is associated with a
                                                ney, Toronto, Vienna, and Zurich over the                         rapid shift in the number of people moving
                                                two centuries since 1800.                                         from rural to urban areas. Subsequently,
                                                                                                                  the pace of urbanization slows and density
                                                Again today, rapidly rising                                       levels off as the urban share surpasses 60
                                                concentration, then a leveling off                                percent, and the level of GDP per capita
                                                A similarly shaped pattern reappears in con-                      surpasses $10,000. With only a handful of
                                                temporary comparisons between a country’s                         exceptions, countries with GDPs per capita
                                                level of development and the concentration of                     above $25,000 have an agglomeration index
                                                density. During 2000–05, the average urban                        above 70 percent.
                                                population growth for low-income countries                            Administratively defined areas. Tak-
                                                was 3 percent a year—faster than upper-                           ing individual cities as the geographic
                                                middle-income countries at 1.3 percent and                        unit, a positive concave relationship exists
                                                high-income countries at 0.9 percent. The                         between a country’s level of development
                                                relationship is robust. It holds for a variety                    and its primacy—the share of urban popu-
                                                of concentration measures, ranging from the                       lation living in the country’s primary city, a
                                                agglomeration index, to population, gross                         widely used concentration measure. Similar
                                                product, and household consumption den-                           to the relationship between agglomerations
                                                sity. It is robust to geographic scale: an area                   and the level of development, primacy also
                                                of 1 square kilometer, a city, a grid cell of 1°                  rises rapidly before stabilizing during the
                                                longitude by 1° latitude, and an aggregated                       latter stages of urbanization (see figure 1.8,
                                                urban sector.                                                     panel a). Population and output density are
                                                    Local 1-square kilometer areas. Esti-                         highly correlated, but population density
                                                mated agglomeration indexes produce a                             understates the geographic concentration
                                                pattern similar to the historical time series:                    of economic mass. Agglomeration econo-
                                                rapidly rising density for countries during                       mies, the benefits that fi rms and workers
                                                the early phase of urbanization (figure 1.7).                      enjoy as a result of proximity, make it likely
                                                This strong positive relationship between                         that output density will increase more than
                                                                                                                  proportionately with employment or popu-
                                                                                                                  lation density.
Figure 1.7 Shares of population living in urban agglomerations rise with the level of                                 1° longitude by 1° latitude. Using the
development
                                                                                                                  terrestrial grid cells to estimate concentra-
Agglomeration index                                                                                               tion as the share of the densest cell’s gross
1.0                                                                                                               product in the country’s GDP, concentra-
       Egypt, Arab Rep. of                                                    Japan
                                                                                                                  tion of economic mass rises rapidly among
0.8                                        Korea, Rep. of                                    United States        countries with a GDP per capita of less than
                                                                                                                  $15,000, and then stabilizes and tapers off
                                                                                                                  among higher-income countries (see figure
0.6                                                                                                               1.8, panel b).
                                  Brazil
      India                                                                                                           Urban areas of countries. Concentra-
                                 South Africa
0.4                                                                                          Norway               tion measured by consumption, rather
                                                                                                                  than by population or GDP, suggests the
                 China
                                                                                                                  same concave relationship with the level of
0.2                                                                                                               development. For instance, the urban shares
                Ethiopia
                                                                                                                  of household consumption in Malawi and
                                                                                                                  Cameroon at GDPs per capita of $150 and
 0
        0            5         10          15          20         25         30                  35          40   $700, respectively, are 36 percent and 48
                             GDP per capita (PPP, constant 2000 US$, thousands)                                   percent. At about 63 percent, the shares are
Sources: Calculated by WDR 2009 team using Nelson (2008) and World Bank (2006g).                                  higher for Jordan and the Arab Republic of
Note: The size of each circle indicates the population size of that country. PPP = purchasing power parity. The   Egypt with GDP per capita of around $1,600,
agglomeration index uses the following criteria: density of 150 persons per kilometer or more, access time of
60 minutes or less to a sizable settlement, defined as one that has a population of more than 50,000.             and rise to 80 percent in Panama and Poland
                                                                                                                        Density                                              61



Figure 1.8 Geographic concentration of population, gross product, and household consumption rises sharply with development, then levels off
Cross-country evidence, late 1990s and 2000s
                      a. Population                                             b. Economic mass                                               c. Consumption
                     Spatial unit: city                     Spatial unit: grid cell of 1° longitude by 1° latitude                   Spatial unit: aggregated urban areas

                                                          Gross product in densest area                                  Urban share (%) of
% urban population in largest city                        as % of country’s total GDP                                    household consumption
50                                                        50                                                             100

40                                                         40                                                             80

30                                                         30                                                             60

20                                                         20                                                             40

10                                                         10                                                             20

 0                                                          0                                                              0
     0           10         20         30         40            0          10          20          30        40                0      1     2       3     4     5     6     7
         GDP per capita (constant US$, thousands)                   GDP per capita (constant US$, thousands)                       GDP per capita (constant US$, thousands)
Sources: WDR 2009 team estimates, based on World Bank (2007j), Nordhaus (2006), and more than 120 household surveys for more than 75 countries.



with GDPs per capita of $3,500 and $5,000,                          will approximate a 50/50 urban-rural split.
respectively (see figure 1.8, panel c).26                            During more advanced urbanization—now
                                                                    a within-urban transformation in a postin-
A portfolio of bigger and denser places                             dustrial area—the distribution of popula-
It follows from these stylized facts of geo-                        tion can be approximated as 75 percent
graphic transformation that high-income                             urban and 25 percent rural.
countries have a portfolio of places with a                             This generalization corresponds well to
higher proportion of large settlements and                          the experience of the United States. In 1690,
a lower proportion of small settlements                             when the average GDP per capita was a mere
than do middle-income countries. And the                            $500 (1990 international dollars),27 the pri-
middle-income countries have a signifi-                              mary city in colonial British America was
cantly higher proportion of medium-size                             Boston. With a population of 7,000, how-
settlements than do low-income countries.                           ever, Boston was by modern-day standards
In low-income countries, about three-                               little bigger than a small town. In the urban
quarters of the population live in small                            hierarchy, only three other cities had popu-
settlements of less than 20,000 people,                             lations greater than 2,500, two of them New
and only 10 percent live in urban agglom-                           York and Philadelphia. The early phase of
erations of more than 1 million people. In                          American industrialization brought with it
high-income countries, the opposite is true.                        an increase in the urban share from 7 per-
Less than a quarter of the population live in                       cent in 1820 to 20 percent in 1860, as GDPs
small settlements of less than 20,000 peo-                          per capita rose from $1,257 to $2,170 (1990
ple, and about half of the population live in                       international dollars). During this time, the
settlements of more than 1 million people                           population of the primary city, now New
(see table 1.1).                                                    York, expanded from 123,706 to 805,651. Its
    At an incipient stage of urbanization,                          rapid growth allowed the urban hierarchy
the portfolio of places in a small country                          to expand and stretch out.
or part of a larger country, such as a prov-
ince or even a large district, can be approxi-                      Table 1.1   The size of urban settlements grows with development
mated as 75 percent rural and 25 percent
                                                                                                                     Low-income          Middle-income        High-income
urban, all settlements of relatively low den-                         Population size                                countries (%)       countries (%)        countries(%)
sity. As urbanization accelerates—still pre-                          Small settlements: less than 20,000                 73                    55                  22
dominantly a rural-urban transformation
                                                                      Medium settlements: 20,000 to 1 million             16                    25                  26
driven by industrialization—and the area
or province grows toward a GDP per capita                             Large settlements: more than 1 million              11                    20                  52
of $10,000, its distribution of settlements                         Source: World Bank 2007j.
62   WO R L D D E V E LO P M E N T R E P O RT 2 0 0 9



