Skip to main content


Cities and regions

Economic prosperity is very unevenly distributed across space. What causes some places to be richer and some poorer? Why do these spatial disparities matter – and what can policy do to reduce them? Economists studying urban and spatial economics consider these questions across different countries, developed and developing, and at different spatial scales cities and regions.

At the turn of the century, the UK’s then prime minister Tony Blair promised that he would make sure every child had the opportunity to flourish “regardless of where they grow up”. Similar goals to rebalance, revive or “level up” the country – to reduce the gap between places that are rich and those that are poor – have been set out by British prime ministers ever since.

What makes tackling the economic differences between different places so hard? To answer this, we need to understand the economic forces that create these spatial disparities – that make some places richer and others poorer - what this implies for people living in different places and what, if anything, policy-makers can do. These questions have been a focus of CEP research for the last 30 years.

The crucial role of education

Spatial inequalities in the UK are profound and persistent. There is a clear North-South divide and some of our biggest cities, and too many of our towns and coastal areas, struggle to provide good jobs and wages for the people living there.

What is driving this difference in the distribution of jobs and wages? CEP research by Stephen Gibbons, Henry Overman and Panu Pelkonen suggests that education and skills play a central role in explaining these disparities.

The share of the population with degrees varies significantly across the country. When comparing average wages between different areas, it is such variations in the education and skills level of the local population that account for most of the differences in area wages – up to 90 per cent. The wage gap between people with similar skills in different areas is much smaller.

Highly educated workers have better labour market outcomes on average, wherever they are. But places with a greater number of good jobs attract a greater number of highly educated workers – and this makes places even more productive. The circular nature of this process – concentrations of highly educated workers making areas more productive, which then attracts additional highly educated workers – can mean that the economy turns small “natural advantages” – for example, a favourable climate or geographical position, such as a natural port – into big spatial disparities.

In disadvantaged places, this feedback loop works in the opposite direction. Places with fewer opportunities and jobs are less attractive to graduates and with fewer graduates, these places then become less productive and hence less attractive to employers.

The implication is that to “level up” the UK, policy-makers need to understand what causes these differences in area productivity and how this relates to the uneven distribution of education and skills across the country.

Agglomeration: why being together matters

One big reason for differences in productivity between areas is “agglomeration benefits”, the economic benefits that flow from labour pooling, input sharing and knowledge spillovers when firms and workers gather in big cities. These agglomeration benefits have been a key focus of much of CEP’s research in urban economics. Early studies covered a variety of questions about the channels through which spatial concentration – having lots of firms and people in a particular area – could enhance productivity, including the role of a broad industrial base in increasing innovation.

Photo: Pedestrian commuters walking across London Bridge on iStockPhoto Licenced and Downloaded 04/08/2021
Photo: Pedestrian commuters walking across London Bridge. Credit:

Agglomeration benefits are the fundamental reason for cities to exist, but measuring their strength is difficult due to the "chicken and egg" problem described above: if higher density places are more productive, this attracts more firms and workers. How do you separate out cause and effect?

Ground-breaking work by Gabriel Ahlfeldt, Stephen Redding, Daniel Sturm and Nikolaus Wolf used Berlin’s partition in 1961 and reunification in 1989 as a ”natural experiment”. Specifically, they exploited the fact that the rise and fall of the Berlin Wall changed local densities because it cut some city blocks off from their neighbours. Using detailed data on different city blocks that went all the way back to the 1930s, the researchers studied how the city responded to these changes in density and provided new estimates of the strength of agglomeration externalities.

From theory to data – and back again

The Berlin wall research represented a stepping-stone between CEP’s earlier theoretical work on uneven development - what became known as “New Economic Geography” (NEG) - and newer research, which developed quantitative models that aimed to explain the spatial distribution of economic activity.

Starting with the Nobel Prize-winning work of Paul Krugman in 1991, NEG brought together insights from the field of international trade and economic geography to understand how large spatial disparities might emerge from small initial differences between places. This model put economic interactions between firms and their consumers – and the importance of having market access to both – at the core of understanding the spatial distribution of economic activity. CEP’s Tony Venables worked with Paul Krugman to develop the theoretical insights emerging from NEG. 

By the 2000s, interest had turned to testing the theoretical predictions of these models. Redding and Sturm once again turned to the division of Germany after the Second World War and the reunification of East and West Germany in 1990 as a natural experiment that could provide evidence for the importance of market access for economic development.

Following division, West German cities close to the new border with East Germany went from being at the centre of an integrated Germany to being on the periphery of West Germany. Redding and Sturm showed that the imposition of the East-West border led to a sharp decline in population growth in West German cities close to the new border relative to other West German cities, and that this decline was more pronounced for small cities than for large cities. All these findings were in line with theoretical predictions.

Advances in both the underpinning theoretical research and increases in computing power have allowed researchers to move away from testing theories developed using simplified models of cities and regions, to using complex models that can explain the data from large numbers of locations with different geographies, productivity, amenities and local factors, as well as different trade and commuting costs between locations.

Using such a model, and combining data on commuting, population, value of floor space and more for small areas within London from 1801 to 1921, Stephan Heblich, Redding and Sturm showed how the invention of the steam railway, which dramatically cut travel times, led to the first large-scale separation between where people worked and lived – creating commuters. And they used the model to ask what would have happened if London had never built its railway network.

