84259



                                                                         Overview	

                            The	Urban	Imperative:	Toward	Shared	Prosperity		

                                                    Edward Glaeser, Harvard University

                                                     Abha Joshi-Ghani, The World Bank


                                                                           November 2013 


WHY RETHINK CITIES? ........................................................................................................................................  2 
RETHINKING CITY STRATEGIES ............................................................................................................................  4 

    THE URBAN IMPERATIVE: UNLEASHING ECONOMIC CHANGE ....................................................................................................              4 
                                                                                                                                                                     10 
    TOWARD A MORE INCLUSIVE CITY .................................................................................................................................... 
                                                                                                                                                  13 
RETHINKING CITY INTERVENTIONS .................................................................................................................... 

    THE PROVISION OF CORE URBAN SERVICES .........................................................................................................................              13 
                                                                                                                                                                                14 
    HOUSING .................................................................................................................................................................... 
                                                                                                                                                                                15 
    LAND USE ................................................................................................................................................................... 
    APPROACHES TO AFFORDABILITY .....................................................................................................................................           17 
    THE CHALLENGE OF FINANCING ENVIRONMENTALLY SOUND  INFRASTRUCTURE ..........................................................................                                 17 
                                                                                                                                     21 
CITIES IN A NATIONAL SYSTEM: THE SPATIAL DYNAMICS ................................................................................... 

AREAS FOR FUTURE RESEARCH                                                                                                                          22 
                         .......................................................................................................................... 

                                                                                                                                                               24 
CONCLUSION .................................................................................................................................................... 
REFERENCES                                                                                                                                                      25 
          ...................................................................................................................................................... 
 

                                                       




This paper is an overview of a collection of papers prepared for the World Bank's Sixth Urban Research
and Knowledge Symposium, October 2012. Financial support for this paper was provided by the Urban
Development and Resilience Department and the Research Support Budget of the World Bank. The
findings, interpretations, and conclusions are the authors' own and should not be attributed to the World
Bank, its Executive Directors, or the countries they represent.
 
Some 2.7 billion people will move into cities by 2030. The vast majority of this migration will
take place in developing countries. This transformation represents an unprecedented
demographic shift with profound implications. Urbanization is undoubtedly a key driver of
development—cities provide the national platform for prosperity, job creation, and poverty
reduction. But urbanization also poses enormous challenges—challenges that we are all familiar
with: congestion, air pollution, social divisions, crime, the breakdown of public services and
infrastructure—and the slums that one billion urban resident’s call home. Urbanization is
perhaps the single most important question in development today. How do we empower city
leaders to harness the unprecedented urbanization taking place today for sustainable and
inclusive growth? How do we address the vast inequalities of access and opportunity that cities
present today? How do we avoid the mistakes of the past to ensure that the aspiring cities of the
developing world grow in a way that is inclusive, equitable and sustainable and brings shared
prosperity?

It is clear that cities have not performed as well as can be expected in their transformative role
for more livable, inclusive, people-centered, and sustainable development. But they have
enormous potential as growth escalators, offering the opportunity to lift millions out of poverty
and serve as centers of knowledge, innovations and entrepreneurship. Cities in both the
developed and developing world want to attract more entrepreneurs and create more jobs. People
want affordable housing, less congestion and pollution, and improved access to education,
health, transport, and water and sanitation. Cities also need to be resilient to natural hazards and
the impacts of climate change. If these are left unaddressed, cities will become part of the
problem rather than the solution. Business as usual is not an option. So, what should be done—
and why do we need to get it right?

Why	Rethink	Cities?	
In 1899, Adna Weber began his magisterial survey of urban growth by observing that “the most
remarkable social phenomenon of the present century is the concentration of people in cities”
(Weber 1899). Yet when Weber wrote, populations remained predominantly rural, even in
wealthy countries, and poorer nations were overwhelming rural, as they had been for millennia.
At the end of the 20th century, the developed world was overwhelmingly urban, but more than
60 percent of the populations of Africa and Asia were still rural.1 The 21st century has already
witnessed the halfway point where more than 50 percent of the world’s population is now urban,
and in this century, almost every nation is likely to become predominantly urban.

The great transition from farm to city is filled with economic, social, and political promise. Few
countries have transitioned from poverty to prosperity without urbanizing. The concentration of
people in cities can offer both economic opportunity and the chance for social change toward a
more open, innovative, inclusive, and democratic society. The easy spread of knowledge within
dense areas doesn’t just enable computer programmers to enter the global economy, it also




                                                 2
enables the spread of new ideas about equality and opportunity and gives voice to multiple and
diverse stakeholders, who are the lifeblood of political reform.

While urbanization can mean economic dynamism, inclusion and democracy, it also creates
enormous challenges, and the difficulties of urban life are magnified when the public sector is
weak. Urban proximity enables the exchange of bacteria just as easily as the exchange of new
thoughts. It took centuries for the cities of the West to become places of good health, rather than
contagious disease, but congestion and high housing prices continue to bedevil the world’s more
prosperous urban places. The cities of poorer nations, especially those with more problematic
public sectors, remain beset by enormous challenges.

Despite these difficulties, urbanization will, and should, continue. Widespread prosperity seems
much more likely to emerge from new cities than from the farms of the less developed world.
The right response to the challenges of urban life is not to try to confine aspiring urbanites to
rural poverty, but to defeat the demons that come with density.

Cities can be understood as the product of a triad of forces. First, and most obvious, cities are
physical entities—streets, buildings, and subways. Visitors see these realities first, and while
they are only part of the city, they shape everything else that occurs within an urban area. The
physical bones of the city help determine the cost of housing, the integration of the rich and poor,
the ease of getting around, and the quality of air and water. The second force is the public sector,
which defines the borders of the city and is responsible for providing clean water and public
safety. Public regulations shape private entrepreneurship and the buildings that give any city its
skyline. Close proximity increases human connections, for good and for bad, and the public
sector helps manage those interactions, allowing the benevolent ones to flourish and limiting the
more damaging consequences of urban interaction.

The third force is the magic of human interaction. Certainly cities are physical entities and
governmental units, but the real heart of a city is its people. Shiny buildings do not mean urban
success—true success means that ordinary humans are finding success, by finding each other
within the dense urban enclaves. These interactions generate innovations in art, technology,
religion and politics, as they have for thousands of years since Pericles’s Athens and Kautilya’s
Taxila.

The goal of this essay is to explore how these forces come together and shape cities, and how the
public sector can play a more positive role in enabling cities of the developing world to flourish.
Too many cities in Sub-Saharan Africa continue to suffer from the oldest urban scourge—
unclean water. Crime and murder turn many Latin American neighborhoods into places of terror
rather than opportunity. Limited transport options can turn daily commutes in Asia’s megacities
into arduous treks. Shantytowns are a regular sight in many of the world’s burgeoning cities.

There are no easy solutions. The cities of the West acquired clean water only through massive
expenditures on infrastructure. In America’s cities, the battle against crime seemed unwinnable


                                                 3
just two decades ago. Long commutes and expensive housing infect almost every thriving city.
But even if the great problems of urban life cannot be simply solved, there are better—and
worse—policy approaches to these problems. Human history provides hard-won knowledge on
how to make cities more livable.

There are many different plausible approaches to understanding urban solutions. One approach
emphasizes clearly defined public sectors—such as transportation, police, and public health—
and focuses on solutions in each sectoral silo. Yet some problems—such as economic vitality
and quality of life—span sectors and require solutions outside any one area of public
intervention. Therefore a second approach might start with clearly defined objectives, such as
longevity and economic prosperity, but these can be too broad, and different fields often differ in
their approaches. This essay presents a hybrid approach traversing both.