         The number of cities with a population         and public health facilities in urban areas.
     greater than 1 million increased from just         Along with diverging wages, this promotes
     one, New York, in 1820 to nine in 1860. All        divergence in more basic measures of wel-
     these cities were in the Northeast, where          fare between urban and rural areas.30 But
     industrialization began. As the geographic         rural-urban disparities begin to narrow as
     transformation wore on, and the United             the urbanization process slows, and gov-
     States completed its transition to a mature        ernments become more capable. The exo-
     industrial economy, population density in a        dus of people and workers from rural areas
     consistent sample of U.S. cities with popu-        to towns and cities reduces surplus labor
     lations greater than 25,000 increased from         from the land in agriculture—and reduces
     7,230 persons per square mile to 8,876 per         competition between workers in rural labor
     square mile. The average land area of a city       markets. And labor-saving technological
     increased from about 19 square miles to 40         progress releases labor for migration to
     square miles.28 Cities became more packed          urban areas and improves productivity. In
     and more sprawling at the same time.               time, investments and fiscal redistributions
                                                        give rural residents better local access to
     Convergence—rural-urban and                        basic amenities, such as a clean daily source
     within cities                                      of running water, sanitation, and electricity,
     A “bumpy” economic geography distributing          as well as schooling and health care. Indeed,
     production and people unevenly across the          with development and the passage of time,
     space in a country is a natural feature of the     a country’s economic geography approxi-
     working of a market economy. This bumpi-           mates a “natural” balance that equalizes
     ness tends to become more pronounced as a          welfare between urban and rural residents.
     country develops. The question often asked         In this situation, people choose to live where
     is: what does this do to the geographic distri-    they expect to be best off in material and
     bution of poverty, consumption, and other          nonmaterial well-being. The Islamic Repub-
     living standards? The answer can determine         lic of Iran illustrates this rural-urban con-
     the political and social sustainability of the     vergence (see box 1.5).
     process of concentration.                              Evidence from today’s industrial coun-
                                                        tries suggests that development has largely
     Rural-urban disparities in well-being—             eliminated rural-urban disparities. High
     first wide, then narrow                             urban shares and concentrated economic
     Rural-urban disparities in productivity,           density go hand in hand with small differ-
     wages, and well-being can be expected to be        ences in rural-urban well-being on a range
     large and increasing in the earlier stages of      of indicators. The 15 countries that joined
     development. With the rapidly increasing           the European Union (EU) before 2004, all
     concentration of economic mass in a coun-          with GDPs per capita in excess of $13,000
     try’s towns and cities in the earlier stages of    (1990 international dollars), consider the
     development, significant disparities in pro-        unemployment rate an important policy
     ductivity, wages, and basic welfare occur          target.31 But rural-urban unemployment
     between urban and rural areas. The agglom-         differences should not be a concern. The
     eration of capital, consumers, and workers         unemployment rates are 10.1 percent for
     quickly brings production advantages, and          urban areas, and 9.9 percent for rural areas.
     transport costs restrict the benefits to the        This is also evident for youth: 19.4 percent
     locality. These larger local markets enable        in urban areas compared with 18.7 per-
     firms to spread the fixed costs of production        cent in rural areas. The rates of labor force
     across a wider number of consumers, pro-           participation in urban and rural areas are
     ducing cost and productivity advantages.29         68.3 and 69.4 percent, respectively.32 For
     This means higher wages in towns and cities,       England, the high degree of rural-urban
     and greater availability of a more diversified      equality in well-being is reflected in similar
     range of goods and services.                       disposable incomes: indeed, at £522, weekly
        The concentration of mass also helps to         disposable income in villages is 10 percent
     ensure a better supply of basic infrastructure     higher than the £476 in cities.33
                                                                                                                           Density                                       63




       B OX 1.5         Urbanization and narrowing rural-urban disparities in the Islamic Republic of Iran
       Rural-urban disparities have narrowed                    • First, the share of the urban population                   the lagging provinces. Between 1976
       in the Islamic Republic of Iran. In 1976,                  has increased from 49 to 67 percent                        and 1996, the female literacy rate rose
       on the eve of the Iranian revolution, the                  between 1979 and 2005. This is a con-                      from 17 to 62 percent, while for urban
       mean per capita household income in                        tinuation of a longer-term trend: the                      women it rose from 56 to 82 percent.
       rural areas was 44 percent of that in urban                urban population had grown by 5.4                          During 1994–2000, infant mortality
       areas. By 2005, it had increased to 63                     percent per year (and in Tehran by 6                       and under-5 mortality fell fastest in the
       percent.                                                   percent) between 1966 and 1976.                            poorest provinces.
          The Shah’s government favored cities                                                                             • Finally, overall poverty has fallen. The
                                                                • Second, the rural-urban gap in house-
       over the countryside. Price controls for                                                                              national poverty rate was at 8.1 percent
                                                                  hold incomes has narrowed. Between
       essential foods depressed agricultural                                                                                in 2005, with relatively modest differ-
                                                                  1976 and 1984, agricultural value added
       incomes. High tariffs, import bans, and                                                                               ences in rural and urban poverty of 10
                                                                  grew by 31 percent, twice the rate of
       licensing for industrial goods propped                                                                                and 7.1 percent, respectively. But pov-
                                                                  the nonoil economy. One reason for this
       up prices of manufactured goods and                                                                                   erty rates still vary a lot between prov-
                                                                  growth was that farmgate prices rose 55
       depressed farmers’ purchasing power.                                                                                  inces, ranging from 1.4 to 23.3 percent.
                                                                  percent. Another reason was that more
       An inward-looking development strategy
                                                                  was spent on projects to increase the                    The political commitment to spatial
       oriented toward final domestic demand
                                                                  productivity of small and medium-size                    equity has produced mixed outcomes
       amplified internal migration to Tehran
                                                                  farms. Growth could also be attributed                   during the last 30 years: overall poverty
       and a few other large cities. For every
                                                                  to the fact that agricultural production in              declines and a convergence in rural-
       indicator of development, the center per-
                                                                  the Islamic Republic of Iran is dominated                urban standards of living, but persistent
       formed far better than the periphery. In
                                                                  by the private sector, whereas large                     differences in interprovincial living stan-
       1973, the poverty rate was 23 percent in
                                                                  industrial enterprises and service provid-               dards.
       the central region and 42 percent for the
                                                                  ers were nationalized after the revolu-
       country. This spatial inequality matched
                                                                  tion, which hindered their efficiency.
       the nation’s ethnic map, fueling tensions.
          What has happened since the commit-                   • Third, rural and urban human devel-                      Based on a contribution by Anton Dobro-
       ment in 1979 to address spatial disparities?               opment indicators improved, even in                      nogov, Alexander Kremer, and others.



