The Urban Research Agenda: In Zambia and Elsewhere

Report
Cities in the Developing World
Ed Glaeser
Harvard University
“I regard the growth of cities as an evil thing,
unfortunate for mankind and the world. ”
1
0
.2
.4
.6
.8
Hong Kong SAR,
Singapore
China
Kuwait Qatar Belgium
Venezuela, Argentina
RBIsrael
Uruguay
Japan
Australia
Chile
Bahrain
Denmark
New
Zealand
Gabon
France
Sweden
Brazil Finland
United Arab Emirates
Korea,
Rep.
Jordan Netherlands
United
States
Saudi
Arabia
Canada
United Kingdom
Norway
Libya
Spain
PeruMexico
Cuba
Colombia
Panama
Germany
Russian
Federation
Switzerland
Czech
Republic
Bulgaria
Malaysia
Algeria
Turkey
Cyprus
Estonia
Dominican
Republic
Iran,
Islamic
Rep. Latvia
Hungary
Ukraine
Italy Austria
Mongolia
Lithuania
Ecuador
Iraq
Bolivia
Tunisia
ElRep.
Salvador Armenia
Costa
Rica
Congo,
Ireland
South
Africa
Paraguay
Botswana
Poland
PortugalGreece
Croatia Nicaragua
Gambia, The Morocco
Syrian
Arab Republic
Slovak
Republic
Romania
Albania
Haiti
Jamaica Kazakhstan
Honduras
Cameroon
Ghana
Cote
d'Ivoire
Slovenia
Indonesia
Guatemala
China
Philippines
Liberia
Moldova
Benin
Egypt, Arab Rep.
Senegal Mauritius
Mauritania
Sierra
Central
Leone
African Republic
Zambia
Zimbabwe
Namibia
Togo
Pakistan
Kyrgyz
Mali
Thailand
Congo, Dem.
Rep.Republic
Lao
Sudan
PDR
Myanmar
Yemen,
Rep.
Mozambique
India
Vietnam
Bangladesh
Lesotho
Tajikistan
Tanzania
Kenya
Afghanistan
Swaziland
Cambodia
Rwanda
Niger
Nepal
Malawi
Uganda
Sri
Lanka and Tobago
Trinidad
Papua New Guinea
Burundi
0
.2
.4
.6
Urbanization in 1960
.8
1
0
.2
.4
.6
.8
1
Share of Countries over 1/3 Urbanized, by GDP per Capita (2012 $)
1960 and 2010
$0-1000
$1000-2000
$2000-3000
1960
Source: World Bank
$3000-4000
2010
$4000-5000
Poor (under $1200 p.c.), Populous
Countries that are one-third urban,
Country
Largest City
(Population)
Congo
(Dem. Rep)
Kinshasa
(9 Million)
Zimbabwe
Harare
(1.6-2.8 Million)
Mali
Bamako
(1.8 million)
Haiti
Port-au-Prince
(1 – 2.4 Million)
Pakistan
Karachi
(23.5 Million)
Senegal
Dakar
(1-2.5 Million)
Percent
Urbanized
.34
.38
.34
.52
.36
.42
Percent in
Million+
Agglomeration>
GDP P.C. 2010$
(PPP Adjusted)
.17
210
(330)
.12
625
(missing)
.13
650
(1100)
.21
700
(1000)
.19
1100
(2400)
.24
1100
(1700)
1
Belgium
.8
Australia
United Kingdom
Argentina
Sweden
Germany
United States
Chile
Austria
Japan
France
Venezuela, RB
Netherlands
Italy
Cuba
Spain
Hungary
Switzerland
Mexico
Poland
Peru
South
Africa
Brazil
Colombia
Iraq
Greece
Egypt, Arab Rep.
Bulgaria
Portugal
Romania
Iran, Islamic Rep.
Turkey
Algeria
Philippines
Morocco
Korea, Rep.
Malaysia
Ghana
Congo,
Dem. Rep.
Pakistan
Thailand
India China Myanmar
SriVietnam
Lanka
Indonesia
Cameroon
Mali Sudan
Cambodia
Yemen,Afghanistan
Rep.