Rethinking	City	Strategies	
Urbanization is a key driver of development. But it also poses enormous challenges. Policy
makers and city mayors need to tackle a wide range of problems, from debilitating conditions in
urban slums to the lack of basic services such as clean water and sanitation, inadequate housing,
the exclusion of the poor from the city’s socioeconomic fabric, and the management of natural
hazards and pollution. If these problems are left unaddressed, cities can become a source of
social and political instability. But cities can instead become the engines of transformative
change toward inclusive, people-centered, and sustainable development.

The key questions examined here center on cities and economic growth and the changing
dynamics of world economic geography. Should we expect the economic success of cities like
Bangalore and Shenzhen to trickle down to the less fortunate? What policies can ensure that the
economic benefits of urbanization help a broad swath of the population?

From the economy, pivot to the physical city: the market failures that lead to sprawling
congested cities also cause environmental damage, increase greenhouse gas emissions, and
inflict deleterious impacts on the most vulnerable citizens, the urban poor. How do cities
manage urban land, housing, and transportation to become sustainable and livable? No one
doubts that cities need infrastructure, including sewage facilities, roads, and power generation.
Yet while the need for infrastructure is clear, the best means of providing it are less obvious.
Should it be provided by the public or private sectors? How can limited financing be used more
effectively? National policies matter for cities, and national policy makers can no longer afford
to ignore this if they want to harness urbanization for economic growth, job creation, inclusion,
and sustainability.

The	urban	imperative:	unleashing	economic	change	
Few cities expand without economic success. Just a handful of cities have had such powerful
noneconomic attractions that they have grown without a core economic comparative advantage.
In the United States, early Boston attracted religious migrants, and Los Angeles drew in


                                                4
Midwestern farmers interested in a more Mediterranean climate. But even in these cases, urban
pioneers wanted to earn a living, and later urbanites were attracted by the economic potential
unleashed by the first residents. A more common historical pattern is that a city forms around a
political nucleus. But even if political power is the primary source of the city’s external
revenues, people cluster around that power searching for economic advantage.

Every large city that is also a capital—from Washington, DC, to Kinshasa and Ankara—has
grown in part because of the economic activity generated by the central authority. Yet even those
cities are usually eager to see more nonpolitical sources of prosperity. In commercial cities, the
fight for economic success is even more critical.

What do we know about urban productivity? The first, and most central, fact about the prosperity
of cities is that productivity and incomes typically increase with city size and density. Glaeser
and Gottlieb (2009) provide evidence for the economies of agglomerations in the United States,
and there are many studies documenting similar patterns in other wealthy nations (for example,
Combes, Mayer, and Thisse [2008]). If anything, the connection between urban density and
prosperity seems to be even stronger in the developing world. Wages typically rise by 6 percent
as density doubles in the United States. The connection between density and incomes appears to
double in India and triple in China (Chauvin, Glaeser, and Tobio 2013). These facts just confirm
what seems obvious to the naked eye—economic growth in the developing world is uneven and
highly concentrated in cities.

Economists since Adam Smith have sought to understand why prosperity is higher in large
population centers. One hypothesis is that some areas have innate productivity advantages as
a result of natural factors, such as access to waterways or mines, and people come to cities to
be near those productivity advantages. A related view is that some cities have manmade
advantages, including the presence of national government, which attracts people. According
to this view, the link between density and prosperity reflects largely reverse causality—
productivity causes density, not the reverse.

There surely was much truth to this view historically. The ports of Liverpool, New York, and
Shanghai were major geographic advantages that helped those cities grow. Bleakley and Lin
(2012) document how U.S. cities grew at fault lines, the natural stopping point of inland
navigation. Coal mines were critical to Essen and Pittsburgh. The perceived political
advantages of British control during the colonial era were surely part of the appeal of
Singapore and Hong Kong.

The economies of agglomeration

Few scholars see natural resources as the cause of most urban productivity today. Commodities,
like coal, can be easily shipped and the natural harbors of many cities have become irrelevant.
Soil quality may explain historical density levels (Combes, Mayer, and Thisse 2008), but it is
hard to imagine that soil quality is having any direct impact on productivity today. Even in


                                                5
Africa, where urbanization is often associated with commodities, there is increasing evidence of
agglomeration economies. Instead, most scholars believe that the agglomeration produces its
own productivity, perhaps abetted by governmental factors and connections to the outside world.

The benefits of being around other people are typically labeled agglomeration economies, the
starting point for understanding the sources of urban success. Since agglomeration economies are
so basic, and so ubiquitous, it is useful to review them, for they underlie everything else that
follows. At their core, agglomeration economies occur because density increases the ease of
moving goods, people, and ideas. Cities remove the physical spaces between people and firms,
and proximity is valuable precisely because it makes connections easier.

The most basic agglomeration economy is the reduction of transport costs for goods, which is
emphasized by Krugman (1991) and others: if a supplier locates near customers, the costs of
shipping decline. Ellison, Glaeser, and Kerr (2007) show that manufacturing industries that buy
and sell to each other also locate near one another. This agglomeration economy is closely linked
to economies of scale at the firm level, because if factories were infinitely indivisible, they
would just spread out to be located near every conceivable customer. Cities enable scale
economies and proximity for trading partners.

The decline in transport costs has made this agglomeration force less important in the developed
cities of the West. For example, cities such as New York and London were once manufacturing
powerhouses, places where factories located to be close to customers and transport infrastructure.
As it became easier to ship goods, manufacturing left older, wealthier cities for cheaper locales
in suburbs or lower-cost regions or lower-cost countries. In the 1960s and 1970s, many older
cities had to struggle with the challenges created by this deindustrialization. A similar trend can
be observed in some developing countries today, where manufacturing is gradually moving out
of cities in search of cheaper locales.

But even if manufacturing is now far less important to San Francisco and Milan, it remains a
mainstay for urban areas in the developing world. Developing countries are at an earlier stage
of development, and lower labor costs give them an advantage in producing manufactured
goods. Moreover, transportation networks are far less reliable in the developing world, which
give a greater edge to urban areas, especially if they are also well-functioning ports. It is far
easier to ship from rural Ohio than from rural Maharashtra, so new manufacturing is far more
likely to locate in Mumbai than in Cleveland.

Specialized business services

Adam Smith focused on the benefits of the division of labor, which he connected with urban
densities, noting that in “the Highlands of Scotland, every farmer must be butcher, baker, and
brewer for his own family” (Smith 1905 [reprinted 2003]). During Smith’s era, factory
production was strongly associated with specialized labor, so Smith’s observation can be
interpreted as suggesting that benefits of factories were easier to exploit in cities with large home


                                                 6
markets. His exact words, however, refer to the service industry, which is particularly targeted
toward local customers, and the benefits of specialization in urban service industries continue
even in cities that have largely deindustrialized.

At its most tangible level, Smith’s observation can be seen in the abundance of specialized
shops, salons, and restaurants in large cities. Yet in most cities, specialized business services are
even more important. Kolko (2010) documents how business services have largely replaced
manufacturing as the heart of urban economies in the developed world, and that they are
becoming an important source of employment in developing countries too. These sectors have a
natural affinity for high density locales, because they typically involve a lot of face-to-face
contact. One advantage of locating in London is proximity to an abundance of legal, accounting
and financial expertise, and that expertise can be extremely specialized, an added benefit to local
customers.

The urban edge in providing business services essentially reflects the fact that cities reduce
transportation costs for people, but there are other ways for that edge to make cities economically
dynamic—by creating thick urban labor markets (Marshall 1890). In a city with a panoply of
diverse employers, the economic difficulties of a single firm may not cause much hardship for
those firms’ employees who can readily work somewhere else. In a company town, however, the
travails of that company will lead to far more distress.