   For 21 of the 30 OECD countries, the                             three times higher. But for OECD countries
higher the GDP per capita in 2003, 34 the                           with average GDPs per capita above $10,000,
lower the ratio of GDP per capita in predom-                        the ratio is between one and two (except for
inantly urban areas to that in rural areas                          Norway). Given the well-developed fiscal
(see figure 1.9).35 For the Czech Republic,                          redistribution mechanisms in OECD coun-
Hungary, Poland, the Slovak Republic, and                           tries, and differences in age-demographic
Turkey, with an average GDP per capita                              profi les between urban and rural areas,
below $10,000 (1990 international dollars),                         these disparities in GDP per capita will
GDP per capita in urban areas is two to                             overstate rural-urban differences in, say,

Figure 1.9        Rural-urban disparities in GDP per capita tend to be smaller in richer OECD countries

Ratio of urban to rural GDP per capita                               Ratio of urban to rural GDP per capita
3.50                                                                 3.50

3.00                                                                 3.00

2.50                                                                 2.50

2.00                                                                 2.00

1.50                                                                 1.50

1.00                                                                 1.00

0.50                                                                 0.50
0.00                                                                 0.00
       5           10        15        20        25        30            40      50        60      70      80         90       100
           GDP per capita (constant 1990 int’l $, thousands)                          Agglomeration Index, 2000 (%)
Source: WDR 2009 team, based on data from OECD (2007), pp. 1–256.
64                                               WO R L D D E V E LO P M E N T R E P O RT 2 0 0 9



Table 1.2    Rural-urban disparities in earnings, wealth, and consumption characterize development over the last two centuries
                               Rural-urban
 Country (year)                disparity (%)        Description and country sample
 Sweden (1805)                      221.0           Wealth per male adult in urban and rural areas.
 Finland (1805)                     146.0           Wealth per male adult in urban and rural areas.
 England (1830s)                      73.2          Urban wages are wages per laborer in the building trades, and rural wages are for agricultural laborers.
 France (1882)                        29.0          Urban wages are for unskilled wages in the regional capital city (department chef lieu), and rural wages are
 France (1911)                        51.0          based on average farm wages .
 United States (1925)                 28.0          Urban earnings are manufacturing earnings, and rural earnings are agricultural earnings.
 United States (1935)                 75.0
 Developing countries                 51.2          Urban wages are for unskilled general laborers, and rural wages are agricultural wages, including payments
 (nineteenth century)                               in kind. The countries included are Argentina 1872; Australia 1887; Denmark 1872; France 1892, 1801; Hungary
                                                    1865; Japan 1887; and the United States 1820–29, 1890.
 Developing countries                 41.4          Urban wages are based on wages for unskilled construction workers, and rural wages are agricultural cash
 (twentieth century)                                wages. There are 19 countries (1960–70) underlying this average: Argentina, Cameroon, Chile, Costa Rica,
                                                    Côte d’Ivoire, Guatemala, Kenya, Pakistan, Malawi, Malaysia, Mexico, Morocco, Panama, Sri Lanka, Tanzania,
                                                    Trinidad and Tobago, Tunisia, Uruguay, and R. B. de Venezuela.
 Developing countries                 42.0          Based on per capita household consumption, after controlling for household characteristics. There are 72
 (twenty-first century)                              countries (2000–05) underlying this average disparity: Armenia, Angola, Bangladesh, Belize, Benin, Bhutan,
                                                    Bolivia, Brazil, Burkina Faso, Burundi, Bulgaria, Cambodia, Cameroon, Chad, Chile, Colombia, Dem. Rep. of
                                                    Congo, Costa Rica, Côte d’Ivoire, Croatia, Djibouti, Ecuador, Arab Rep. of Egypt, El Salvador, Ethiopia, The
                                                    Gambia, Georgia, Ghana, Guatemala, Guinea, Guyana, Honduras, Hungary, India, Indonesia, Jamaica, Jordan,
                                                    Kyrgyz Republic, Madagascar, Malawi, Maldives, Mali, Mauritania, Mexico, Moldova, Mongolia, Morocco,
                                                    Mozambique, Nepal, Nicaragua, Nigeria, Pakistan, Panama, Paraguay, Peru, Philippines, Poland, Romania,
                                                    Russian Federation, Rwanda, Senegal, South Africa, Sri Lanka, Swaziland, Tajikistan, Tanzania, Thailand,
                                                    Timor-Leste, Uganda, Ukraine, Vietnam, and Zambia.
Sources: Sweden and Finland 1805: Soltow 1989, table 1, p. 48; England 1830s: Williamson 1987, table 3, p. 652; France 1882, 1911: Sicsic 1992, table 2, p. 685; United States 1925,
1935: Alston and Hatton 1991, table 3, p. 93; Developing countries (nineteenth century): Clark 1957, table II pp. 526–31;
Developing countries (twentieth century): Squire 1981, table 30, p. 102; Developing countries (twenty-first century): WDR 2009 team estimates based on individual country’s
household survey for 72 countries; the data set is described in detail in Montenegro and Hirn (2008).
Note: Rural-urban disparity (in nominal terms) is computed as the difference in wages, earnings, wealth, or consumption between urban and rural areas relative to the rural
averages.

