Global Urbanization Today: Low- and Middle-Income Countries · Urbanization in Poor CountriesLarge increases in world population lie ahead; almost all growth to occur in thecities

Global Urbanization Today:Low- and Middle-Income Countries

Mark R. Montgomery

Stony Brook Universityand

Population Council, NY

May 25, 2015

Montgomery (Stony Brook UniversityandPopulation Council, NY)Urban May 25, 2015 1 / 36

Urban Challenges in the 21st Century:

Urbanization in Poor Countries Large increases in world population lie ahead; almost allgrowth to occur in the cities and towns of poor countries. Today’surban dwellers, and those of the future, will live mainly in small andmedium-sized cities.

Decentralization National governments transferring responsibilities into hands of stateand municipal governments, which typically lack resources and allmanner of bureaucratic expertise.

Governance and the Urban Poor As urbanization proceeds, the poor will increasingly befound in developing-country cities and towns; success in povertyalleviation will become a test of urban governance.

Extreme-event risks As global warming takes hold, the consequences (floods, droughts)will be borne by city and town dwellers of poor countries, as well asrural dwellers. But national adaptation plans typically ignore urbanareas and the urban poor.

Post–2015 Sustainable Development Goals How will such urban concerns berecognized, measured, and monitored?


Emergence of New National Governmental Systems:Decentralization

National governments are passing to lower-level tiers of governmentimportant functions, responsibilities, revenue-raising authority, and(sometimes) revenues.

Powerful notion of moving government “closer to the people,” intheory improving responsiveness

Highly variable implementation

Municipal and “state” governments increasingly important in settingpolicies and programs — but often poorly resourced and under-staffed


Data and Governance in Poor Countries

Almost all poor countries collect spatially-specific population andsocioeconomic data via population censuses

But unlike the high-income countries of Europe and North America,poor countries do not systematically analyze and distributespatially-specific data to their local governments and civil society

Latin American countries the prominent exception: Detailed, easilyaccessible data seen as essential to good governance. Mexico, Brazil,Uruguay and other examples. This view gaining adherents (e.g., Indiafor its 2011 census)

But elsewhere, national statistical offices do not view data analysisand distribution as part of their work

Smaller-city governments especially disadvantaged.


How Can International Research Help?

Development assistance for analysis as well as collection of populationcensus data

Publication of jurisdictional boundary shapefiles, and summaries ofsocioeconomic data by jurisdiction (as in Latin America)

Move the development and scientific conversation away fromrural–urban to a focus on places: specific cities and towns

Emphasize the need to link population data to spatial data andpopulation researchers to remote-sensing and other spatially-informedresearchers.


Time to Jettison “Urban Sector” and “Rural Sector”?

We think that the term “urban” aids understanding of development,and helps to guide programs and development policies in aconstructive way.

But does it? Can we not substitute “rural–urban continuum” or“connectedness” for the increasingly stale conception of an“urban–rural dichotomy”?

Rural villagers participate (directly or indirectly) in urban economic,social, governance and health systems. Two-way connections viamigration, remittances, and multiple input and output markets.

Spatial data illuminate these connections


For Climate Adaptation, Focus on Places: Cities andTowns

“Urban” is an abstraction, not really an actionable concept. But cities andtowns are specific places, overseen by municipal and other governments.

This shift of perspective brings gaps and deficiencies of national statisticalsystems into plain view. Most countries collect very little reliabledemographic, economic, health, and social data at the level of individualcities and towns. How will the post-2015 statistical systems monitor citiesand towns?

Population censuses are the major exception. They must—in my view—beassigned a major role in the post-2015 measurement and monitoringagenda.


The Demography

The Urban Demography of Poor Countries:Who lives where today?

Forecasts of size and location of population growth


The Demography

UN Projects Enormous Growth in Urban Populations

-100

0

100

200

300

1950 1975 2000 2025 2050Year

Tot

al P

opul

atio

n G

row

th in

Pre

viou

s 5

Yea

rs (

mill

ions

)

Development Level (Dashed Line shows Rural Growth): Developed Developing


The Demography

Urban Population Size by Region, Low- and Middle-IncomeCountries

01

23

4U

rban

Pop

ulat

ion,

in B

illio

ns

1950 1975 2000 2025 2050Year

AfricaLatin America

AsiaOceania


The Demography

What’s Distinctive About Today’s Urban Transition?

Let’s correct some common misunderstandings:

Urban population growth rates are higher than in comparable periodsin the West—lower urban mortality, stubbornly high urban fertility insome cases, urban population momentum from past growth

But the pace of urban growth—that is, the change in percentageurban over time—is not historically unusual. It is not correct to saythat “urbanization” is proceeding faster than in the history of theWest.

Although information is limited, it is not obvious that rural-to-urbanmigration rates for today’s poor countries are distinctively higher orlower than in the past.


