VoxEU Column Development

Making city lights burn brighter

Urbanisation and GDP per capita are positively correlated across countries. However, when the sample is restricted to developing countries, urbanisation and growth are more loosely related – particularly in Africa. This column argues that the low share of manufacturing in developing-country cities may help to explain this discrepancy. Strengthening urban finances, embracing technology, improving skills, and stimulating the formal sector will help cities to promote growth. Since decisions affecting urban development can have lasting impact, longer-term planning deserves greater attention than it is currently receiving.

Urbanisation and per capita GDP are well correlated.1 According to a recent estimate by Gilles Duranton using cross-country data for 2012 (see Figure 1), each percentage point of urbanisation is associated with a five-percentage-point increase in GDP per capita, with urbanisation apparently explaining 60% of the variation in incomes.

Figure 1. Urbanisation and GDP per capita

But what causes what? Urbanisation could be a major outcome of the development process, or alternatively, rapid urbanisation could become a driver – or a facilitator – of growth.

  • Data from developed countries suggests that urbanisation has been growth-promoting;
  • Data for Africa, Latin America, and south Asia alone suggests that on the contrary, urbanisation is only loosely coupled with growth.

Developing countries are urbanising at rates of 2–3% per annum, but per capita incomes are increasing more modestly than in the past.

  • The growth-inducing effects of urbanisation are even weaker if correlations are limited to African countries.
Urbanisation 2.0: In the developing world

A striking difference between the urbanising experiences of developed and developing economies lies in manufacturing. Typically, manufacturing might account for less than 15% of the GDP in medium and large cities in developing nations, and can be as little as 5% in many African cities (e.g. Lagos). It was often one-third or more of GDP in Western countries when their urbanisation was in full swing.

Without the productivity dividend arising from the spread of manufacturing in cities, urbanisation has been paralleled with lower GDP growth and fewer formal-sector jobs, with an increasing percentage of the expanding urban workforce engaged in services occupations.

Continuing rapid urbanisation juxtaposed with a slow increase in the numbers of ‘good jobs’ – whether in manufacturing or services – not only strains the absorptive capacity of cities, but also leads to negative externalities that can circumscribe the productivity-enhancing agglomeration economies associated with city size, especially in the earlier stages of development (Brülhart and Sbergami 2008). These derive from economies of scale, scope, diversity, clustering, and specialisation, and through the thickening of labour markets.2 But these economies do not accrue automatically as a city expands – they are a function of the macroeconomic environment, and depend upon certain acquired urban attributes that can be undermined by poor policy choices and a lack of planning.

Keys to productive urbanisation

Urbanisation needs to go hand-in-hand with four sets of reform measures:

  • Strengthening urban finances so that cities can address their deficits in housing, infrastructure, and services;
  • Embracing technologies that improve the design and performance of essential urban hardware and services;
  • Increasing the supply of skills and improving quality; and
  • Stimulating private investment in tradable activities that will create a multitude of jobs in the formal sector, thereby gradually narrowing the scope of the informal economy.

The development orientation of the state and a conducive macroeconomic regime are necessary preludes to city-level efforts, starting with resource mobilisation by local authorities to fund infrastructure and services. The good news is that macroeconomic management has improved in most regions of the world. The bad news is that relations between national governments and city administrations are often strained, and urban financing remains a challenge. Where possible, coordinated regional solutions to problems of transportation and environmental management offer high returns for sustainable, urban-led economic growth.

There are new digital technologies and techniques for analysing ‘big data’ that can improve and organise urban life, and bolster agglomeration economies. These advances can make urbanisation a more efficient process. New technologies can make traffic more manageable and reduce congestion. Urban design can become more flexible, conserving of resources, and performance-based with the help of the huge data streams being generated through myriad sensors. Moreover, better cooperative arrangements between governments and the private sector open up new avenues for urban planning and management.