The abundance of local employers also eases the job search process, part of connecting labor and
capital. Young workers may not know what sector or employer is a good match for their talents
and inclinations. A city filled with different workplaces allows workers to try things out, to
experiment with acting before entering into a career in accounting. The fallbacks may also
support more risk taking by urban workers, who can try something uncertain, knowing that there
will be other opportunities if things don’t pan out.

The ease of connecting with other people also produces nonpecuniary benefits of urban living.
Young people come to cities, in part, to be near other young people and because cities are
marriage markets (Costa and Kahn 2000). The diversity of urban pleasures helps to make cities
even more attractive than urban wages on their own.

Speeding the flow of ideas

Perhaps the most important urban advantage, however, comes from speeding the flow of ideas.
Jane Jacobs (1969) considered cities the well-spring of innovation because they enable old ideas
to be transformed into new ideas. There is an abundance of examples, from the creation of the
skyscraper in 1880s Chicago to the development of modern finance in post-1970 New York, of
how personal connections among innovators helped create a collaborative chain of brilliance.

There are at least two distinct ways for workers to benefit from the free flow of ideas in cities.
First, when ideas generate entrepreneurship and jobs, the extra labor demand increases wages.


                                                 7
Second, when the workers learn from the people around them, their human capital increases,
and that makes them more productive. Both forces tend to raise wages in urban areas. One
piece of evidence supporting the second hypothesis is that workers who come to cities typically
see their wages rise slowly with their experience in the city, rather than instantly increasing to
the average in the area (Glaeser and Mare 2001). This pattern is perhaps most compatible with
the view that cities are drivers of human capital, where workers become skilled by being
around other skilled people.

General agglomeration economies should occur everywhere, but policy makers should seek to
understand what forces are most powerful in their community. The more critical question for
policy makers is how to harness related forces that explain differences across space. Empirical
work has documented two major factors that determine the success of cities: education and
entrepreneurship. In India, the quality of infrastructure and the education of the local labor force
are the strongest predictors of city competitiveness for new firms (Ghani, Kerr, and O’Connell
2013).

Spillovers of knowledge

In the 1980s, the new growth theory, led by scholars such as Paul Romer and Robert Lucas,
focused on knowledge spillovers as the source of increasing returns economywide that enabled
long-run economic growth. The focus on the flow of knowledge pointed toward both education
and cities.

In the United States, workers who live in more educated cities typically earn more, even holding
their years of schooling constant (Rauch 1993). And the connection between an area’s education
level and earnings appears to be even stronger in China and India than it is in the United States
(Chauvin, Glaeser, and Tobio 2012). Policy makers now focus on job creation in cities in
addition to increasing productivity and raising incomes.

The Moretti paper sifts through the evidence on education as a source of local economic success.
Moretti’s earlier work (2004) has shown how educational institutions appear to be particularly
powerful predictors of urban productivity. U.S. cities that had land grant colleges before 1940
have experienced higher wages and faster population growth in recent decades. These facts
lend support to Senator Moynihan’s claim, “If you want to build a world class city build a great
university and wait 200 years.”

But most public officials have neither the ability to build a world class university nor the
patience to wait 200 years. So, the importance of education must lead to a different set of
policy levers. The education of an area reflects both selection, the in-migration of the skilled,
and treatment—the education of the citizenry. Education strengthens a nation as a whole;
therefore, attracting more skilled migrants means strengthening one community at the expense
of another. In many cases, better local school systems will do double duty both by building a



                                                 8
more educated workforce and by attracting skilled parents interested in better training for their
children.

What policies will attract more skilled workers? The literature on this question is limited, but it
seems to call for investing in amenities that are particularly relevant to the more skilled. One line
of thought emphasizes core urban amenities such as safe streets, good schools and swift
commutes, which should be valued by everyone. A second view focuses on urban attributes, like
the arts and a thriving nightlife, which may be particularly attractive to more educated workers.
The threat of inducing an exodus of the most skilled remains a major argument against attempts
at local redistribution.

The literature on education is far larger, but there remains relatively little consensus on the
determinants of a first-rate education system. High-quality teachers appear to be a primary
ingredient in educational success (for example, Chetty, Friedman, and Rockoff [2011]), but
better management and governance are also important. In the developing world, the problem
of teacher absenteeism remains particularly acute. In some countries, like the United States,
charter schools have achieved remarkable success, especially where public school systems are
underperforming (Angrist et al. 2010). Similarly, the mushrooming of private schools in India
is meeting the gap created by underperforming and low-quality public education in many of
the country’s urban areas.

Nurturing entrepreneurs

While the skills learned in school are surely important, the skills acquired out of school are likely
to have an even greater impact. While many different skills are assuredly important to urban
success, one particularly important knowledge base concerns entrepreneurship. The most
traditional forms of economics tend to assume that the supply of entrepreneurship is constant
across space, but since Chinitz (1961), economists have been aware of at least the possibility that
the supply of entrepreneurs differs across space.

In Chinitz’s (1961) comparison of New York and Pittsburgh, New York appeared to be more
resilient even in the 1950s—in part thanks to the supply of entrepreneurship, a legacy of the
city’s industrial past. New York’s largest industry had been garment production, which had few
barriers to entry and provided a starting point for thousands of would-be entrepreneurs. Those
entrepreneurs gained experience not just in making clothes, but also in starting a business—and
they often took those skills into different industries, including banking and real estate. They also
appear to have passed along some of those skills to their children. Pittsburgh, by contrast, was
dominated by large firms, such as U.S. Steel. These companies may have been extremely
efficient, but they tended to train middle managers rather than entrepreneurs. According to
Chinitz (1961), the absence of entrepreneurs in Pittsburgh thus contributed to making the city less
effective at reinventing itself.




                                                 9
Typically, area-level entrepreneurship has been measured either using average firm size,
controlling for industry, or the share of employment in newer establishments. Both measures are
strongly correlated with subsequent employment growth, both across and within metropolitan
areas (Glaeser, Kerr, and Ponzetto 2009). Chinitz’s hypothesis that natural resources, such as
coal mines, led to heavy industries with fewer entrepreneurs and less capacity for urban
regeneration also seems to be supported by the data (Chatterji, Glaeser, and Kerr 2013).

While the statistical work on entrepreneurship is at an early stage, the important role of
entrepreneurship in generating local success is also supported by abundant urban history.
Henry Ford played an outsized role in Detroit’s history, and the resurgence of Seattle since the
1970s seems linked to an abundance of start-ups, including Microsoft, Amazon, and
Starbucks. Entrepreneurs, like most people, are not perfectly mobile, and they often locate
their businesses where they have spent their early adult lives. These businesses can often
become mainstays of local employment, often for decades.

But even if the value of entrepreneurs is widely accepted, it is less clear what that implies for
local policies. This is the topic of the Lerner essay in this series. One activist school sees a role
for large-scale government leadership in funding entrepreneurship and establishing a spatial
cluster. Public support did play some role in creating Silicon Valley and the Bangalore Cluster.
An alternative school argues that entrepreneurship is far too complicated for the public sector to
effectively micromanage and instead argues for a more laissez-faire approach, emphasizing
limited regulations and low taxes. A middle position focuses on improving the local business
environment, improving physical infrastructure, attracting skilled workers, funding basic
research, and providing an appropriate legal infrastructure to allow that research to be
commercialized.

Toward	a	more	inclusive	city	
Successful urban entrepreneurs can earn billions, but urban success requires that prosperity
comes to more than just a few. Inequality is a long-standing feature of urban life. Plato wrote,
“Any city, however small, is in fact divided into two, one the city of the poor, the other of the
rich” (Plato 363BCE [reprinted 2012]). In a highly unequal world, we shouldn’t be surprised that
cities are also unequal, especially when poorer people can choose to come to cities. Urban
poverty reflects, in part, the enduring attraction that cities have for poorer people, because cities
provide economic opportunity and often a more robust safety net. Yet we should still worry
about whether cities are doing enough to provide economic opportunity for everyone.