The Demography

Clearly Unprecedented: Number of Cities of 1 Million+

36

3 2

97

6 5

102

12 5

33

2 0

238

2212

431

3821

010

020

030

040

0T

otal

Num

ber

of C

ities

More developed regions Less developed regions

1950 2000 2025 1950 2000 2025

Cities 1−5 Million


Cities 10 Million+

Preston(1980) termedthis “urbangigantism”.


The Demography

Cities of 1 Million+ in 2000, by Region

34

2 1

44

3 4

160

17

7

050

100

150

Tot

al N

umbe

r of

Citi

es

Africa LAC Asia



Cities 10 Million+


The Demography

Urban Population by City Size: Smaller Places Matter!

50.2%

10.4%

21.2%

8.6%

9.6%

25

50

75

0/100

CitySize

Fewer than 500,000

500,000 to 1 million

1 million to 5 million

5 million to 10 million

10 million or more

However, amongall urban dwellers,fewer than 1 in 10live in so-called“mega-cities”. Farlarger percentagesin smaller citiesand towns.


The Demography

Sources of City and Urban Population Growth

Natural increase Multiple credible estimates indicate that the share ofurban growth due to natural increase is about 60 percent.The exact figure varies across countries (depending onfertility levels) and over time for given cities—but 60 percentis a passable rule of thumb. City growth is only partly due tomigration.

Origin of urban in-migrants Today, even in relatively low-urbanizedcountries, more urban in-migrants come from other urbanareas than from rural villages!

Are in-migrants disadvantaged? Do they live in slums? Some aredisadvantaged and live in slums; others aren’t and don’t.The common perception needs critical examination.


The Demography

The 60/40 Rule

The United Nations rule-of-thumb: In developing countries, about 60percent of the urban growth rate is due to natural growth, thedifference between urban fertility and mortality rates. The remaining40 percent is due to migration and spatial expansion.

Similar rule applies to countries followed over time: 60 percent figurecited in Handbook of Indian Urbanization (2005:32) for four decadesfrom 1961 to 2001.

China the great exception to the 60/40 rule—low fertility, migrationtightly controlled, then unleashed. Population of “floating migrants”estimated at 150 million.


The Demography

Keep Urban Natural Increase in Mind!

Many countries are nervous about city growth and have forcefullydiscouraged migration—often via punitive, brutal, welfare-reducingand (ultimately) ineffective actions such as forced evictions.

More enlightened regional development policies are promising, butseldom produce rapid change in pace and spatial distribution ofpopulation growth.

But voluntary urban family planning programs have proveneffectiveness against a major component of city growth. This simplepoint is almost universally ignored.


The Demography

Cities and Towns:Linking Population to Spatial Extents


Change Over Time: The UN Cities Database

UN Population Division: City Population Time-Series

Longitudinal data: Population counts for individual cities over time,reported at irregular intervals. Long time-series, in many cases (5+observations)

The UN monitors all cities of 100,000 population and above; when agiven city crosses this threshold, the UN attempts to reconstruct itshistory.

Since 1970s, UN asks for data on urban agglomerations—butcountries often supply counts for the city proper, the metropolitanarea, or in unspecified types.

No boundaries as such—only the type of boundary.



Improving the UN Cities Database

1 Bring spatial content to a currently aspatial database.

Short-term goal: Vigorously clean and update the city data, includebasic spatial identifiers, and explore spatial factors influencing growth.Longer-term goal: Need to assemble a geographic database using thesmallest available geographic “building blocks” to which populationdata can be attached.Need to add meaningful jurisdictional boundaries

2 Need better coverage of smaller cities and towns, where a largepercentage of urban residents live, and to areas on the peripheries oflarge cities or lying between cities that are likely to fuse with theseurban populations.

3 Need to incorporate land cover rasters



Heterogeneous City Histories: Luanda, Angola

● ●●

●

●●

●

●

0

1000

2000

3000

4000

1950 1960 1970 1980 1990 2000 2010Year

Pop

ulat

ion

(000

s) BoundaryType

●●●

City Proper

Unknown

Urban Agglomeration



Spatial Expressions of Population Growth: GHSL

Percent_1975 Percent_1990

Percent_2000 Percent_2014

0

20

40

60

80

100



Heterogeneity across Cities and Time

● ● ●

●● ● ●

● ●● ● ●●

●

●

●

●●

●

●

●

●

●

●

● ●●

0

1000

2000

3000

4000

1940 1960 1980 2000Year

Pop

ulat

ion

(000

s)

BoundaryType

●

●

●

City Proper

Unknown

Urban Agglomeration



Forecasting City Size and Growth

The Science Behind the UN’s City and Urban Forecasts

Since the early 1980s, forecasts based on a simple extrapolation methodthat produces substantial, well-documented forecast errors. A high-priorityarea for research—new work would be welcomed by the UN.