                                                 average levels of personal disposable income                          disparities in productivity and income.
                                                 and consumption. The agglomeration index                              For a sample of developing countries in the
                                                 produces the same qualitative pattern.                                1960s—among them Malaysia, Mexico,
                                                     Rural-urban disparities in these countries                        and Trinidad and Tobago, which have since
                                                 were wide throughout the nineteenth and                               reached upper-middle-income or high-in-
                                                 early twentieth centuries. Wealth per male                            come status—urban wages exceeded rural
                                                 adult in nineteenth century Sweden was more                           wages by more than 40 percent. Similar gaps
                                                 than 200 percent higher in urban areas than                           can be observed in per capita consumption
                                                 in rural areas, and 150 percent higher in Fin-                        between urban and rural areas for a recent
                                                 land (see table 1.2). Meanwhile, for rapidly                          sample of 72 developing countries.
                                                 urbanizing England, urban wages were more                                 The rural-urban discrepancy between
                                                 than 70 percent higher than rural wages in                            economic mass and population distributions
                                                 the 1830s. France and the United States saw                           diminishes with urbanization. Another way
                                                 big increases in the urban wage premium                               to examine consumption disparities between
                                                 from 1882 to 1911 and from 1925 to 1935.                              urban and rural areas is to look at the popu-
                                                 Indeed, in the United States, the premium                             lation share of a country’s urban areas and
                                                 increased almost threefold in a decade.36 For                         compare it with the share of consumption
                                                 developing countries in the nineteenth cen-                           in these areas. If this ratio is greater than
                                                 tury, including Australia, Denmark, France,                           one, consumption per capita is, on average,
                                                 Japan, and the United States, urban nominal                           higher in urban areas than in rural areas,
                                                 wages were 50 percent higher.                                         while the converse is true if the ratio is less
                                                     Today’s developing countries are still                            than one.
                                                 in the first phase of urbanization and,                                   Rural-urban disparities in consump-
                                                 not surprisingly, have large rural-urban                              tion fall with density in today’s developing
                                                                                                                Density                                              65



Figure 1.10 Rural-urban gaps in per capita                  points. For countries where urbanization is
consumption become smaller with urbanization                advanced and the urban share is approach-
Ratio of urban consumption share                            ing its natural maximum, almost no differ-
to urban population share                                   ence exists between urban and rural areas
4.0                                                         in access to basic services. Equalization of
3.5                                                         access to basic services can be expected to
3.0                                                         promote a corresponding convergence in
2.5                                                         nonmaterial indicators of welfare and liv-
2.0                                                         ing standards (see table 1.3).
1.5                                                             Narrowing rural-urban disparities is
1.0
                                                            important, but the progress in absolute
0.5
                                                            measures of basic welfare in the rural areas
 0
      0      20         40         60        80       100
                                                            of the world’s poorest countries is even more
                  Urban population share (%)                important. Rising rural-urban disparities are
Source: WDR 2009 team estimates from more than 120 house-
                                                            consistent with an absolute improvement in
hold surveys for more than 75 countries.                    basic welfare in both rural and urban areas.
                                                            The overall evidence is encouraging. Over
countries (see figure 1.10).37 In Malawi and                 the past decade, most low- and middle-in-
Sri Lanka the ratio is around two: urban                    come countries have experienced absolute
areas account for about 10 percent of the                   improvements on a range of basic welfare
population but 20 percent of consumption.                   indicators, including infant and under-5
For countries with higher levels of urbaniza-               mortality rates, malnutrition, immuniza-
tion, the spatial distribution of population                tion, and school participation in rural and
more closely resembles that of production.                  urban areas. Of 32 low-income countries,
Madagascar and Tanzania have urban popu-                    three-quarters reduced infant and under-5
lation shares of around 20 to 25 percent and                mortality rates and the incidence of severe
urban consumption shares of about 30 to                     stunting and severe underweight, especially
35 percent. By the time a country enters an                 in rural areas.40 And since 1990, school
advanced stage of urbanization, population                  attendance rose in four-fifths of these
is more or less proportionately distributed                 countries, especially in rural areas.41 Both
with economic mass, so that the ratio is close
to one. In Chile 85 percent of the popula-                  Table 1.3   Rural-urban disparity in basic services narrows with development
tion reside in urban areas, and these urban                                             Disparity in          Disparity in access
residents account for 92 percent of national                 Urban population         access to clean            to sanitation
consumption. In Brazil 80 percent of people                  share (mean GDP         water (percentage           (percentage            Examples of countries in
                                                             per capita)                  points)                   points)             the sample
live in urban settlements, and these 80 per-
cent are responsible for 85 percent of con-                  75% or higher                      8                         8             United States, Norway,
                                                             (mean GDP per                                                              Switzerland, Spain,
sumption. As development progresses and                      capita: $21,602)                                                           Germany, Canada, Mexico,
the concentration of economic activity in                                                                                               Chile, Brazil, Argentina,
areas of high density increases, rural-urban                                                                                            Gabon, R. B. de Venezuela,
                                                                                                                                        Djibouti, Lebanon, Jordan,
disparities narrow. A downward sloping line                                                                                             United Kingdom
at all levels of urbanization is a good omen:
                                                             50%–70%                           15                       20              Estonia, Panama, Turkey,
most developing countries may have passed                    (mean GDP per                                                              Hungary, Ecuador,
the peak in their rural-urban disparities.38                 capita: $9,672)                                                            Colombia, Malaysia, Syria,
    What is true for private consumption is                                                                                             Azerbaijan, South Africa,
                                                                                                                                        Rep. of Congo, Algeria,
true for basic amenities. Among low-income                                                                                              Tunisia, Bolivia
countries with urban population shares of
                                                             25% or lower                      24                       26              India, Rep. of Yemen,
less than 25 percent, access to water and                    (mean GDP per                                                              Madagascar, Chad,
sanitation in towns and cities is around                     capita: $2,585)                                                            Tajikistan, Bangladesh,
25 percentage points higher than in rural                                                                                               Tanzania, Kenya, Nepal,
                                                                                                                                        Cambodia, Malawi, Uganda,
areas.39 But for more urbanized countries,                                                                                              Sri Lanka, Bhutan
such as Algeria, Colombia, and South Africa,                Source: World Bank 2007j.
the disparity in access is 15 to 20 percentage              Note: Disparity refers to the percentage point difference between urban and rural areas.
66                                               WO R L D D E V E LO P M E N T R E P O RT 2 0 0 9



Figure 1.11    Even at the subnational level, rural-urban disparities fall as density increases

                Philippines, 2000                                                 China, 1999 and 2006                                                 India, 1983 and 1994
Ratio of urban to                                            Ratio of urban disposable income                                      Disparity in life expectancy
rural incomes                                                to rural net income                                                   urban-rural ratio (by state)
4.0                                                          6                                                                     1.25
3.5                                                          5
3.0                                                                                                                                1.20
2.5                                                          4
                                                                                                                                   1.15
2.0                                                          3                                                                                                                1983
1.5                                                                                                 2006                           1.10
                                                             2
1.0                                                                                              1999                                                                                1994
                                                             1                                                                     1.05
0.5
0.0                                                          0                                                                       0
    0      20     40         60       80      100                0          20          40       60        80        100                  0      0.1         0.2        0.3        0.4      0.5
                 Urban share (%)                                               Urban population share (%)                                       State-specific urban share (%)
Sources: Balisacan, Hill, and Piza forthcoming; Yao forthcoming; Cali 2008.