The great-growing cities of the developing world are attracting millions of people from poor,
rural areas. This migration process makes urban poverty inevitable. When those cities succeed,
the flow of poorer people only increases, as it does when cities help poor people become more
prosperous. So, successful cities cannot eradicate poverty, but they do provide opportunities to
reduce the number of poor people within their borders. Wealthy countries often have
communities without poverty, like the United States’ Greenwich, Connecticut, but such places


                                                 10
are uniformly prosperous because they are unaffordable and have no space for the poor. That
certainly is not success.

A more inclusive city is not without slums and poverty, but it does manage to regularly enable
poor people to escape poverty. It creates jobs and provides ladders of wealth and occupations
that provide a means of moving toward prosperity. There are many causes of exclusion within
metropolitan areas. Neighborhoods differ in core amenities, including access to public transport,
employment, or parks and safe public spaces. The rich are often willing to pay more for such
local assets, and that leads to clusters of wealth in a city’s most desirable neighborhoods, and
clusters of poverty where prices are lowest. If the rich are also willing to pay more for the
privilege of living around other rich people, possibly because of better local schools or less
crime, this will also lead to segregation by income.

As natural as some sorting by income within an urban area might be, the isolation of the poor in
specific neighborhoods works against the ideal of the inclusive city. After all, isolated pockets of
poverty can seem very far from the thriving heart of a city. Children raised in segregated slums
would seem to lose the benefits of being part of a connected city. Unsurprisingly, social reforms
have long desired to build cities with more social and economic integration.

The segregation of the poor should be a concern—divided cities are neither productive nor
sustainable. At the least, governments should worry about public policies that might contribute to
segregation, such as artificial differences in the levels of public services across space or land-use
controls that make it difficult to build affordable housing in more expensive neighborhoods.
Public policies must address the unnecessary segregation resulting from significant differences in
the availability of public services over space.

The returns to education

Historically, education has been a primary tool for creating inclusion at the level of the city and
the country. The returns to education are typically high in the developing world, and education
provides a far more sustainable path toward social mobility. Despite the extremely high
motivation of many students, education in developing cities is often beset by significant
difficulties, including limited resources, teacher absenteeism, and extreme heterogeneity in
student preparation. These problems can be even more extreme in the rural areas of the
developing world.

However, this note is not an education policy primer, and does not take any stand on the
appropriate path for education reform. This is a golden age of education research, where
randomized trials and other statistical methods are enabling us to learn much more about how to
make teaching effective. In some cases, private providers, such as charter schools in the United
States, have achieved remarkable results. Other work shows that teacher quality is of enormous
importance and teacher incentives can make a great difference. There are clearly different paths
toward educational quality, and local circumstances may dictate which path will be more


                                                 11
effective within any country. The point is that improving education, particularly for the poorest
in society, is likely to be the most effective way of ensuring more inclusive cities.

The mix of industries

The Sassen essay approaches inclusiveness by contemplating the industrial mix of an area, and
specifically discusses the connection between services and manufacturing. There are two
important ways the industrial mix relates to earnings for the less advantaged. First, some
industries may simply have more demand for less-skilled workers than others, and this will tend
to increase earnings for the less skilled. Indeed, one cause of the rise in inequality in the West
since the 1970s has been a relative decline in industries and occupations, especially in
manufacturing, which disproportionately employed those with less education. The rise of urban
manufacturing in the developing world, by contrast, increases the demand for less-skilled
workers.

Wage differences alone do not tell much about the quality of the jobs on offer. One reason that
service sector jobs traditionally offered lower wages than manufacturing jobs, for comparably
skilled workers, is that the service jobs had less physical discomfort than traditional
manufacturing or mining. And an increase in demand for less-skilled workers in the city will also
tend to attract less-skilled workers, limiting the local wage effects of the increase in demand.

From a national perspective, there is nothing bad about the fact that rural migrants come to take
up urban jobs seeking better economic opportunities and a better life for their families. However,
as the Sassen essay reminds us, current urbanites will often oppose their migration. Antipathy
toward urban growth is then reflected in overly draconian land-use policies or other rules
favoring insiders over outsiders. In some cases, as in China, internal migration is limited by
rules, and the migrants that do arrive are often given far fewer privileges than the insiders.
Managing the conflict between current insiders and future migrants will be a major political task
for developing world cities for decades.

But the industrial mix affects worker productivity in a second and perhaps a more important way.
A job is not just a paycheck; it is also an opportunity to learn. The disadvantaged benefit most
from jobs that give them skills to help them move up. One plausible interpretation of the
apparent success of industrial policy in Singapore is that, as workers moved from textiles to
electronics and then up the quality ladder, they acquired human capital as they went. In this
view, industrial policy is really human capital policy.

Despite Singapore’s success, there are good reasons to suspect that industrial policy will more
often produce cronyism than concrete benefits. Economists for centuries have urged a more
laissez-faire approach to managing firm growth. But even if a government wants to avoid
picking winners, choices regarding infrastructure and investment will inevitably favor one
industry or another.



                                               12
The overwhelmingly large informal economy

One major feature of developing world cities is that the poorer laborers often work in an informal
sector that is free from regulations and taxes, but where workers lack access to public services
and some parts of the safety net. The Ghani-Kanbur essay points out the overwhelming scale of
the informal economy, which reflects a state that engages in ambitious regulations, but lacks the
ability to enforce them. While we may wish that every government had the inclination to engage
in only limited, sensible regulations and the ability to enforce them, this is not the case in much
of the developing world. In the world’s poorer cities, the informal sector is an escape valve,
enabling entrepreneurs to succeed and employ less-advantaged workers despite significant
regulatory barriers.

The Ghani-Kanbur essay finds that, in Indian cities, the informal sector is increasing in size
relative to the formal sector. One interpretation of this increase is that density is even more
valuable in the informal sector than in the formal sector. Since informal firms may rely more on
tacit agreements between suppliers and customers, the benefits of easy face-to-face
communication may be greater. Alternatively, informality may be more prevalent in smaller
firms that disproportionately locate in cities because of their need to partner with other firms. A
third force pulling informality into cities is that large-scale manufacturing, which tends to be
formal, is typically more land intensive and thus more suburban. Small service firms that prize
informality are more naturally urban.

Millions of less-advantaged workers benefit from the employment the informal sector provides,
but informality is far from a perfect solution. In the “gray” market, there are no rules protecting
worker safety and often little legal recourse for workers who have been maltreated. Informality
may be a reasonable response to a bad situation, and it may foster entrepreneurship and create
jobs, but that doesn’t mean there aren’t significant costs to having millions of workers without
social safety nets.

Rethinking	City	Interventions	
Cities are people and political units, but they are also physical entities, filled with vast amounts
of housing and infrastructure and deeply shaped by public policies. Barriers to building can
produce a flat city that sprawls and a proliferation of slums. New transport infrastructure can pull
a city outward, as new settlements occur along the highway or subway line. The physical
challenges facing developing world cities can be daunting, and historically many of these cities
have made decisions with severe adverse consequences. Housing, land, and transportation are at
the nexus of how a city develops and how livable and inclusive it is for those who come
attracted by new opportunities.