Kenya
Tanzania
Bangladesh
Uganda
Mozambique
Nepal
0
.2
.4
.6
Canada
0
1000
2000
3000
Cereal Yield per Hectare 1961
4000
Table 3:
Urbanization and Agricultural Productivity, 1961 and 2010
Year:
Log of Agricultural Productivity
Log of Agricultural Productivity * Demeaned Log of Population
Log of Population
Observations
R-squared
(1)
1961
(2)
1961
(3)
2010
(4)
2010
0.095***
(0.024)
0.038***
(0.012)
-0.205***
(0.070)
119
0.189
0.051**
(0.021)
0.025**
(0.010)
-0.134**
(0.056)
119
0.531
0.054***
(0.018)
0.025***
(0.009)
-0.157***
(0.056)
139
0.085
0.00
(0.019)
0.021**
(0.008)
-0.122**
(0.051)
139
0.304
Notes:
Standard errors in parentheses (*** p<0.01, ** p<0.05, * p<0.1)
Agricultural productivity is defined as cereal yield in kilograms per hectare times hectares per capita. The interaction between
agricultural and population has demeaned the population in the given year so that the raw coefficient on agricultural productivity can be
interpreted as the impact of agricultural productivity at the mean level of population. Data comes from the World Bank. Standard errors
are in parentheses. A constant is included in the regression but not reported.
The Strength of Urban Poverty?
.2
India
Mali
Ghana
.1
South Africa
Moldova
Guatemala
0
Rwanda
Mexico
Poland
Sweden
Malaysia
France
Finland
Indonesia Romania
Russian Federation
United Kingdom
Egypt, Arab Rep.Brazil Slovenia
Japan
Morocco Bulgaria
Australia
Norway
United States
Canada
Germany
Netherlands
Ukraine
Cyprus
Spain
Argentina
Italy
Uruguay
New Zealand
-.1
Iraq
Thailand
4
6
8
Log of Per Capita GDP
10
12
Happiness across the United States
The Urban Triad
The Economic Magic of Human Interaction by ‫חדוה שנדרוביץ‬
Government battling the Demons of Density
The Physical City by rulto
11
10
9
8
7
6
0
.2
.4
.6
% Urbanization, 2010
Log of P.C. GDP 2010 PPP
.8
Fitted values
1
Index of Earnings for Urban Areas
1.4
1.35
1.3
Earnings Index
1.25
1.2
1.15
1.1
1.05
1
US
India
China
0.95
0.9
1
2
3
4
Quintile of Population Density
5
USA data is from the 2005 ACS. China data is from the 2005 Census. India data is from the 2005 IHDS (India Human Development
Survey)
Do Cities Increase Productivity?
• Random Shocks to Location for Individuals
– Roots in the individual fixed effect literature
– Similarities with Randomizing Across Peers
– Doesn’t deal with omitted place level factors
• Random Shocks to the Density Level
– If long term, these must be orthogonal to current
productivity (Combes et al. 2010)
– Or taking advantage of high frequency variation
(Greenstone, Hornbeck, Moretti)
• An alternative is to estimate the flow of ideas within a
network (Duflo and Saez, 2003)
• Real estate prices as a sign of WTP for proximity
Combes, Duranton, Gobillon and Roux
Human Capital and Urban Success
0
.05
.1
.15
Average Population Growth by Share with BA in 2000
(Quintiles)
1
2
3
4
5
Per Capita GDP 2010
.
100000
SAN JOSE
80000
WASHINGT
SAN FRAN
o
o
o
BOSTON
o
NEW YORK
o
oo
o
o
o
o
DALLAS
o
oo o
LOS ANGE
CHICAGO
o o oo
oo
o
ATLANTA
o
o
o
o
o
ooo
o
oooooo ooo
o
o o
LAS VEGA o oo o oDETROIT
o o o o oo
oo o o
oo o o
o
o o oo o
o
o
o o o o o oo o ooooo
ooo
o
o
o
o
o
o
o oo o o o o o ooo o o
o
oo
o o
ooo
o
o
o
o
o
o
oo o o
o ooo oo
o oo o o oo
o
o o oo o
o
o
RIVERSIDo
o
o
o
o
o
60000
40000
20000
.1
.2
.3
Share w. BAs 2000
o
o
o
.4
.5
Log Per Capita GDP
.
