MORE ON CITIES
In my latest essay for The Africa Report, I laid out some very simple arguments on Africa and the importance of urbanization. There is always room for more details on the subject and based on the feedback, I want to expand on a few points in the following note.
IS AFRICA’S FUTURE TRULY URBAN?
Relative to other regions, Africa still has a large share of its workforce in agriculture — and public finance still largely relies on the export of primary agricultural goods and natural resources. But some simple facts and extrapolations suggest that cities will be very prominent in Africa’s future. The share of the global population living in cities will reach 70% in 2050 and Africa’s share of that will increase from 16% to about 25%. Population projections predict Africa will host more than a quarter of the most populous cities in the world in the next two decades.
Urbanization levels in Africa are currently at about 50%, and with projected increases, there is a reason to conclude that most economic activity will happen in cities. About 41% of the continent’s population is under the age of 15, and there is some evidence that young people are not interested in farming. The growth of the urban service sectors is also dwarfing manufacturing and agriculture. Cities are going to be very important to future socioeconomic development in Africa.
“URBAN AND POOR”
As I have explained in the past, city-dwellers in Africa are poor relative to past urbanization patterns — and it is important to understand why this is happening. For example, Edward Glaeser thinks this might be due to food production. In the past, countries have had to grow their own food to feed a teaming urban population. But with rising agricultural productivity, some countries pursued their own comparative advantages while importing food from elsewhere to feed the urban population. This is not unique to Africa and makes Glaeser’s assertion less plausible to me.
The explanation of Africa urban poverty that I find most persuasive is by economist Remi Jedwab. The basic story is that natural resources are the main driver of “structural transformation” in Africa and this is why there is urbanization without growth. At least since the Industrial Revolution, urbanization has been driven by the “push and pull” effects of economic development. Agricultural productivity meant that a lot more food can be grown with a lot fewer farmers, thanks to technology. With excess farming labour, more people moved to cities to find other ways of making a living besides farming. Also, the growth of industries in cities with more high paying jobs meant that many people were “pulled” to cities to work in the expanding industrial sector. Urbanization during this period was driven by the transformation of patterns of economic activities. Cities were production hubs for firms.
According to Jedwab, urbanization in Africa show a different pattern because cities are characterized by consumption rather than production. First of all, agricultural productivity never really took off on the continent. The transfer of technical competence and knowledge had institutional constraints. Also, prices of agricultural produce were held subject to price controls — removing the incentives for investment in technology and expansion of the agriculture value chain. But even more importantly, trade and exports were concentrated in primary agricultural products (cocoa, rubber, etc) and natural resources like crude oil. This generated a lot of economic surplus, and increased the demand for urban services. Hence, the service sector in cities exercised a “pull effect” not just on rural agricultural labour, but also on the burgeoning urban industrial sector. If you are the landowner of a large cocoa plantation, or you are the fashion designer to the wife of the president of an oil-exporting country which got you awarded an oil block — then you have more need for banking services, insurance, real estate management than the locally made garment, shoes or even cars. This means that the services you and people like you consume will grow and demand more workers than others. This is why African cities are “consumption cities” and the continent got “de-industrialized”. Please note that it is foolish to think that the remedy to this is Import-Substitution Industrialization, but explaining that will come another day.
FRAGMENTATION AND GETTING MORE OUT OF URBAN LAND
Besides all the cool things that cities are known for — like high culture, fine dining, cinemas or even cultural monuments. What makes cities engines of prosperity is that cities are essentially labour markets. People move to cities to seek economic opportunities for the betterment of their lives. Cities are good at fulfilling this goal because they match workers and firms efficiently. Firms cluster around cities because it has more people and gives them economies of scale. More people move to cities because they can easily get jobs at those firms and learn from proximity to other people. But this relationship becomes fragile or broken if a city has a fragmented form. Fragmentation occurs when people are too far from their work and opportunities. This causes a city to be disconnected and raises the cost of accessing opportunities.
Basically there are three ways to keep building a city. The first way is by expansion. People living “inside” the city of Abuja will be familiar with this. Expansion is using more land to add to the building stock of an existing built area. People living in an expanded area can still be connected, but as more building is done, people will be farther away from each other and away from business centers and economic hubs. The second way to build is to leapfrog. Sometimes new settlements that are completely cut-off from existing built areas spring up. This moves people even farther from each other than expansion and raises the cost of access and connecting those new areas with infrastructure like roads and connecting to the power grid. The third pattern of urban development is called infill. This is when new structures are added to an already built area with minimal additional land — an existing area is optimized for height rather than additional land space. Infill is the best way to build a dense city. And density reduces fragmentation, disconnection and lowers costs.