The	provision	of	core	urban	services	
To be inclusive, a city needs to provide more than just decent wages to its poorer workers, it
needs to provide core urban services as well, such as clean water and sanitation. Workers need


                                                13
safety and decent commutes too, and they need decent housing. These issues are discussed by
Woetzel and Pohl, who observe that basic services are often concentrated in the city core, and
access to clean water and sanitation drops rapidly as one moves out of the core. The disparities
between core and periphery are most acute for megacities. The benefits of urbanization do not
seem to extend beyond the core. While density makes it more cost-effective to provide basic
infrastructure services to urban residents, city governments often don’t collect enough revenues
to extend these services to the expanding urban population, or even to maintain existing services
for current residents.

Housing	
The Collier and Venables essay provides a panoramic look at the physical problems of African
cities, especially their housing challenge. Shantytowns have always existed in cities, though
colder climates have tended to limit people’s willingness to urbanize in completely substandard
dwellings. In warmer locations, where rural settings have become perilous due to conflict or crop
failures, migrants are willing to come to cities even if housing is wildly inadequate. The
challenge of providing decent housing involves a number of constraints—the limited supply of
urban land, the lack of property rights, outdated and unpractical regulations on minimum lot
sizes and floor area ratios, as well as weak construction sectors and a lack of suitable housing
finance mechanisms for the poor and the middle class.

Sonia Hammam’s essay provides a critical review of the nature of housing markets throughout
the world. Housing is really a combination of two attributes—a physical home and a location.
Housing values differ so much within and across cities precisely because locations differ so
much in their value. Hammam’s essay underlines that the fact that housing markets are
essentially local, so interventions that provide more public housing in a far-flung suburb do little
to ease the cost of living within a dense city. The key to solving the urban housing problem is not
just to deliver more cheap housing—it is also to make sure that housing is in places that provide
plenty of value for residents. Understanding the constraints on supply of adequate housing and
the rising demand for affordable and adequate housing is important.

Shaping the demand for housing is the demand for structures and particular locations, but
housing policy often does more to influence housing supply than housing demand. Supply is
shaped by building technology, land availability, and the rules that influence construction. In
some cases, those rules are widely flouted, so cities can develop an ocean of substandard housing
despite having extremely tight building rules—Mumbai is one such example. Similarly, the
provision of public subsidized housing in far-flung places, away from jobs and adequate
transport, fails to generate complementary private investments in housing maintenance or
upgrading and creates inefficient and divided cities.

The supply of housing is built up over time, and understanding supply means understanding both
the stock of homes and the flow. The flow of homes is determined by the combination of supply
and demand, and then the stock represents the history of past flows. Typically, the flows are


                                                14
positive and the stock gradually increases, but there are times when the stock can be radically
reduced either through planned demolitions or natural disasters. A radical reduction in the stock
of housing due to a calamity can represent one of the greatest urban challenges.

Both rules and building technology determine the speed of the flow of housing supply. Complex
multistory dwellings can take many months or years to build. A Byzantine permitting process
can double or triple the construction time. Conversely, simple self-made or prefabricated homes
can be put up in days, especially if there is no effective regulation. The speed of building can be
considerably faster if residents are doing the building, avoiding the need to bring in outsiders to
build more formal structures.

Collier and Venables essay also reminds us that housing in many developing countries is often
marked by physical and legal inadequacies. Physically, the housing can be unsafe and lacking in
core amenities, including running water and electricity. Legally, many residents lack clear title to
their property, usually because homes were built without the permission of the land’s owner.
Developing world cities have an enormous informal housing market, just as they have an
enormous informal labor market, and the results are similar. The informal housing market
provides usable space for poorer people who could not otherwise live in the city, but the
residents of this housing have few of the protections of a more legal market.

As we contemplate solutions to the housing problems of the developing world, Collier and
Venables wisely warn that we should be cautious about applying wealthy world standards to
poorer nations. Housing that would be torn down in New York or London would often be seen as
luxurious by the poorer residents of Addis Ababa. If housing is kept to standards that seem
adequate to the wealthy world, the price will be prohibitive for poorer urbanites if they must pay
for it themselves. If the housing is provided by the government, the public sector both risks
crowding out other potentially more valuable forms of social spending, like education, and
creates powerful, artificial incentives encouraging the poor to migrate to cities.

The gulf in wealth between developing world and developed world cities also suggests the value
of planning for obsolescence. Any housing that is appropriate for Africa or India today will (it is
hoped) be inadequate 40 years down the road, when both areas are far wealthier. So, any housing
policy needs to be self-consciously temporary and adaptable to the rapidly changing nature of
developing world economies.

Land	use	
The public approach to housing in developing world cities always includes land-use regulations,
although they may be only sporadically enforced. Those regulations can have many purposes,
but one typical goal is to ensure a minimum standard of living. Yet the impact of those
requirements often increases the cost of housing precisely for people the rules are trying to help.
This limits the ability of land-use regulations to ensure better housing standards.




                                                15
It is perhaps possible to accept that poorer urbanites will live in very small, very simple homes. It
is harder to accept that these homes will be unsafe because of unsound building, have sanitary
issues related to lack of sewage and other infrastructure, or be in precarious areas prone to
natural disasters like flooding or landslides. These safety challenges are particularly hard to
solve, because poorer residents often take on considerable risks in urban labor markets and
disregard rules against living in dangerous locales.

For the developing world, slums are often built ignoring both land-use restrictions and indeed
property ownership itself. Many favelas and shantytowns were built on government land that was
not intended to provide space for housing. In some cases, settlements grow up on private land
that was being kept vacant for another purpose. There are difficult tradeoffs among the rule of
law: the preservation of space for alternative uses, or the very real housing needs of the urban
poor.

Housing involves structures as well as land, and the quest for better housing also involves
innovative means of delivering inexpensive structures. A number of private companies have
been experimenting with smaller, prefabricated homes, often using inexpensive recycled
materials. These strategies hark back to the days when Sears Roebuck sold thousands of cheap,
do-it-yourself home-building kits throughout America, and private innovation helped create
cheaper housing.

But it is hard to imagine that cheap housing structures will solve the problem of affordable urban
housing unless the legal infrastructure is also effective and sensible. Alain Bertaud’s essay turns
to the often murky world of land use planning in cities. Ideally, land-use planning should aim at
providing safe, affordable housing in a connected city, with functional transport corridors and
easy connections between jobs and housing. Yet, too often land-use systems produce sclerotic
cities that provide insufficient legal, dense housing and too many extremely arduous commutes.
The focus on static urban design plans has resulted in massive urban sprawl and the development
of new towns with little market demand.

Bertaud emphasizes that land use is closely tied to transportation planning. Housing may be the
combination of structures and location, but the value of location is determined by the ease of
access to other locations. The core model of urban housing prices, first articulated by William
Alonso almost 50 years ago, describes how density and price reflect the ease of access to jobs
at the center of the city. The importance of transportation has not diminished since Alonso’s
day (Alonso 1964). Bertaud correctly insists that we see transportation and land-use planning
as a single topic.

For example, the impact of an urban height limitation will be determined in part by the ability to
get to urban jobs from more distant locales. When a city with abundant public transportation
limits the urban core, development for rich and poor will move away from the core, people will
still be able to reach jobs, and the city will continue to grow. If only car transportation is



                                                 16
possible, the poor are more likely to try to remain at the center, leading to overcrowding, but the
rich will suburbanize. If all travel links are poor, either the city’s growth will stop or it will still
sprawl outward, but at an enormous cost of time and comfort as the new populations engage in
long, difficult commutes.

Approaches	to	affordability	
Bertaud also reminds us of the foolishness of many affordability measures. The ratio of income
to housing costs is one measure, but it is essentially arbitrary and often misleading. Consider a
city and a rural area: in the city, workers can earn $5,000 a year, but must pay $2,000 in housing.
In the rural area, housing costs only $500 a year, but earnings are only $2,500. The housing cost
to income ratio is much higher in the city, suggesting an affordability problem, but earnings after
paying for housing costs are 50 percent higher in the city, suggesting that the urban residents
have a good deal. Indeed, the logic of spatial equilibrium would suggest that if there were no
other particularly urban problems, housing costs should rise to $3,000 a year, so that after-
housing earnings were equalized between the two locales.