Of course challenges remain, especially in the areas of housing and human capital development. Turning the tide against the spread of shantytowns and the many-sided environmental degradation that follows in their wake calls for a multi-pronged strategy, aimed at the upgrading of selected existing slum developments – as done in parts of Latin America – together with increased investment in affordable housing – as demonstrated by a few east Asian countries. Because manufacturing is rarely the engine of growth in newly forming cities, migrants must find jobs mainly in the services sector. The better-paid jobs in tradable services go to those with appropriate education and skills. Hence successful urbanisation is now associated with the acquisition of skills, including ‘soft skills’ not needed in traditional manufacturing jobs.

Each of these ‘facilitators/necessary conditions’ only provides traction for urban economies if existing businesses invest and expand production, and new entry thickens the ecosystem of firms and contributes to technological upgrading, diversification, and innovation. The skilling of cities needs to be supplemented by measures that remove some of the roadblocks deterring entrepreneurship. Research also points to the concentration of human capital created by good schools and training entities connected to urban productivity, entrepreneurship, and innovation – especially in high-tech activities.3 Creative cities exhibit numerous positive externalities, especially when they harness ICT. Hence, near the top of the urban strategy is the nurturing and retention of human capital through investment up to the university level and in research institutes; and investment in amenities that can make a city attractive.

Concluding observations

Cities may not drive growth, but growth will be seriously constrained if cities are unable to meet demand for infrastructure and services; fail to create an attractive and secure business environment; and are unable to develop thick labour markets supplying businesses with the desired mix of skills. Their efforts will also fail if they cannot deal with problems posed by slums, congestion, and pollution; and if they are slow at harnessing technologies that feed productivity and are critical to global connectivity.

City-led growth can be an effective component of national development strategies if investment decisions take into account where future populations will be living, and plan for the delivery of basic services and infrastructures tailored to the likely demand from productive activities. Because decisions affecting urban development can have lasting impact, longer-term planning deserves greater attention than it is currently receiving. Given the speed of innovation and technical change, avoiding poor decisions while strengthening institutions is the right path to leveraging urban capabilities to accelerate growth.


Baldwin, Richard E and Phillipe Martin (2004), “Agglomeration and Regional Growth”, in J V Henderson and J-F Thisse, eds., Handbook of Regional and Urban Economics, 4: 2671–2711.

Behrens, Kristian, Gilles Duranton, and Frédéric Robert-Nicoud (2013), “Productive cities: Sorting, selection, and agglomeration”, University of Geneva Department of Economics Research Paper 13111. 

Brülhart, Marius (2009), “Is the New Economic Geography Passé?”, VoxEU.org, 7 January. 

Brülhart, Marius and Federica Sbergami (2009), “Agglomeration and Growth: Cross Country Evidence”, Journal of Urban Economics, 65(1). 

Duranton, Gilles and Diego Puga (2004), “Micro-Foundations of Urban Agglomeration Economies”, in J V Henderson and J-F Thisse, eds., Handbook of Regional and Urban Economics, 4: 2063–2117. 

Economist (2013), “Urbanisation and growth: City chickens and country eggs”, Analects blog, 4 August. 

Glaeser, Edward L (2011), “Unleash the Entrepreneurs”, City Journal, 21(4).

Glaeser, Edward L and Joshua Gottlieb (2009), “The Wealth of Cities: Agglomeration Economies and Service Equilibriums in the United States”, NBER Working Paper 14806. 

Hofman, Anett and Guanghua Wan (2013), “Determinants of Urbanization”, Asian Development Bank Economics Working Paper 355. 

Leipziger, D and S Yusuf (2014), "Making city lights shine brighter", CEPR Policy Insight 71, 3 March.

United Nations Human Settlements Programme (2010), “Urban Trends: Urbanization and Economic Growth”, March.

1 Henderson (2002, 2010), United Nations Human Settlements Programme (2010). Hofman and Wan (2013) find that the causation runs from growth to urbanisation. See also Economist (2013).

2 Duranton and Puga (2004), Glaeser and Gottleib (2009), Brülhart (2009), Baldwin and Martin (2004).

3 Entrepreneurship is the critical ingredient, and it is not necessarily correlated with human capital (Glaeser 2011). Larger cities could have an edge according to Behrens et al. (2013).