12
Norway
o
Israel
10
o
Brazil
o
o
o
o
o
o
o o
o
o
o
o
o
o
Thailan
o
o
U.S.A.
o oo
o
o o
o
o
o oo
Portugal o
o
o
o
oo
o
o
o
ooo
o
Singapor
o
o
o
o
o
o
8
Kyrgyzst
6
300
400
500
PISA MATH SCORE
600
Index of Earnings for Urban Areas
3
2.5
Earnings Index
2
1.5
1
China
India
USA
0.5
1
2
3
4
5
Quintile of Years of Education
US data is from the 2000 IPUMS. India data is from the 2001 Census. Chinese data is from the Household Survey Income Project of
2002.
Relocation of Departments in the 50s
(joint with Lu Ming)
11
3
0
0
8
210
0
9
0
17
21
0
8
5
0
24
4
13
11
29
21
14
0
16
7
13
14
20
(20,29]
(13,20]
(8,13]
[0,8]
No data
8
0
8
Joint with Yueran Ma
Qujing
6
Chongzuo
0
2
4
Hezhou
Lanzhou
Laibin
Guangyuan
Bazhong
Suining
Loudi
Baiyin
Zhangjiakou
Nanchong
Dazhou
Yiyang
Mianyang
HeyuanHechi
Yaan
Deyang
Xiangtan
Tianjin
Chengdu
Yichang
Guyuan
Yulin
Kunming
Ziyang YuxiBaoshan
Xiaogan
Nanning
Suizhou
Hefei
Xiangfan
Leshan
Fuyang
Chongqing
Qingyuan
Guangan
Meishan
Zhangjiajie
Meizhou
Hengyang Yulin
Shangqiu
Hebi
Guigang
Shaoyang
Beihai
Yangjiang
Zhoukou
Maoming
Bengbu
Zhumadian
Tangshan
Datong
Tongling
Nantong
Nanyang
Shenzhen
Chaohu
Ankang
Shiyan
Hanzhong
Changsha
Huangshi
Yibin
Liuzhou
Xinyang
Zhuzhou
Qinhuangdao
Changde
Xuzhou
Longnan
Shijiazhuang
Heze
Qingdao
Weihai
Jieyang
Luohe
Wuhan
Xiamen
Shantou
Zigong
Maanshan
Beijing
Jining
Liuan
Xianning
Shuozhou
Suzhou
Yueyang
Huizhou
Zhengzhou
Laiwu
Yongzhou
Luzhou
Guangzhou
Zaozhuang
Anshun
Zhanjiang
Jingmen
Tianshui
Cangzhou
Huaihua
Sanya
Baoding
Yunfu
Taian
Chuzhou
Anqing
Guilin
Yangzhou
Jingzhou
Xingtai
Yangquan
Quanzhou
Chenzhou
Xuchang
Linyi
Xianyang
Pingdingshan
Shaoguan
Ganzhou
Taizhou
Pingliang
Hengshui
XianHuanggang
Puyang
Jiujiang
Taiyuan
ZhongShan
Chizhou
Handan
Qingyang
Wuzhou
Suqian
Wuhu
Anyang
Zhaoqing
Langfang
Sanmenxia
Ningde
Baoji
Luoyang
Jinan Taizhou
Huaian
Jiangmen
Shangluo
Tongchuan
Haikou
Xinxiang
Putian
Pingxiang
Liaocheng
Dongying
Longyan
Zibo
Zhangzhou
Yantai
Dezhou
Xinzhou
Weinan
Kaifeng
Jiaozuo
Zhoushan
Lvliang
Yanan
Wenzhou
Nanjing
Chaozhou
Dongguan
RizhaoWeifang
Linfen
Foshan
Yuncheng
Jinzhong
Shanghai
Binzhou
Nanchang
Shangrao
Shaoxing
Hangzhou
Zhenjiang
Changzhou
Xuancheng
Ningbo
Changzhi
Jingdezhen
Wuxi
Sanming
Yingtan
Jinhua
Yichun
Xinyu
Jincheng
Quzhou Huangshan
Lishui
Nanping
Suzhou
Huzhou
Jian
Jiaxing
Fuzhou
0
.5
1
1.5
# of Jinshi in Ming/Bianhu
2
2.5
Chinitz: Contrasts in Agglomeration:
New York and Pittsburgh
Economic Growth and Firm Size
0
.5
1
1.5
2
MSA Employment Growth (1977-2010)
by Average Firm Size (1977) Quintiles
1
2
Smallest firms are in Quintile 1
3
4
5
Copper Mines and Entrepreneurship
Logan
Airport
Downtown
South Boston
Waterfront
Innovation District
Policy Related Questions
• Explicit Spatial Policies
– Should be city sizes be limited? Optimal city size?