Another approach to housing affordability is the physical cost of delivering housing and
whether housing prices substantially exceed physical costs. This process does not reveal
whether poorer urbanites can afford the houses, but it does give some information about
whether the housing market is functioning well. If the physical costs are relatively high, this
suggests a problem in building technology. If prices are much higher than physical construction
costs, this implies either high land values or regulatory barriers to construction.

The other side of land-use planning involves commercial structures, and comes with many of the
same concerns: affordability is still desirable and safety remains a concern. Land-use regulations
also affect the physical landscape. There has, however, been something of a sea change in the
thinking on zoning for nonresidential uses. A century ago, planners were very concerned with
separating industrial and commercial from residential uses of land. They saw negative impacts of
industry on residential neighborhoods. Today, mixed-use planning is far more popular, because
planners now recognize the environmental and social costs of artificially separating people from
their work. Jane Jacobs (1969) was one of the great advocates of mixed-use cities. There may
still be cases where particularly noxious factories need to be isolated, but in many cases, mixing
rather than separating seems the more sensible policy.

The	challenge	of	financing	environmentally	sound		infrastructure	
The environment, particularly the impacts of climate change, is an important reason to focus on
minimizing travel distances. Shorter commutes mean less energy use, which is of even more
consequence if developing world cities develop automobile habits similar to the developed
world. Cities are full of instances where individuals impose costs on third parties—externalities
in the language of economics. The costs potentially related to carbon emissions provide an
example of one such externality. This series does not methodically consider the externalities
inherent in urban congestion and contagious disease, although these provide a major motivation


                                                  17
for spending on urban infrastructure. The essay by Matthew Kahn does, however, specifically
focus on cities and the environment.

Kahn reminds us that there is both local and a global side to urban environmentalism. The global
side focuses on city-level activities that may affect the entire planet, particularly through climate
change. Cities do have higher carbon emissions than many traditional rural areas, but that is true
of all more prosperous lifestyles. The relevant comparison is not between city living and rural
poverty, but between city living and less dense, but still affluent, lifestyles, like sprawling
suburbia. In that comparison, city living looks pretty green, because people drive less at higher
densities and typically inhabit smaller living spaces, even holding family size and income
constant (Glaeser and Kahn 2010). Recent literature also offers evidence of this relationship
between urban form and per capita emissions and, specifically, the association of urban density
with lower per capita emissions (Hoornweg, Sugar, and Gomez 2011).

If India and China see their per capita carbon emissions rise to the level in the sprawling United
States, global carbon emissions will increase by more than 125 percent. But if they rise only to
the level seen in wealthy but hyperdense Hong Kong SAR, China, global carbon emissions will
rise by less than 30 percent. The environmental advantages of dense living provide one reason to
rethink land-use policies that artificially limit building within the city center. Policies that limit
driving, such as Singapore’s electronic road pricing, were implemented mainly to limit
congestion for economic and social reasons, yet they also support global environmental
concerns.

Kahn is also focused on the local environmental issues that occur within cities. Air quality, water
pollution, and green spaces are three primary local concerns that help determine the local quality
of life. In each case, private individuals tend to ignore the local impacts of their actions, such as
when they deposit waste in common land areas, contaminating it, or drive polluting cars. For
cities to be pleasant as well as productive, they must pursue policies that counter these costs of
dense living.

Kahn emphasizes that developing world cities are likely to pay more attention to environmental
concerns as they become wealthier and better educated. A pleasant environment is a luxury good,
and richer people are willing to pay more for a pleasant location. The poorer residents of
developing world cities are more likely to put basic necessities ahead of parks and clean air.
Education also matters, because a more educated population is likely to be more effective at
lobbying for local concerns, including the environment.

Even if developing world cities don’t get wealthier, many of them may still have to grapple
with the challenges of climate change. If greenhouses gases lead to rising sea levels, or more
extreme weather events, developing country cities are likely to face the consequences. Nearly
two-thirds of urban settlements with more than 5 million people are located partly in low
elevation coastal zones, which contain 13 percent of the world’s population (McGranahan,



                                                 18
Balk, and Anderson 2007). Many of them lack the infrastructure needed to cope with extreme
weather events, so the possibility of major disasters is a looming challenge for the developing
world.

Alternative approaches to environmental issues

Economics and engineering offer alternative approaches to these environmental issues.
Economics emphasizes using incentives to change individual behavior. A carbon tax that makes
energy users pay for the social costs of their actions is a classic example of a tax aimed at
reducing socially costly behavior. Congestion charging is another example. Engineering focuses
on changing the physical environment to reduce the negative consequences of dense living. If
congestion charging is the economists’ approach to traffic congestion, the engineering approach
is to build more highway lanes or to provide more public transit options. For highways and
driving, the economic approach seems quite sensible: vehicle miles traveled increase roughly
one-for-one with highway miles built (Duranton and Turner 2011). If this continues to hold in
the developing world, just building more roads will not solve the traffic, since this will just
induce more people to drive. It is hard to imagine that people will start limiting their driving
unless they are charged for the social costs of driving and a viable alternative (such as reliable
public transport) is available.

But in other cases, the engineering approach is far more feasible than using economics. It is
difficult to imagine relying on private provision and taxes for clean water and sewage. How
could such a system be enforced in the informal communities of the developing world? The
engineering approach to clean water and sewage, which involves massive public investment in
infrastructure, seems the only feasible solution.

Building the right infrastructure

But providing infrastructure in the developing world is also daunting, as the Woetzel and Pohl
essay makes clear. The infrastructure need is enormous, but so is the possibility of building the
wrong infrastructure. Moreover, the building process is extraordinarily difficult, with plenty of
opportunities for corruption and mismanagement. The public sector is often ill-suited to manage
such projects, because of both a lack of technical expertise and because of the short time
horizons that often influence political leaders.

Infrastructure challenges start with project selection. For decades economists have promoted
cost-benefit analysis, and for centuries they have argued that infrastructure is best financed by
user fees. Adam Smith argued that funding infrastructure with tolls can ensure that the projects
built are also those that deliver sufficient social valued added (Smith 1905 [reprinted 2003]).
User fee financing may be politically difficult in some cases, and just plainly infeasible in others.
Occasionally the externalities that justified the infrastructure to begin with push against charging
too much for the infrastructure’s use. For example, if the goal is to eliminate contagious disease,
then charging too much for the water may push people to continue using unsafe sources.


                                                 19
Paying for infrastructure

When user fees cannot pay for the infrastructure, the push is toward conventional cost-benefit
analysis and the use of more general tax funds. Even in this case, however, it is probably sensible
to pay for infrastructure with local taxes paid by the people who use the infrastructure. Tax
increment financing, which pays for infrastructure by taxing the increases in value of
neighboring properties, provides an alternative way of making sure that the costs of
infrastructure are paid for by people who actually benefit from it.

While the basic tools for cost-benefit analysis are well understood, there are numerous problems
with implementation. Often project supporters will try to significantly overstate benefits by
overestimating future use. In other cases, costs may be underestimated. When infrastructure
creates significant benefits for the local community, it may be hard to estimate these benefits,
especially if they are diffuse. For example, it is hard to determine the value of a project meant to
make a neighborhood more attractive. This unpredictability leaves plenty of room for
manipulating estimates for political ends.

The Woetzel and Pohl essay also emphasizes the value of getting the best out of existing
infrastructure. Too often governments invest too much in new infrastructure and too little in
maintaining the existing infrastructure stock. Getting the most out of existing infrastructure can
include policies such as congestion pricing, which efficiently rations access, but politics tends to
resist such charges.