                                                 urban and rural areas in these nations have                           and deprivation. Disparities within cities
                                                 achieved progress toward the Millennium                               can be large. In Nairobi poverty is high in
                                                 Development Goals.                                                    the inner city but much lower in the rest of
                                                    Rural-urban convergence takes place                                the city and the suburbs (see figure 1.12). In
                                                 sooner in more urbanized subnational                                  Mombasa, Kenya’s second-most-populous
                                                 areas. In both China and the Philippines,                             city, marked geographic divisions in the
                                                 urbanized provinces exhibit lower internal                            poverty rate are evident (see map 1.2). South
                                                 urban-rural disparities in incomes (see fig-                           African cities also show internal disparities
                                                 ure 1.11). In China the entire relationship                           in the poverty rate. Cape Town has a low
                                                 has shifted upward over the past decade                               poverty rate in the coastal areas, but a higher
                                                 so that, in general, rural-urban disparities                          poverty rate in the interior of the city. Simi-
                                                 have increased over time, consistent with                             larly, both Johannesburg-Pretoria-Tshwane
                                                 China’s early stage of development, which                             and Durban have visible divisions. But the
                                                 is marked by rapid urbanization. In India                             geography of poverty in Durban is different
                                                 rural-urban gaps in life expectancy were                              from that in Cape Town and Johannesburg:
                                                 smaller in the more urbanized states in                               the poverty rate is, in general, higher outside
                                                 both 1983 and 1994. But the entire relation-                          the city boundaries than inside.
                                                 ship has shifted downward over time.                                      The most obvious sign of divisions within
                                                                                                                       cities is slums. Slums have chronically over-
                                                 Slums—divergence and convergence                                      crowded dwellings of poor quality in under-
                                                 within cities                                                         served areas. The reason for the lack of basic
                                                 In poor countries, higher average living                              public services and infrastructure is the
                                                 standards in cities do not rule out poverty                           inability or unwillingness of many urban


                                                 Figure 1.12         Slums grow with the pace of urbanization, and fall with its level
                                                    Annual growth of slum population (%)                                   Percentage of slums
                                                    20                                                                     120

                                                    15                                                                     100

                                                    10                                                                      80

                                                     5                                                                      60

                                                     0                                                                      40

                                                    –5                                                                      20

                                                 –10                                                                         0
                                                    –2           0         2        4        6         8        10           –20      0         20      40         60         80     100     120
                                                             Annual growth of urban population (%)                                            Urban share of a country (%)
                                                 Source: Kilroy 2008.
                                                                                                                        Density                                                   67



Map 1.2 Local divisions—spatial disparities within urban settlements can be large
Poverty rates in African cities


                                                                                             KENYA


                                                                                         SOUTH
                                                                                         AFRICA




                                                            Nairobi




                                                                                                                                                   Johannesburg and Pretoria



                                                                                                                                Poverty rate:
   Poverty rate:                                            Mombasa                                                             Proportion of
   Proportion of                                                                                                                poor (%)
   poor (%)                                                                                                                        0–11
      17–42                                                                                                                        11–25
      42–48                                                                                                                        25–40
      48–58                                                                                                                        40–57
                                                                                                                                   57–92                              Durban
      58–63
      63–84                                                                                     Cape Town                          No data
      No data

Source: The Poverty Mapping Project, Columbia University, using data from Alderman and others (2002); Statistics South Africa; the Central Bureau of Statistics, Kenya; and the
Ministry of Planning and National Development, Kenya.




governments, utilities, and service provid-                            Slums are part of rapid urbanization, and
ers to operate in slums, generally because of                      it is not uncommon for a fifth to a third of
the informality and illegality of such settle-                     a city’s population in a contemporary devel-
ments.42 So living standards, especially                           oping country to reside in slums (see figure
health, security, and sanitation, are lower in                     1.12).44 Goiâna, the capital of the Brazilian
slums than in formal settlements close by.                         state of Goiàs, a medium-size city of 40,000
Mumbai’s Dharavi, believed to be Asia’s big-                       in 1950, is today a city of more than 1 mil-
gest slum, has “maybe a million residents . . .                    lion, with much of the population increase
crammed into a square mile of low rise wood,                       accommodated in slums.45 Since 1950,
concrete and rusted iron . . . a family of 12 liv-                 Delhi’s population has risen more than
ing in a 90-square-foot room.” In Shiva Shakti                     tenfold, from 1.4 million to 15.6 million,46
Nagar, again in Mumbai, each community                             accompanied by an increase in the number
tap is shared by roughly 100 people.43                             of slum clusters from 200 to 1,160.
   The growth of slums in major cities                                 “A dirtier or more wretched place he
is characteristic of rapid urbanization.                           had never seen. The street was narrow and
Because rapid population growth cannot                             muddy, and the air was impregnated with
be satisfactorily accommodated, slums and                          fi lthy odors. . . . Covered ways and yards,
shantytowns grow bigger and more visible.                          which here and there diverged from the
This contributes to wide and increasing                            main street, disclosed little knots of houses,
geographic divisions in well-being within                          where drunken men and women were posi-
urban areas. Development—both economic                             tively wallowing in filth.” A contemporary
and institutional—and better infrastruc-                           description of a developing country slum
ture, combined with focused interventions,                         such as Nairobi’s Kibera or Huruma, Abi-
eventually bring about a convergence in liv-                       djan’s Washington, Delhi’s Majboor Nagar or
ing standards in urban areas.                                      Kanchan Puri, Buenos Aires’s San Fernando,
68                                      WO R L D D E V E LO P M E N T R E P O RT 2 0 0 9