• Investment in Infrastructure
– Roads and railroads but endogeneity is difficult
• Education and human capital spillovers
– Old questions remain (primary vs. secondary)
– New crop of great experimental work.
• Regulation, credit market, labor market matching
questions all relate to cities.
• Encouraging Entrepreneurship in Cities
– One Stop Permitting
Government Effectiveness and
Urbanization
-2
-1
0
1
2
Denmark
Singapore
Finland
SwitzerlandSweden
New Zealand
Canada
Netherlands
Australia Hong Kong SAR, China
Norway
Luxembourg
Austria
Iceland
Belgium
Germany
UnitedFrance
Kingdom
Barbados
Cyprus
United
States
Japan Macao SAR, China
Ireland
Portugal
Slovenia
Israel
Estonia
Malta
Korea,Chile
Rep.
Qatar
Czech
Republic
Malaysia
Brunei
SpainDarussalam
Slovak Republic
United Arab Emirates
Hungary
Mauritius
Lithuania
Croatia
Latvia
Uruguay
Bahrain
Poland
Greece
Botswana
South Italy
Africa
Tunisia
Turkey
Costa
Rica
Trinidad and Tobago
Jordan
Mexico
Jamaica
Thailand
Panama
Kuwait
Namibia China
Bulgaria
Brazil
Armenia
India
El
Salvador
Colombia
Ghana
Serbia
Sri Lanka
Philippines
Rwanda Guyana
Morocco
Kazakhstan
Albania
Cuba Saudi Arabia
Romania
Indonesia
Lesotho
Vietnam
Egypt,
Arab
Rep.
Argentina
Russian
Tonga
Peru Federation
Senegal
Maldives
Malawi Tanzania
Belize
Benin
Mozambique
Bolivia
Republic
MoldovaSyrian Arab
Dominican
UgandaKenya
Iran,
IslamicRepublic
Rep.
Gambia, The
Algeria
Guatemala
Honduras
Mongolia
Niger
Ecuador
Swaziland
Papua
New
Guinea
Zambia
Pakistan
Cambodia
Ukraine
Mali MauritaniaCameroon
Nepal
Gabon Venezuela, RB
Bangladesh
Paraguay
Kyrgyz RepublicFiji Nicaragua
Burundi
Lao PDR
Tajikistan
Yemen,
Rep.
Cote d'IvoireCongo,
Sierra Leone
Iraq Rep. Libya
Liberia
Sudan
Togo
Afghanistan Zimbabwe
Central African
Haiti
Republic
Myanmar
Congo, Dem. Rep.
0
.2
.4
.6
Urbanization Share
.8
1
GDP Under $1500, Pop>2 Million
0
Figure 5
India
Rwanda
Ghana
Vietnam
Lesotho
-.5
Malawi
Uganda
-1.5
-1
Kenya
Senegal
TanzaniaMozambique
Benin
Kyrgyz
Republic
Niger
Papua New Guinea
Bangladesh
Pakistan
Zambia
Mali
Nepal
Lao PDR
Cambodia Tajikistan
Mauritania
Yemen, Rep.
Burundi
Sierra Leone
Afghanistan
Cameroon
Cote d'Ivoire
Liberia
Togo
Central African Republic
Zimbabwe
Haiti
-2
Congo, Dem. Rep.