Public-private partnerships

The Woetzel and Pohl article emphasizes the need for greater technical expertise in local
governments is crucial if the government is managing the project, but it is also necessary if
infrastructure provision has been farmed out to the private sector. As the Engel and Galetovic
article emphasizes, public-private partnerships (PPPs) are not a panacea. There is plenty of room
for failure in PPPs, especially if the government ends up expropriating the private firm or if the
private firm ends up subverting the government. At their best, PPPs enable private expertise and
incentives to accomplish a socially desirable task, like building a new bridge or highway. The
basic structure is that the private builder receives the right to build the project and agrees to a
time path of user fees. Those fees then provide incentives for the private sector to maintain the
project, since it can cover its costs only if it continues to attract users. In cases where user fees do
not cover the costs, the government will need to supplement these fees with a direct subsidy.

PPPs offer tremendous promise, but there are many paths toward disaster as well. One possibility
is that the private entity will convince the government to give them far too generous a subsidy.
Any time the government is giving cash to a private entity, there is potential for corruption, since
side payments to public officials can bring huge profits. Technical expertise is helpful in
evaluating the project, but it will do little good if the official is on the payroll of the private
company. A second problem is if the government decides not to honor its contract. Once the


                                                  20
infrastructure is in place, the government has an incentive to lean on the private entity to reduce
its user charges. If firms anticipate such expropriation, they will be unwilling to make
investments in the first place. A third problem is that firms may skimp on investments that could
prevent low probability disasters. The advocates of PPPs correctly emphasize that a private firm
will not let the project deteriorate to the level where users disappear, but they may take risks,
betting that disaster will not materialize. If there is a catastrophe, the company may trust that its
liability will be limited.

These fears have materialized in many PPP projects throughout the world, as Engel and
Galetovic illustrate. Weaker public sectors are less likely to deliver projects efficiently on their
own, but they are also less effective at monitoring private providers. PPPs are well worth
investigating, but their downsides must also be anticipated.

The infrastructure needs of developing world cities are among the greatest challenges of the
developing world. Vast sums will be spent, and there is sure to be corresponding waste. In his
essay, Romer emphasizes on the role of government, new rules, the importance of local capacity
and the critical role of strong institutions and governance. Good governance is most important in
areas where the government and the private sector can be most productive. Further, the capacity
to enforce rules should be complementary to and not a substitute for market activity. The hope is
that an increasingly sophisticated local public sector will be able to manage the process so that
the lives of ordinary citizens are truly transformed with clean water, reliable electricity, and
faster commutes. But this is predicated on investing in and bolstering local capacity, as Romer
makes clear.

Cities	in	a	National	System:	The	Spatial	Dynamics	
Apart from Singapore, cities are part of a nationwide economic system that contains other cities
and rural areas. The remarkable urban success of the Republic of Korea was driven mainly by
the national government’s comprehensive planning and investment policies. National
governments make policies that will influence the growth of the city, including investments in
nationwide transport systems and more location-specific interventions. In many countries,
subnational political units, such as states or provinces, exert a fair amount of control over
individual cities. How should higher levels of government react to urban growth?

The Desmet and Rossi-Hansberg paper asks a very specific question. Should national
governments favor large cities, or should they push populations into more medium-sized cities?
As cities grow, they generate agglomeration economies, but there are also externalities
associated with greater population densities. Optimal policies toward medium and large cities
need to not only recognize the existence of these effects, but also to understand how these effects
differ by city size.

The existence of agglomeration economies on their own does not imply favoring big cities.
When people and businesses are pushed from medium to large cities, the large cities become


                                                 21
more productive, but the smaller cities become less productive. Those effects need to be weighed
against each other. If there were no negative effects of big city density, then even with these
losses, there would be a good reason to channel people toward the largest metropolitan areas. Yet
because density has downsides, the benefits must be weighed against the costs. Desmet and
Rossi-Hansberg present a tool for calibrating the relative size of the benefits and costs of city
bigness. There is an associated online tool, which national leaders can use to actually determine
whether their country might use more or less urban concentration. The tool attempts to measure
these positive and negative effects using commonly available data and then use this measurement
to predict optimal public policy.

The Henderson paper reminds us that agricultural productivity is often a precondition for
urban growth: a nation rarely moves to urban living until its farms can feed its citizens.
Henderson also notes that big factories often leave urban areas, leaving services in the urban
core. This is because in the long term, the largest cities are expensive places for industry, and
with further development, modern industry tends to decentralize from the biggest cities, first
to their peri-urban areas, and then to smaller cities and rural areas. Gradually, there emerges a
geography of production in the urban hierarchy, whereby smaller cities tend to specialize in
standardized manufacturing or certain services, while the largest cities tend have a diverse set
of business and financial services, dependent on an information-rich, dense local environment.

In practice, national leaders will rarely consider explicit policies that favor one locale over
another, except if a particular place is economically disadvantaged relative to the rest of the
country. Still, in many cases, national governments do make choices that implicitly favor one
area over another. The location of highways and rail networks lead to choices between cities, and
hopefully the Desmet and Rossi-Hansberg essay will be helpful in thinking through these
choices.

Areas	for	Future	Research	
The challenges facing the world’s developing cities also call for more research on economic
development, social mobility, and the impacts of government policies. Development research has
experienced a revolution over the last 15 years, as increasingly sophisticated methods, especially
randomized trials, have begun to emerge. In many situations these new methods have proved
crucial, but other contexts require different approaches.

While there is widespread agreement that entrepreneurship can be an important ingredient for
economic success, there is less agreement about what drives entrepreneurship or what
government policies encourage and incubate new economic enterprises. Most of the research on
entrepreneurship has been in the developed world, but even that is just beginning to focus on the
causes of entrepreneurial differences across space. An important new branch of research in the
developing world would be to document the connections between entrepreneurship and growth,
but more work is also needed that recognizes that entrepreneurship and growth may be the result
of deeper forces. More research is also needed on government policies, including regulation, that


                                               22
may affect entrepreneurship. The World Bank’s ground-breaking work on measuring regulations
can inform this research. Randomization is most likely to occur in this area around policies that
target aid to individual entrepreneurs, rather than in regulations or other more widespread
economic policies.

It is critical that more research be conducted on the impact of infrastructure on economic, social,
and health outcomes. There has been a lot of water-related research on health outcomes, on
private provision and on targeted subsidies, but even more is needed. There may be scope for
randomized trials, but large-scale infrastructure cannot be randomly allocated. In these cases, the
best option may be to hope for natural experiments that allow estimations of full-range impacts
of particular forms of development.

Human capital and skills are at the center of urban success, and there are already a large number
of researchers examining the determinants of educational outcomes. This is an area where
randomized trials are particularly useful. There are numerous examples of settings where
classrooms have been randomly selected to use different techniques or to offer varying
incentives to students. The main failing of this research is that outcomes are typically test scores,
and we are primarily interested in adult outcomes. Some research in the United States has
documented the connection between teacher quality and adult earnings, but more work in this
area is needed in the developing world.

The phenomenon of growing informality in the face of rapid urbanization is relatively new to the
research and policy-making arena. Further research is needed to identify the causes underlying
this trend, such as:

      the nature of agglomeration externalities and how they play out for the formal and
       informal sectors;
      the precise reasons for productivity in the informal sector to be lower than in the formal
       sector, especially if agglomeration benefits seem to accrue equally, if not more so, to
       informal enterprises;
      the reasons for informality to persist in the face of rapid growth and continued
       deregulation in many countries;
      how to bring informal enterprises into the fiscal net so that they can contribute to the
       services needed for them to flourish; and
      the policy measures that can be undertaken to support the informal sector in creating jobs
       and addressing poverty, given that it does not look as though job creation in the formal
       sector has been or is likely to be vigorous enough to meet the employment and poverty
       challenges of the next two decades.

