Blog
Spatial Inequality and Development

by Anthony L. Venables

Economic activity is distributed extremely unevenly across space. At the international level there are rich countries and poor ones-underdevelopment can be viewed as a manifestation of spatial inequality. Within countries there are regional disparities; per capita income in the Southeast region of Brazil is three times higher than it is in the Northeast. Urbanization provides a powerful illustration of the extent to which population, manufacturing and services cluster together. Even within cities there are examples of particular sectors concentrating in small neighbourhoods.

There is some evidence that spatial inequalities within countries increase during the early stages of development and during periods of rapid economic change. In China, Russia, India, Mexico and South Africa evidence suggests that spatial and regional inequality of economic activity, incomes and social indicators, is on the increase. Spatial inequality is a dimension of overall inter-personal inequality, but it has added significance when spatial and regional divisions align with political and ethnic tensions to undermine social and political stability. Also important in the policy debate is a perceived sense that increasing internal spatial inequality may be related to greater openness of economies and to globalization in general. The WIDER project on Spatial Inequality and Development has brought together economists and geographers studying a wide range of countries to investigate these issues.

Three broad questions need to be addressed. The first is, how important is spatial inequality, and how is it evolving? Most economic variables are unevenly distributed across space-population is concentrated in particular regions, industrial sectors clustered in particular towns. The ultimate concern is, however, with spatial inequalities in per capita income. As already noted in the case of Brazil, these can be very large, but research suggests that within country spatial inequality accounts for at most one-third of the total inequality in personal incomes. This one- third is however particularly important, as it not due to underlying differences in individual characteristics-such as ability- but simply a consequence of where people live. Individuals may have allegiance to spatial units, perhaps because these units are aligned with language, ethnicity or religion. In this case increasing spatial disparities may lead to tensions and conflict.

How has spatial inequality been evolving over the past two decades? There is evidence that, within many countries, it has been on the increase. In Mexico and China where trade liberalization has been associated with overall growth the benefits of this growth have not flowed evenly across space. The same has been true in many transition countries. However, a number of key questions remain. To what extent is some increase in spatial inequality a natural feature of development, as growth is initially concentrated in a few regions? Is this increase temporary, and how long is it likely to take for growth to spread from region to region?

The second broad question is, what are the determinants of spatial inequalities? If the world was a 'featureless plane', and if economic activity had the standard neoclassical properties, then economic activity would be evenly distributed across space and there would be no spatial dimension to inequality. But the world does not satisfy either of these two assumptions. There are real geographical features such as mountains and coasts and forests and rivers that can affect the distribution of economic activity and spatial inequality in wellbeing. More importantly, activity has a propensity to cluster together. This occurs as firms benefit from forwards and backwards linkages with proximate supplier and customer firms; as firms and workers benefit from the development of large pools of skilled labour; and as firms learn from observing the behaviour of close by competitors. Given these forces, it is to be expected that activity should cluster together.

Does this matter, and what are the implications for policy? If economic activity tends to cluster then it suggests that development is unlikely to take the form of smooth convergence in the economic performance of regions or countries. Some places will boom, while others will lag behind. Development becomes an inherently 'lumpy' process, as growth is spatially concentrated.

Whether or not spatially unequal development also creates income inequalities depends to a large extent on the extent to which labour can move from lagging regions to fast growing ones. This suggests that policies to facilitate migration should tend to eliminate spatial inequalities, as workers move to high wage areas. However, migration is not always easy, as investment in location-specific physical and human capital can mean that individuals get trapped in a declining region.

Furthermore, it is not always clear that promoting migration is an appropriate response. There are typically multiple market failures associated with the location decisions of firms and individuals. Some of the positive ones were noted above - setting up a new firm in a location might have positive spillover effects for local workers and neighbouring firms. But there are also negative externalities. Bringing more activity into a large city might create congestion costs, and could also damage the source regions, from which the activity has moved. This is why investing directly in infrastructure in regions, to allow individuals to increase the return to their general and location specific human capital, is a policy option that must always be kept on the table.

The specifics of policy recommendations still need to be developed. They need to be based on understanding and quantifying the importance of all the microeconomic factors-to do with endowments, institutions, and also spillovers and linkages between economic agents-that make some locations more attractive destinations for investment than others.

Tony L. Venables, from the London School of Economics, is the co-director (with Ravi Kanbur from Cornell University and Guanghua Wan from WIDER) of the WIDER project on Regional Disparities in Human Development