.1
.2
Source: World Bank
.3
.4
% Urbanization, 2010
.5
.6
Waste and Fraud in Infrastructure:
The Institutional Challenge
Author: Branille
Engineering vs. Economics
Transportation and Congestion
• The traditional literature used engineering estimates
and engaged in cost-benefit analysis
– Fundamental conclusion is “bus good; train bad”
• A newer literature uses cross-metropolitan area
estimates to determine the impact of new
infrastructure projects.
– Baum-Snow (military map) on suburbanization
– Duranton-Turner on the fundamental law of highway
traffic and impact of roads on MSA gdp
• A parallel literature in developing economies
– Banerjee, Duflo and Qian (2012).
• Requires the exogenous location of infrastructure
Anarchy vs. Authority
RayKelly by David Shankbone
Photo by SuSanA Secretariat
The Brazil Model:
The Dentist and the Supermarket
Supermarket by Wonderlane
NIMBYism vs. Monumentalism
Astana by ChelseaFunNumberOne -
Housing Markets,
Property Rights and Regulation
• Successful cities have high land costs which translates into
high costs of living.
– Housing costs are mediated by levels of regulation
• Impact of property ownership on outcomes (DeSoto, Erica
Field) through self-protection or ability to finance new
investment.
• Impact of structure on outcomes through health– the bore
holes problem and the Tenements law.
• The Western pattern was that property rights were
developed first (12th-18th centuries) but regulation
followed. Not so in the developing world.
• Large unregulated communities with dimly defined
property rights.
Median Housing Value by Population Growth
New York County, New York
Pitkin County, Colorado
Nantucket County, Massachusetts
0
Marin County, California
San Mateo County, California
Santa Clara County, California
-.5
0
.5
Population Growth, 2000-2010
1
.25
Change in Housing Prices, 2001-2006 vs. 2006-2011
0
Houston
-.25
New York
DC
-.5
Detroit
-.75
Phoenix
-1
Las Vegas
0
.2
.4
.6
Change in FHFA Price, 2001-2006
.8
Fact 1 (obvious one):
Price increases are spectacular
Source: Wharton/NUS/Tsinghua Chinese Residential Land Price Indexes
.2
0
.1
Density
.3
.4
Fact 2 (relatively obvious):
Prices are not cheap compared to income
0
2
4
6
Annual Disposable Income/Price per m^2 (2005)
8
A conservative estimate: In many of the 35 major cities by 2005, average
annual disposable income is worth around 2-3 square meters (20-30 square
feet) of housing . It is even worse now.
10
Fact 3 (not so obvious):
Prices are way higher than physical costs of construction
8
Beijing
6
Guangzhou
Hangzhou
Nanjing
Shanghai
Chengdu
Shenyang
Changsha
Nanchang
Jinan
Shijiazhuang
Zhengzhou
Chongqing
Tianjin
XiAn
Urumqi
Kunming
Changchun
GuiyangHefei
Haikou
Lanzhou
Hohhot
Yinchuan
Xining
2
4
Nanning
Wuhan
5000
10000
Price (RMB/m2)
15000
20000
Fact 4: Real estate is the largest component
of household wealth in China
Not an entirely fair comparison as real estate asset is perhaps the largest
component of household wealth for US households outside of the highest
80
Fact 5: Even when things turn sour,
prices don’t adjust immediately
60
XiAn
Nanchang
Fuzhou
Changsha
XiningUrumqi
Lanzhou
Xiamen
Kunming
40
Chengdu
Chongqing
HefeiZhengzhou
Jinan
Ningbo
20
Qingdao
Shenyang
Guangzhou
Hohhot
Wuhan
Changchun
Nanning
Shijiazhuang
Guiyang
Haikou
Hangzhou
Taiyuan
Total
Nanjing
Yinchuan
Dalian
Shenzhen
Harbin
Beijing
Tianjin
0
Shanghai
5
10
15
20
Areas Not Sold/Recent 3Years Areas Completed (Res)
Some moderate correlation between area not sold and price
declines as of 2012, but literally no correlation before 2011.
25
Image by QuarterCircleS
The Boston Hypothesis
Image by Ramy Raoof

similar documents