Other areas for research are on how regulations restrict and constrain the provision of affordable
housing and on what can be done on both the supply and demand side for housing, including
stimulating the private sector’s provision of affordable housing. More research and city-based


                                                 23
analyses of the impact of specific regulations on housing costs, housing supply, and housing
affordability are needed.

Most knowledge on resilience today is based on rates of urban and economic growth and
infrastructure typology of the last 50 years. But the pace and magnitude of urbanization are
currently such that the built up area will triple in the next 30 years. One of the biggest challenges
today is resilience in the face of much greater uncertainty; a more rapidly changing climate;
significantly increased variability in political stability and commodities pricing; more demands
on food and water; and the rapidly changing geopolitical landscape. Research on resilience in an
integrated context—resilient systems design—is a critical aspect of urban system design.

While many relevant policy decisions are made and implemented locally, most statistical data are
collected at the national level. A lack of globally comparable data on cities has fragmented urban
research agendas and limited the growth of cumulative knowledge generation across countries,
regions, and disciplines. This in turn has limited the capacity to get fact-based understanding of
the patterns of city performance across the globe. That is why it is so important to collect
verifiable and comparable city-level data.

Finally, there also is need for research on the institutions that shape urban politics. We know
relatively little about how institutions and politics affect the economic life of the city—case
studies and more historically grounded work can provide more data in this area.

Conclusion	
Perhaps the most hopeful aspect of urban life is that cities tend to create their own solutions.
London spread cholera, but also gave John Snow the information he needed to understand the
sources of the disease. Urban connections are empowering communities throughout the world to
press for urban improvements. There is hope in the future of cities because cities have so often
done miraculous things. The cities of the West were once places of suffering, crime and grime,
but now they are thriving centers of innovation, culture and entrepreneurship, and far more
pleasant and safe. Why? Because urban leaders effected change, and that change will continue
only with better governance and stronger institutions.

For most of human history, people lived on the edge of survival. In the past two centuries, we
have miraculously moved toward far greater prosperity through transformations, above all, in
cities. Urbanization now has the potential of transforming the developing world, and that’s why
getting urban policies right is so important. There is no future in rural poverty—the path to
prosperity inevitably runs through cities. The right approach is not to accept the urban failures
that often exist now, but to rethink cities and try to imagine how to get to a brighter urban future.




                                                 24
References	
Alonso, W. 1964. Location and Land Use: Toward a General Theory of Land Rent. Cambridge,
      MA: Harvard University Press.
Angrist, Joshua D., Susan M. Dynarski, Thomas J. Kane, Parag A. Pathak, and Christopher R.
       Walters. 2010. “Who Benefits from KIPP?” NBER Working Paper No. 15740,
       Cambridge, MA.
Bleakley, Hoyt, and Jeffrey Lin. 2012. “Portage and Path Dependence.” The Quarterly Journal of
       Economics 127 (2): 587–644.
Chatterji, Aaron, Edward L. Glaeser, and William R. Kerr. 2013. “Clusters of Entrepreneurship
       and Innovation.” NBER Working Paper No. 19013, Cambridge, MA.
Chauvin, Juan Pablo, Edward L. Glaeser, and Kristina Tobio. 2013. “Urban Economics in the
      U.S. and India and China.” Manuscript in progress, Harvard University.
Chetty, Raj, John N. Friedman, and Jonah E. Rockoff. 2011. “The Long-Term Impacts of
       Teachers: Teacher Value-Added and Student Outcomes in Adulthood.” NBER Working
       Paper No. 17699, Cambridge, MA.
Chinitz, B. J. 1961. “Contrasts in Agglomeration: New York and Pittsburgh.” American
       Economic Review 51: 279–89.
Combes, Pierre-Philippe, Thierry Mayer, and Jacques-François Thisse. 2008. Economic
     Geography: The Integration of Regions and Nations. Princeton: Princeton UP.
Costa, Dora L., and Matthew E. Kahn. 2000. “Power Couples: Changes in the Locational Choice
       of the College Educated, 1940–1990.” Quarterly Journal of Economics 115 (4): 1287–
       1315.
Duranton, Gilles, and Matthew Turner. 2011. “The Fundamental Law of Road Congestion:
      Evidence from the U.S.” American Economic Review 101 (6): 2616–52.
Ellison, Glenn, Edward Glaeser, and William Kerr. 2007. “What Causes Industry
        Agglomeration? Evidence from Coagglomeration Patterns.” NBER Working Paper No.
        13068, Cambridge, MA.
Ghani, Ejaz, and Ravi Kanbur. 2013. “Urbanization and (In)Formalization.” Policy Research
       Working Paper No. 6374, World Bank, Washington, DC.
Ghani, Ejaz, William Kerr, and Stephen O’Connell. 2012. “What Makes Cities More
      Competitive? Spatial Determinants of Entrepreneurship in India.” Policy Research
      Working Paper Series No. 6198, World Bank, Washington, DC.
Glaeser, Edward L., and Joshua D. Gottlieb. 2009. “The Wealth of Cities: Agglomeration
       Economies and Spatial Equilibrium in the United States.” Journal of Economic Literature
       47 (4): 983–1028.
Glaeser, Edward, and Abha Joshi-Ghani. Forthcoming. The Urban Imperative: Towards Shared
       Prosperity. Washington, DC: World Bank.
Glaeser, Edward, and Matthew Kahn. 2010. “The Greenness of Cities: Carbon Dioxide
       Emissions and Urban Development.” Journal of Urban Economics 67 (3): 404–18.



                                             25
Glaeser, Edward L., William R. Kerr, and Giacomo Ponzetto. 2009. “Clusters of
       Entrepreneurship.” NBER Working Paper No. 15377, Cambridge, MA.
Glaeser, Edward, and David Mare. 2001. “Cities and Skills.” Journal of Labor Economics 19 (2):
       316–42.
Hoornweg, Daniel, Lorraine Sugar, and Claudia Lorena Trejos Gomez. 2011. “Cities and
      Greenhouse Gas Emissions: Moving Forward.” Environment and Urbanization.
      http://eau.sagepub.com.
Jacobs, Jane. 1970. The Economy of Cities. New York: Vintage.
Kolko, Jed. 2010. “Urbanization, Agglomeration, and Coagglomeration of Service Industries.” In
       Agglomeration Economics, ed. Edward Glaeser. NBER Books.
Krugman, Paul. 1991. “Increasing Returns and Economic Geography.” Journal of Political
     Economy 99 (3): 483–99.
Marshall, Alfred. 1890. Principles of Economics. London: Macmillan.
McGranahan, Gordon, Deborah Balk, and Bridget Anderson. 2007. “The Rising Tide: Assessing
     the Risks of Climate Change and Human Settlements in Low Elevations Coastal Zones.”
     Environment & Urbanization 19 (1): 17–37.
Moretti, Enrico. 2004. “Human Capital Externalities in Cities.” Handbook of Regional and
       Urban Economics. North Holland-Elsevier.
Plato. 363 BCE (reprinted 2012). The Republic. New York, NY: Penguin Classics.
Rauch, James. 1993. “Productivity Gains from Geographic Concentration of Human Capital:
      Evidence from the Cities.” Journal of Urban Economics 34 (3): 380–400.
Adam Smith. 1905 (reprinted 2003). The Wealth of Nations (6th ed.). New York NY: Bantam
     Classics.
Weber, Adna. 1899. The Growth of Cities in Nineteenth Century America. New York:
      Macmillan.




                                             26