                                                                                           in multistory tenements arranged along nar-
     B OX 1.6    Slums, then and now                                                       row, unlit foot passages. This “housing was
                                                                                           hopelessly inadequate in all respects—in
     The term “slum,” probably originating     reached out beyond the working              quantity, in quality and environmental
     from an old English or German word        class, finally motivating strong politi-    amenities, if needs as basic as clean water
     meaning a poorly drained or muddy         cal action. But rather than attempting      and safe sewage disposal can be described as
     place, was applied to housing in the      to stop more workers from coming,
     early Industrial Revolution in the        or clearing out these areas of disease
                                                                                           amenities.”47 Apart from the obvious mis-
     United Kingdom before the railways        and poverty, the government in the          ery, slums were prone to deadly outbreaks of
     were in place, when canals trans-         1870s passed legislation for strict         measles and scarlet fever and high rates of
     ported heavy goods along the length       building regulations, prescribing the       mortality attributable to diarrheal diseases,
     and breadth of the country. During        dimensions of streets and houses, and       typhus, and respiratory diseases.48
     Britain’s rapid industrialization, most   making it mandatory that all dwell-             Yesterday’s slums are today’s world-class
     factories were built beside canals, the   ings be connected to newly built            cities. Britain is not the only industrial coun-
     main channel for transporting coal        sewerage systems. Major municipal
     for their steam engines and other         investments in water works, sewage
                                                                                           try to suffer from slums and wide intracity
     inputs of production.                     facilities, and public health dramati-      divisions in welfare during the earlier phases
        Poor workers, migrating to cities      cally reduced mortality in Britain’s        of development and rapid urbanization (see
     for factory jobs, could ill afford to     cities between 1874 and 1907.               box 1.7). The stylized pattern of divergence
     walk long distances to and from their        Despite atrocious and filthy con-        followed by convergence is a hallmark of
     places of work. Before electric trams,    ditions, millions of migrants keep          other modern-day developed countries as
     other forms of transport were expen-      leaving rural areas for the teeming
                                                                                           well. Slums for these cities are now much a
     sive. So workers settled close to fac-    economic opportunity offered in the
     tories. Cheap housing grew around         cities of poor and middle-income
                                                                                           thing of the past. Aided by improving land
     these factories in low-lying, poorly      countries. Even though health hazards       markets, investments in infrastructure, and
     drained areas. Housing was over-          and mortality rates are far worse in the    targeted incentives, within-city welfare dis-
     crowded. Sanitation was inadequate        shanties around many cities in Africa,      parities tend to narrow, but only in the more
     and in most cases nonexistent. And        people there are trading, working, and      advanced stages of urbanization. Indeed, for
     air quality was poor, with soot and       sending large sums of money home.           “world” cities such as London, New York,
     other pollutants. Sickness was com-       The challenge facing policy makers
                                                                                           Paris, Singapore, and Tokyo, slums can, with
     monplace. Diarrhea, typhus, respira-      today is similar to that faced by the
     tory diseases, measles, and scarlet       Victorians in London: how to nurture
                                                                                           the benefit of hindsight, be viewed as part of
     fever cut the life expectancy of those    these agglomerations with functional        their “growing pains.” Britain cleaned up its
     born in cities by 12 years compared       land markets, better transport, and         Dark Satanic Mills over a century, and if it
     with those born in rural areas.           public health infrastructure to capture     had started the cleanup sooner, the working
        The growing public health hazards      the benefits of economic growth.            class would have suffered from slower wage
     in Britain’s urban slums exacted a        Sources: Satterthwaite and others 2007;     growth and lower consumption.49
     terrible health toll that eventually      Crafts 2008; The Economist 2007a.
                                                                                               The emergence and growth of slums in
                                                                                           the early and intermediate stages of a coun-
                                                                                           try’s development can be explained by the
                                        or Rio de Janeiro’s Rocinha? No, this is an        interaction of functioning labor markets
                                        excerpt from Charles Dickens’s Parish Boy’s        with dysfunctional land markets. In the
                                        Progress, published in 1838, describing the        rapid phase of urbanization, the labor mar-
                                        rapidly expanding city of London in the            ket signals higher labor demand in urban
                                        nineteenth century (see box 1.6).                  areas, the higher demand that arises from
                                            London was by no means the only city           growth in industries and services. Labor
                                        or urban area in nineteenth century Britain        responds by moving to towns and cities.
                                        with large slum settlements. Chronically               As a reflection of this, slum dwellers in
                                        overcrowded and inadequately serviced              developing countries are often productively
                                        housing was a common feature of British            engaged, taking advantage of the economic
                                        cities and industrial towns of the time. In        opportunities the city offers. Mumbai’s
                                        Edinburgh rapid population growth and a            Dharavi has 15,000 “hutment” factories,
                                        first wave of suburbanization by the then-          and “the clothes, pots, toys and recycled
                                        rising middle classes meant that by the            materials its residents produce earn the fac-
                                        1860s, the core of the city had a large slum       tories millions of dollars a year.” Many slum
                                        area with population densities as high as 600      residents started businesses after the state
                                        persons per acre. Residents in this area lived     government provided them with limited
                                                                                                      Density                                          69