Urban Population Forecast Errors for the Year 2000Source: Cities Transformed, 2003

Mean Percentage Forecast Errors1980–2000 1990–2000 1995–2000

RegionEast Asia and Pacific 3.9 26.7 -2.8EAP excluding China 18.4 9.8 -0.4Latin America and Carib. 19.8 5.4 -0.9Middle East and North Africa 13.3 6.8 8.5South Asia 27.2 19.7 2.7Sub-Saharan Africa 21.8 23.4 5.5

Level of DevelopmentLow 23.1 18.3 3.2Lower Middle 6.9 26.1 -1.3Lower Middle excluding China 25.6 9.9 3.7Upper Middle 12.8 8.9 0.8

Figures show population-weighted averages of(U f

t,2000/U2000 − 1)· 100 with U f

t,2000 the

UN forecast made in year t = 1980, 1990, 1995 of a country’s total urban population in

the year 2000, and U2000 the actual population in 2000.


Urban Poverty and Governance




Urban Poverty in the Millennium Development Goals

Target 11

The United Nations Millennium Declaration specified a target of achievingby 2020 “significant improvement in the lives of at least 100 million slumdwellers.”

UN-HABITAT estimates that there are already nearly 1 billion slumdwellers in developing countries.



Recent Estimates of “Slum Dwellers”

050

100

150

200

250

Slu

m P

opul

atio

n in

Mill

ions

Southern AsiaEastern Asia

Sub−Saharan AfricaLatin America

South−Eastern AsiaWestern Asia

North AfricaOceania



Poverty Rates Higher in Secondary Cities: Cote d’Ivoire

1984 1987 1990 1993 1996 1999

Year

0

20

40

60

Pro

port

ion

Und

er U

S$2

Per

Day

AbidjanOther Cities



Is Urban Poverty Spatially Concentrated?Major unanswered questions:

What exactly are “slums”? And where are they? (Officiallyrecognized?)

What proportion of the urban poor live in slums?

What proportion of slum residents are poor?

How long do slum residents stay in those neighborhoods?



Important Limitations

UN-Habitat estimates the number of urban dwellers with inadequatehousing, access to drinking water, and sanitation, from a careful analysis ofover one hundred DHS and MICS surveys.

Despite “slum-dwellers” label, no consideration of spatial concentration

Spatial concentration difficult to measure from survey data alone

Should be interpreted as estimates of the number of urban poor, notslum-dwellers as such.



Are Most of the Urban Poor Slum-Dwellers?

We don’t know. For India, however, Chandrasekhar and Montgomery(2009) found:

UnlistedSlums

ListedSlums Non-Slums Total

Poor UrbanHouseholds

6.7 11.4 81.2 ≈ 100.0%

Non-Poor UrbanHouseholds

2.2 5.0 92.4 ≈ 100.0%

Siddharth Agarwal (Urban Health Resource Centre, New Delhi) shows thatofficial records of unlisted slums are woefully incomplete—in several cities,number understated by half. Even with this adjustment, slum-dwellerswould not represent the majority of the urban poor.


The Post-2015 Sustainable Development Goals

What’s Being Considered for the Post–2015 SDGs?

A Sustainable Urbanization Goal Appears all but certain to be adopted inSeptember 2015: “Goal 11. Make cities and human settlementsinclusive, safe, resilient and sustainable”

Targets and Indicators The upcoming tasks:

Frame goals in terms of sub-national, disaggregated targetsand indicatorsWhere possible, localize these to city–regions, with roles forlocal urban governments?Must overlay statistical monitoring systems and governmentalsystems to know what is feasible in the near term.



Proposed Targets for the Urbanization Goal Include:

Adequate, safe and affordable housing and basic services and upgrade slums

Significantly reduce the number of deaths and the number of people affected anddecrease by [x] per cent the economic losses . . . disasters, including water-relateddisasters, with a focus on protecting the poor and people in vulnerable situations

Reduce per capita environmental impact of cities (air quality and municipal andother waste management)

Integrated policies for mitigation and adaptation to climate change, resilience todisasters, in line with the forthcoming Hyogo Framework

Participatory, integrated and sustainable human settlement planning

Support positive economic, social and environmental links between urban,peri-urban and rural areas by strengthening national and regional developmentplanning



Other SDG Goals Relevant to Cities and Towns Include:

Reduce extreme poverty and poverty as nationally defined

Reduce hunger and malnutrition (recognizes internal markets)

Reduce maternal, child mortality and mortality from both communicable andnon-communicable disease. (Road traffic accidents specifically mentioned.)Universal access to family planning. Reductions in air, water, and soilpollution

Adequate access to drinking water, sanitation; reductions of water pollution

Effective climate change-related planning and management in leastdeveloped countries, focusing on women, youth and local and marginalizedcommunities



New Space is Opening for International Research

The proposed SDGs recognize (implicitly) the value of spatially-refinedestimates and forecasts, at the subnational and even city levels

Much work remains to construct national monitoring systemsadequate to the task ahead

Population censuses are still under-exploited!

Vast array of spatially-explicit remotely-sensed and other data anenormously valuable, unprecedented set of assets for the SDGs.Countries and international agencies will need assistance to make themost of these new assets.


Global Urbanization Today: Low- and Middle-Income Countries · Urbanization in Poor CountriesLarge increases in world population lie ahead; almost all growth to occur in thecities

Documents