There are two central issues to consider when building a city. One is the role of urban planning and how much room to leave for spontaneous development. Rigid planning can stifle a city’s productivity and restrict the supply of housing — which inevitably raises costs and lock people out. Allowing too much spontaneity can lead to urban sprawl — that is uncontrolled expansion and leapfrog. The result is usually slum formation, congestion, and a crowded but not dense city. Conventional wisdom is that planning is efficient when ruthlessly applied to public spaces where streets, drainage channels, parks, etc. are properly built, demarcated and their use for exclusively private purposes is disallowed.
The second thing to consider when building a city is how to allocate urban land. Land is scarce — and scarcer still in cities. Allocating for its best use is one of the most under-appreciated features of a good city. But land allocation is a subtle art. It is often a case of not knowing exactly what to do, but knowing what not to do. No urban planner can know the best use of a piece of land or street. But using political power to allocate land to cronies and friends is one of the worst use of land. Jedwab observed in the cited paper that “state capture affects the composition and spatial distribution of cities” in Africa. Elites often use their political connection to acquire lands with good network effects and positive externalities. Entire neighborhoods can be populated by people from the same income class and make economic mobility costly for poor people. A good example is the case of a popular Nigerian bank that owned an entire block in the most economically vibrant neighborhood in Lagos.
There is also the issue of financing urban development. City governments in Africa often rely on transfers from the central government for financing. This can reinforce the effects of “consumption cities” as transfers from the central government can be treated as surplus by city government elites and spent on consuming services. There is often no pressure to monetize the allocation of scarce urban land, and there is less emphasis on the marketability of land so as to tax efficiently. A former governor of Lagos state is reportedly the “biggest landlord” in the city. If those claims are true, then the city is losing billions of dollars in taxes to such a graft. Economists (notably Paul Collier) have argued that it is ethically wrong for individuals to get disproportionate economic gains from simply owning a piece of land. This may not be true in all contexts, but the positive externalities and location advantages of certain lands are sound justifications for land taxes. A cafe or shopping mall located near a publicly paved access road in a popular business district will increase its sales because of that road and hence is required to pay taxes. The public finance benefits of land taxes are a good incentive to take land allocation seriously. Providing public goods like access roads, bridges and power grids are not cheap and can be an unprofitable investment. Allocating proximate land to valuable public goods for the most marketable and efficient use can recoup some of that public investment and provide decent leverage for future investments.
CHARTER CITIES
I noted in my referenced essay that cities in Africa have a coordination problem — this creates a negative path dependence or self-fulfilling vicious circle. When a city attains an inefficient physical and economic form, there is no positive future expectation by firms and investors. The lack of investment then worsen the physical state and the economic fate of the city, which then form future negative expectations — and the circle keeps turning. This is the “low development trap”. I also noted that it will take more than singular infrastructure spending or an isolated reform to break this trap. That is because cities are resilient creatures. A city’s physical and economic equilibrium can be persistent for centuries — long after the city is no longer competitive or functioning. Venice is a good example of such a city. A large-scale coordinated action is needed to fix a dysfunctional city. However, this can be a difficult task. There have been projects like city-wide slum upgrading, but it sometimes involves displacing people and exploiting the non-existence of land titles for some of those settlements to build condominiums that are not affordable for former residents.
Enter charter cities. The idea was first proposed by economist Paul Romer at a TED talk in 2008, but the underlying principle predates Romer’s exposition. Charter cities are quite similar to special economic zones, and the literature on economic geography has long recognized the divergent economic fates of different locations. One of the most successful recent examples of SEZs is the Chinese city of Shenzhen, which was just a fishing village four decades ago. Shenzhen is now one of the most prosperous cities on earth and hosts some of the largest companies in the world. I have been fascinated by charter cities for a while (I wrote about them here and here) and I think they can be an effective way of bringing prosperity to millions in the developing world. Charter cities can solve the coordination problem of dysfunctional urban dynamics in African cities. But implementation will need to confront some realities.
In Romer’s original formulation, charter cities are a way of experimenting with new rules and institutional forms over dysfunctional existing ones. Rather than do difficult “workouts” in inefficient cities, you can do a “startup” by building a new city with new rules. This was Deng Xiaoping’s brave experiment with places like Shenzhen. A communist regime allowed foreign capital and companies into a zone and allowed Chinese people to freely move there without wage and price controls. New rules can be a powerful means of changing old norms (see this talk by Romer). There are several minor experiments like this in developing countries. Developers build residential enclaves called “estates”, which provide public goods like power, roads and armed security for a service charge. Cities are the right scale for this experiment to have a widespread positive effect.
The problem with the initial trials of charter cities was to rest the idea on external crafting and enforcement of rules. This led to failed trials in Honduras and Madagascar — and it will be even more problematic in other parts of Africa because of colonial history. This top-down approach will have to give way to getting local elites, investors and other social actors excited about the idea of charter cities. Gladly, new people are trying to build an ecosystem around the idea of charter cities and deserve at least the curious attention of everyone who cares about human prosperity in places where it is still lacking.
NB — Mark Lutter of the Charter Cities Institute has an interesting essay on the subject.