B OX 1.7     Many of today’s world-class cities were littered with slums
“In Antwerp and in most Belgian towns the           “Here the background embraces the                 tal soldier turned milkman, Vergniaud.
basic problem in matters of working class        pauper burial-ground, the station of the             There the Colonel lives in a single room
housing was . . . no individual sanitation or    Liverpool and Leeds railway, and, in the             with a dirt floor and a straw bed.”
individual water supply. . . . The three heavy   rear of this, the Workhouse, the “Poor-Law
                                                                                                         “Between 1815 and 1851 France’s popula-
cholera epidemics of the 19th century had        Bastille” of Manchester, which, . . . looks
                                                                                                      tion grew from 29 to 36 million . . . it was
terrific effects in these slums . . . “          threateningly upon the working-people’s
                                                                                                      the cities that absorbed the thousands of
                                                 quarter below. . . . Passing along a rough
   “The first encampments of Baltimore’s                                                              migrants unable to find work in the country-
                                                 bank, among stakes and washing-lines,
poor were at the water’s edge. Time and                                                               side. . . . But there were simply not enough
                                                 one penetrates into this chaos of small
again, outbreaks of yellow fever, malaria,                                                            jobs. Unemployment and overcrowding cre-
                                                 one-storied, one-roomed huts, in most of
cholera, typhoid fever swept the town.                                                                ated appalling living conditions. Only one in
                                                 which there is no artificial floor; kitchen,
These epidemics seemed peculiarly asso-                                                               five houses had running water. In 1832 chol-
                                                 living and sleeping-room all in one. In
ciated with the low-lying encampments                                                                 era wiped out some 20,000 Parisians.”
                                                 such a hole, scarcely five feet long by six
of the poor. The yellow fever epidemic of
                                                 broad, I found two beds—and such bed-                  “Like so many other European cities,
1797, for example, was said to have begun
                                                 steads and beds!—which, with a staircase             Paris suffered from chronic post-war
in the stagnant waters of the Fells Point
                                                 and chimney-place, exactly filled the                housing shortages. Of the 17 slum areas
cove and to have spread . . . to the huts
                                                 room.”                                               designed for clearance, most were still
and hovels on the banks of the Jones Falls
                                                                                                      intact in the 1950s.”
and thence on to the shacks and shanties             “Melbourne’s most infamous slum,
at the foot of Federal Hill.”                    Little Bourke Street, . . . by the 1880s . . . was       “One of the worst outrages of indus-
                                                 crowded, bustling and growing. . . . The             trialism in China against humanity is the
   “By the 1890s, Polish immigrants had
                                                 lane is completely filled up with all kinds of       herding of these workers in noisome
supplanted the Irish and Germans, creat-
                                                 filth comprising garbage tips, putrid liquid,        slums in the factory districts, . . . so foul
ing a ghetto of a new dimension. Single
                                                 straw rags, and other rubbish. A most dis-           and revolting . . . in Shanghai. . . . There
dwellings housed from six to eight families,
                                                 agreeable odor arose from this offensive             are no sanitary provisions of any kind, and
one [family] to a room. . . . Fells Point was
                                                 mass . . . the loathsome mass . . . exposed          the passages between the rows of houses
described by a health official as an Augean
                                                 and allowed to rot and spread its contami-           are practically open latrines. Overcrowd-
stable . . . a mass of nuisance . . . Open
                                                 nating influences.”                                  ing exists to a distressing extent. The
drains, great lots filled with high weeds,
                                                                                                      many children who are reared in these
ashes and garbage accumulated in the                “About 200 years ago, Lower Manhat-
                                                                                                      filthy quarters are covered with running
alleyways, cellars filled with black water,      tan was adorned by a pretty five-acre
                                                                                                      sores from dirt and bodily neglect.”
houses that are total strangers to the           lake known as the Collect. . . . By the mid-
touch of whitewash or scrubbing brush,           1700s, however, the Collect was already                 “In the 15 years between 1930 and the
human bodies that have been strangers            rimmed with slaughterhouses and tan-                 end of the war, the population of Singa-
for months to soap and water . . . that’s        neries. The effusions from these bloody              pore doubled to a million people. The
Pigtown.”                                        businesses were poured directly into the             population explosion had generated a
                                                 lake and more industries, more trash,                housing shortage of epidemic propor-
    “The slums of Dublin were among the
                                                 quickly followed. By 1800 the Collect was            tions. Small shophouses gave shelter to
worst in Europe, rivaled only by Glas-
                                                 a reeking cesspool. By 1813 it had been              as many as 100 people. The average living
gow. Tall town houses, originally built as
                                                 entirely filled in and by 1825 something             space was 9 feet by 9 feet, about the size
elegant homes for the rich in the eigh-
                                                 entirely new stood on the site—America’s             of a prison cell.”
teenth century, fell into the Tomae hands
                                                 first real slum, the Five Points.”
of avaricious and pitiless landlords who                                                                “All of the ghettos of the 1920s within
filled them to bursting point with the             “Although this is a hugely expensive               the city of Tokyo were products of Tokyo’s
desperate and impoverished urban poor.           area in Paris to live today, in Victor Hugo’s        urban development and Japan’s modern
Conditions were often unspeakably vile,          day it was a slum area, close to the Bastille        economic growth. . . . The sheer size of these
with massive over- crowding and utterly          Prison.”                                             ghettos was astonishing. . . . Poverty pockets
inadequate sanitation.”                                                                               re-emerged in all parts of the metropolis of
                                                    “[T]he lawyer Derville ventures into the
                                                                                                      Tokyo after the Second World War, even in
   “Katajanokka’s transformation in its          slums of Saint Marceau, the poorest sec-
                                                                                                      the midst of the old city of Tokyo.”
entirety from a low-income housing area          tion at the outskirts of Paris. Taking his
to an enclave for the city’s civil service       coach through the filthy rutted lanes, he            Sources: Belgium: Lis; Baltimore: Garrett
elite and bourgeoisie represented an             arrives at a broken-down building, made              2002; Dublin: Kearns 2006; Helsinki: Mäki-
                                                                                                      nen; Manchester: Engels 1987; Melbourne:
urban growth pattern that emerged for            entirely of second-hand materials and
                                                                                                      Mountford; Manhattan: Baker 2001; Paris:
the first time in the history of Helsinki.       poorly built, where Colonel Chabert is               Sanderson, Villon 2000, The Economist;
A former slum had become a prestigious           lodged with the cows, goats, rabbits and             Shanghai: Schwenning 1927; Singapore:
residential area for the privileged classes.”    impoverished family of a former regimen-             Baker 1999; Tokyo: Koji 1969.
70   WO R L D D E V E LO P M E N T R E P O RT 2 0 0 9



     rights over their dwellings in 1976 and began                     European countries lived in urban settle-
     to supply water and power to parts of the                         ments of 5,000 inhabitants or more.52 In this
     settlement. Because Dharavi is sandwiched                         respect, at least, little had changed from the
     between the city’s two main railway lines                         previous five centuries. So the takeoff into
     and is surrounded by six stations, it also acts                   urbanization over the next century broke
     as Mumbai’s transportation hub.50 In short,                       dramatically from the past.
     slums arise in many developing countries as
     low-income households take advantage of                           The pace and pattern of urbanization
     spatially concentrated employment oppor-                          is similar
     tunities and as businesses take advantage of                      It started in Great Britain. In 1800 Britain’s
     their location in a land-constrained envi-                        urban share stood at 19.2 percent, about
     ronment. Consistent with today’s industrial                       twice the European average. But in the first
     countries, the correct response is not to                         two decades of the century, the number of
     slow, stop, or reverse urbanization. It is to                     people living in urban areas doubled. By
     tackle dysfunctional land markets.                                1820 the urban share was 40 percent. By the
        The interplay of such market forces and                        close of the century, seven of every 10 Brit-
     responses from rational market actors can                         ons were living in urban settlements. Britain
     also be seen in many Sub-Saharan African                          was joined in its headlong rush into urban-
     countries. But inefficient land markets,                          ization by other early European industrial-
     often thanks to misguided urban plan-                             izers. By the second half of the nineteenth
     ning and zoning, produce only a limited                           century, urbanization spread beyond the
     and unresponsive supply of affordable, legal                      Old World to the United States and Canada.
     land sites for building housing to keep pace                      By World War I, four of every 10 Americans
     with the demand.51                                                were living in urban settlements with popu-
                                                                       lations of 5,000 or greater; just 60 years ear-
     What’s different for today’s                                      lier, the ratio was one in 20.
     developers?                                                           So if anything is different for today’s
     At the beginning of the nineteenth century,                       developers, it is certainly not the pace of
     one person in every 10 in today’s developed                       urbanization. Indeed, the average pace of


     Figure 1.13    Urbanization’s speed has precedents

     Percentage point difference in urban population, 1985–2005 (except where specified)
     35


     30

                Canada,
                1880–1900
     25

                                            United Kingdom,
                            Germany,        1830–1850
     20
                            1880–1900             Denmark and United States,
                                                  respectively, 1880–1900
     15                                                Switzerland,        Mean of high-income
                                                       1880–1900           countries, 1880–1900
                                                                            Mean of developing
     10                                                                     countries, 1985–2005
                                                                              Median of developing
                                                                              countries,1985–2005
      5


