The world’s new growth frontier: Midsize cities in emerging markets

This interactive presents one scenario for the urban world’s evolution. Like any scenario, it is subject to large bands of uncertainty about population trends, migration, business innovation, per capita GDP, the evolution of city structure and management, and the outlook for exchange rates. Like these uncertainties, methodological issues influence urban economic forecasts. Below are details on how we approached two key methodological issues.

Defining cities. Where possible, the cities in our database refer to integrated metropolitan areas rather than specific city jurisdictions; we aggregate neighboring cities into a single urban center. To do so, we have relied on the Functional Urban Area definition from Eurostat’s ESPON project, as well as the metropolitan statistical area definition of the US Bureau of Economic Analysis.

Our approach often results in a relatively broad definition of “city,” because in many instances the city center and the legal city make up only a fraction of a total integrated urban region, in both population and area. Examples of such aggregations include Rhein–Ruhr in Germany; Los Angeles, Long Beach, and Santa Ana in California; and Mumbai and Thane in India. In some cases, functional urban regions cut across national boundaries (for instance, Geneva in Switzerland and France, as well as Copenhagen–Malmo in Denmark and Sweden). However, to be consistent across regions, we also strive to separate urban entities that are located closely together but have relatively little cross-city integration measured by commute flows (for instance, Seoul, Incheon, and Suwon in South Korea, as well as Beijing and Tianjin in China).

Measuring GDP. The MGI Cityscope database encompasses several GDP-measurement approaches because different ones are better suited to some purposes than to others. Per capita GDP figures, for example, are expressed in 2007 purchasing-power-parity (PPP) exchange rates, which are useful for shedding light on differences in living standards. On the other hand, in this interactive we express overall urban GDP data in US dollars measured at the real exchange rate (RER) because it most closely approximates the expected dollar value of revenues or income earned in different currencies. The RER for 2007 is the market exchange rate. We predict the 2025 RER from differences in per capita GDP growth rates, in order to approximate the combined effect of changes in local prices and market exchange rates that affect the dollar value of GDP or income in each country.¹ 1. The logic underlying this approach is that the faster per capita GDP grows, the more rapidly the relative cost of nontradable goods and services is likely to increase, leading to a higher dollar value of local sales. We account for this factor by assuming relative RER appreciation proportional to the gap in per capita GDP growth relative to the growth of the United States, adjusted for the share of nontradables in the economy.

For further details, see the full report, Urban world: Mapping the economic power of cities.