Such a big change to London’s infrastructure would have had striking effects – reducing both the population and the value of land and buildings in London in 1921 by a little over 50 per cent, and decreasing net commuting into the historical centre of London by more than 300,000 workers. Such thought experiments, using real world data, help to illustrate the importance of transport networks in underpinning the economies of our largest cities.

Why are UK house prices so high? 

Sophisticated quantitative modelling is not the only way to understand the causes and consequences of spatial disparities. One important strand of CEP research has focused on housing and the cost of living in cities, and how high house prices lead to high costs of living that can offset the benefits of agglomeration and constrain city growth.

The central housing policy issue in the UK is what many call the “affordability crisis” – the fact that housing has become ever more unaffordable, especially for younger households trying to get on the housing ladder. House prices in the UK – particularly in London and South East England – are among the highest in the world. UK housing costs are not only very high in absolute terms and relative to other countries but also relative to income.  The UK’s affordability crisis has its epicentre in London, but it extends over most of England and all of the South East. It has been developing slowly since the 1960s and accelerated in the early 2000s.

Photo: For Sale sign displayed outside a terraced house in Crouch End, London on iStockPhoto Licenced and Downloaded 04/08/2021
Photo: House for sale, Crouch End. Credit:

What is causing the housing affordability crisis? CEP research points clearly to the main cause being the UK’s planning system – in conjunction with strong demand for housing in some regions, notably the South East. In England, the evidence suggests that local planning constraints magnify the impact of growing housing demand on house prices across the country.

Christian Hilber and Wouter Vermeulen, who analysed data on house prices and earnings over more than three decades, found that house prices in places in England with tight regulatory constraints responded much more strongly to changes in local earnings than in places where constraints were less tight. They estimate that house prices would have risen by about half as much as they did in real terms between 1974 and 2008 if, hypothetically, all regulatory constraints were removed.

How house prices reveal what we value

Alongside productivity differences and the cost of housing, a wide range of amenities determines the attractiveness of different urban locations. And when the supply of housing is constrained, as it is in much of the UK, we can figure out how much people value these different amenities by looking at their effects on house prices.

In a series of widely quoted studies, CEP researchers, have used this insight to work out how much people value everything – from analysis of high quality schools (which shows that a 1 percentage point increase in the proportion of children reaching target grades pushes up house prices by 0.67 per cent) to access to canals and proximity to wind farms.

Urbanisation and development

CEP researchers also look beyond the UK and Europe to consider the role that cities play across the world and how this has changed over time. Vernon Henderson and co-authors have explored the role of natural characteristics – such as ruggedness, temperature, precipitation, access to navigable water and proneness to malaria – in determining how economic activity is distributed across the world.

Among countries that developed early – such as the UK and Western Europe – differences in things that make agriculture productive were much more important for explaining the spatial distribution of economic activity than differences in things that make trade easier, even though the group of countries that developed later is far more dependent on agriculture.

For early developers, structural transformation – the rise of cities – due to rising agricultural productivity began when transport costs were still high, so cities were localised in agricultural regions. When transport costs fell, these cities persisted. In late-developing countries, transport costs fell before structural transformation began, which meant that cities could develop away from agricultural regions in places with good transport links, even though agriculture remains more important for those countries than the early developers.

This work used satellite data on night lights to measure economic growth from outer space – a technique that has been widely used following Henderson and co-authors’ research showing how such data can provide a useful proxy for income growth, especially in countries with weaker statistical infrastructure and where there is a larger share of informal economic activity.

Photo: Manhattan, New York City, United States downloaded 04/08/2021
Photo: Satellite image of Manhattan by NASA on Unsplash.

Other research zooms in to consider what happens in particular cities in developing countries. Africa’s demand for urban housing is soaring and so it is important to find viable ways to meet this demand and reduce the poor living conditions arising from a proliferation of slums.

For example, in Planning Ahead for Better Neighborhoods: Long-Run Evidence from Tanzania, researchers Guy Michaels, Dzhamilya Nigmatulina, Ferdinand Rauch, Tanner Regan, Neeraj Baruah and Amanda Dahlstrand found that providing basic infrastructure, such as roads, water mains and plots for housing – but leaving residents to build their own houses – resulted in higher-quality housing than in areas where housing was allowed to develop organically without much planning or infrastructure.

Where are we now?

CEP research on the causes of spatial disparities has informed key policy debates around reforming the planning system, the way we appraise and evaluate transport investment and how to help improve the economic performance of cities.

As the UK economy adjusts to the immediate impacts of Brexit and Covid, as well as to climate change and the transition to net zero, new research will be needed to understand the implications for spatial disparities. What determines the short- and long-term ability of cities to respond to such profound economic shocks?

While the shocks might be new, the fundamental questions will remain the same. To what extent do different concentrations of high and low skills in an area explain differential city growth? What is the role of population density, roads, railways and various agglomeration benefits? And what role do better amenities play in attracting people to move to successful cities?

Similar questions arise for systems of cities across the world, not just in the UK. But the fact that the fundamental questions remain the same doesn’t mean we are back to where we started. We have learned much from the theoretical and empirical advances of the last 30 years, which puts us in a much better place to continue answering these fundamental questions in an ever-changing world.

By Henry Overman