      0
                                                               All countries
     Sources: WDR 2009 team calculations based on data from the United Nations (2006c); historical data for Canada, the United King-
     dom, and industrial countries’ averages are from Bairoch and Goertz (1986) and Dumke (1994).
                                                                                                                      Density                                             71



urbanization for developing countries over                           Between 1985 and 2005, China added
1985–2005 is remarkably similar to the                            225 million people to its towns and cities,
average for European and North Ameri-                             almost the entire population of the United
can countries53 between 1880 and 1900 (see                        States. Yet China for the same time period,
figure 1.13).54 For the early developers the                       ranked only fifteenth in its absolute increase
average absolute increase in the urban share                      in urban share. In India the number of peo-
over the 20 years was 7.7 percentage points,                      ple in towns and cities rose by 137.8 million,
and for current developers the respec-                            adding a Germany and an Italy to its urban
tive median and mean absolute increases                           areas in just two decades.
were 7.1 and 8.0 percentage points. The                              Today’s developing countries had an
pace of urbanization among most of the                            average increase in their urban popula-
early developers in the last two decades of                       tion of 8.3 million over 1985–2005, almost
the nineteenth century ranked in the top                          three times the increase for many of today’s
quartile of the contemporary distribution                         high-income European and North Ameri-
of urbanization speeds.                                           can countries between 1880 and 1900. But
                                                                  when China and India are excluded from
The volume of urbanization is greater                             the group, the average urban population
for today’s developers                                            increase in recent decades has only been
What then is different? One difference is the                     4.4 million, about 50 percent more than
unprecedented absolute increases in urban                         the average for the early developers during
populations in many developing countries                          1880–1900 (see figure 1.14).57
in recent decades. Today’s developing coun-                          Correspondingly, megacities in devel-
tries simply have larger populations than                         oping countries are unprecedented in their
the industrializing countries of the nine-                        size. Through the nineteenth century the
teenth and early twentieth centuries. The                         world’s largest city was London. But its
urban population today, estimated at 3.3                          1900 population of 6.6 million was only
billion, is far greater than the world’s total                    a third that of modern-day Mumbai or
population as recently as 1960. It took more                      New Delhi, the largest cities in low-income
than 10,000 years for the urban population                        countries. The London of 1900 and, indeed,
to reach 1 billion in 1960, 25 years to add                       even the London of today are also smaller
the second billion, and only 18 to add the                        than modern-day Shanghai (10 million),
third.55 According to the UN projection, it                       the largest city in lower-middle-income
will take just 15 years to add the fourth.56 In                   countries, and several others (Cairo,
East Asia alone, 500 million people will join                     Jakarta, and Manila) among the more suc-
today’s 750 million urbanites over the next                       cessful developers. With more than 22 mil-
25 years, essentially adding another Paris or                     lion people, Mexico City, the largest city in
Kuala Lumpur every month.                                         upper-middle-income countries, is three


Figure 1.14    The population increment in urban areas of today’s developing countries is much larger

Change in urban population (thousands), 1985–2005                                          Change in urban population (thousands), 1985–2005
                                                                                           30,000
140,000                                                                                                     United States, 1880–1900
                                                                                           25,000
120,000

100,000                                                                                    20,000

 80,000
                     United States, 1880–1900                                              15,000
 60,000                                                                                                                Average for developing countries (excluding
                       Average for developing countries, 1985–2005                         10,000                      China and India), 1985–2005
 40,000                          Average for developed countries, 1880–1900                                                 Average for developed countries, 1880 –1900
                                                                                            5,000
 20,000

      0                                                                                          0
                            All countries including China and India                                                   All countries excluding China and India
Sources: WDR team calculations based on data for 1985–2005 from the United Nations (2006c) plus historical data from Bairoch and Goertz (1986).
72   WO R L D D E V E LO P M E N T R E P O RT 2 0 0 9



     times the size of London at the start of the       workers in London earned an urban real
     twentieth century.                                 wage premium of 67 percent, a large part of
                                                        this premium was compensation for the evi-
     Urbanites today enjoy both higher                  dent health hazards of city living.64
     private earnings and better public                     In Germany during the second half of the
     services                                           nineteenth century, infant mortality rates
     Cities now do better than rural areas in both      in rural areas were about 150 per 1,000 live
     income and nonincome indicators of well-           births. But expanding Berlin had the high-
     being. In 2000 the infant mortality rate in        est infant mortality in the Kaiserreich era,
     rural Malawi was 117 per 1,000 live births,        hovering around 300 per 1,000 live births in
     in urban Malawi it was 83. Urban Benin             the 1860s, and peaking at 410 per 1,000 live
     did much better than rural Benin in low-           births in the 1870s. The rural-urban gap in
     ering under-5 mortality rates and reducing         physical well-being remained for decades
     diarrhea and acute respiratory infections.58       during the nineteenth century.65
     Urban Ugandan women were less likely to                As the U.S. economy industrialized and
     suffer from anemia or malnutrition. Supe-          urbanized, people living in high-density
     rior health indicators are repeated in urban       areas at the turn of the twentieth century
     areas throughout the developing world—             were exposed to infectious and parasitic
     from Chad and Cameroon in Sub-Saharan              diseases. In 1880 urban mortality for adults
     Africa, to Nepal in South Asia, Kazakhstan         was 50 percent higher than rural mortality,
     in Central Asia, and Nicaragua in Latin            and two decades later, the urban mortality
     America, and to Morocco and Egypt in               rate was still 18 percent higher. The rural-
     North Africa and Middle East.59                    urban mortality difference was even greater
         But the opposite was true for the devel-       for infants and young children. For infants,
     opers of the nineteenth and early twentieth        excess urban mortality was 63 percent in
     centuries. Migrants to cities could expect         1890 and 49 percent in 1900, and for young
     better material standards of living, offset by     children ages one to four, the respective fig-
     poorer health and shorter lives for them and       ures were 107 percent and 97 percent. In 1900
     their children. In 1881–91 life expectancy at      male life expectancy was 10 years shorter in
     birth was 51 years in English and Welsh vil-       urban areas than in rural areas.66
     lages, but only 44 years in London and 39              That the cities and towns of modern-
     years in large towns.60 In 1850s Britain the       day developing countries do better than
     infant mortality rate in cities with popula-       villages on indicators of health, while the
     tions greater than 100,000 was, at 196 per         opposite was true for the developed coun-
     1,000 live births, far higher than the 138 per     tries at similar incomes in the nineteenth
     1,000 live births in rural communities.61          century, reflects advances in public health
         Even as late as 1937, George Orwell saw it     and medicine, and improvements in sewers
     fit to characterize industrial towns and cit-       and water systems. It also reflects the pub-
     ies as places where “one always feels that the     lic benefits that today’s cities in developing
     smoke and filth must go on for ever and that        countries confer. So the advantages of high
     no part of the earth’s surface can ever escape     density are not limited to income genera-
     them.”62 It is perhaps no surprise, then, that     tion and wealth creation—they also include
     the absence of respiratory diseases attribut-      social services.
     able to poor air quality in the cities would           With these differences in private and
     have resulted in life expectancies 4.7 years       public sources of well-being, it should
     longer in the England and Wales of 1861–70.        hardly be a surprise that cities and towns in
     In the absence of cholera, diarrhea, dysen-        the developing world are growing rapidly.
     tery, and typhus, life expectancy might            The surprise is that this move to density is
     have been 1.7 years longer, and the absence        not faster. And the policy implication? Any
     of measles and scarlet fever, common in            strategy for a less desperate and more delib-
     the cities, would have added 2.3 years to          erate urbanization must include efforts to
     life expectancy.63 Thus in the 1830s, while        improve public services in rural areas.

								
To top