National data are based on the UN's "National Accounts Main Aggregates Database" for historic numbers and the IMF's "World Economic Outlook Database" for future numbers. The International Comparison Program's 2005 revision is used for PPP data. We use these sources to anchor data at the country level and then normalize the detailed subdivision and city data so that they add up to the national anchor points.

Subdivision data are collected from each national statistics agency or Eurostat. Subdivision data are typically not available for every year. Also, some countries only measure GDP at the subdivision level while others only measure household income or spending. We use econometric techniques to estimate missing data so that C-GIDD is fully populated.

City data is collected from a variety of sources. We use the UN's definition of urban agglomerations to define cities. A city is defined by "the de facto population contained within the contours of a contiguous territory inhabited at urban density levels without regard to administrative boundaries. It usually incorporates the population in a city or town plus that in the sub-urban areas lying outside of but being adjacent to the city boundaries." This definition is the same as the Metropolitan Statistical Area (MSA) used in the United States and the Larger Urban Zone used in Europe. City GDP and household income is typically not available for cities, but can in 80% of cities be estimated from sub-subdivision data. For the remaining 20% of cities, we make the estimates based on econometric analysis of subdivision data. Less

Until the late 1994s, it was practically impossible to quantify how many households there were at a certain income level in most countries. Today, it is possible at the national level, but typically not at subdivision or city levels. C-GIDD is the only data source in the world for such data.

The benefit of having access to these statistics is that you can learn how many people can afford a particular product or service. For example, cellphones are usually bought by households making more than PPP $5,000 annually. With C-GIDD, you can easily quantify how many such households there are in Rio de Janeiro, Manila or Dhaka, to name a few large cities. The alternative way of looking at consumer affluence--average income per household--says nothing about how large the market is, only that Rio is more affluent than Manila and Dhaka.

The income distributions have also been used to estimate socioeconomic levels in each geographic unit. Based on a Mexican definition of socioeconomic levels, the database contains the number of people and households belonging to the upper (AB), upper middle (C+), middle (C), lower middle (D+), lower (D), and marginalized (E) classes, respectively . This analysis is done on an adjusted household-size basis to take into account that large households reap economies of scale and children tend to consume less than adults.

The benefit of using this globally consistent definition of socioeconomic levels is that you can create one language system for discussing consumer markets between and within countries. Branded consumer goods and modern trade typically cater to the upper class and upper middle class markets and often reaches into to the middle class. Thus, we--and many other experts--recommend that companies discuss market opportunities in terms of socioeconomic classes when comparing countries, subdivisions or cities. Less