In 2009, no political or economic initiative, no conference, research programme or media report dealing with the urbanization of the planet seemed able to avoid a brief substantiating sentence that captures the imagination: "Today, the majority of people in the world live in cities". Some claim that this has been the case "since 2007". Others are more prudent and recognize that we are in fact uncertain as to the exact date and figure of 50%. Yet there seems to be a broad consensus on the matter as a postscript is invariably added pointing out that the proportion of the planet's urban population will inevitably increase-which is also no more than a hypothesis.
The oft-cited rate of world urbanization is singularly lacking in substance. In fact, no expert is in a position to solve the issue of the heterogeneity of data produced by national statistics institutes. Each country uses its own definition of what is urban, and many countries even propose several conflicting definitions and thus several different figures.
The definition makes the city
The example of Paris illustrates this problem perfectly. According to the French National Institute of Statistics, Insee, and the results of the 1999 population census, Paris counts either 2.1 million, 9.7 million or 11.2 million inhabitants depending on whether one chooses to consider the city's administrative unit per se (105 km2), the urban unit covering the built-up agglomeration around the inner city (2,723 km2), or the urban zone that includes the area in which people commute on a daily basis (14,518 km2). Depending on the definition, the results obtained are very different. All the resulting indicators and discourses that rely on these figures are thus affected: population density, number of jobs, land and property prices... For nearly half a century, the population of the city proper has been decreasing, while that of the urban agglomeration has been constantly on the rise. The boundaries of the city of Paris have not changed since 1860, whereas those of the agglomeration and of the urban zone have continued to expand.
In three-quarters of the countries, the notion of agglomeration (urban unit) does not exist officially. And even when this definition is identical in two countries, the minimum population threshold defining the "urban" category varies considerably: 10,000 inhabitants in Greece, 2,000 in France, 200 inhabitants in Sweden... Nor is the criterion of the administrative city unit viable for making international comparisons. Applied to Paris, it only describes one fifth of the Parisian agglomeration, whereas Chongqing becomes the largest city in the world, with more than 30 million inhabitants in an area of 82,000 km2-which in size is the equivalent of Austria. The central agglomeration of Chongqinq brings together fewer than 4 million inhabitants, the remainder being scattered over 1,300 small towns and villages mostly inhabited by rural people. One could just as well decide that Austria is a city, with the result that 100% of its population would suddenly become "urban" in international statistics.
In order to establish comparisons and trends at the international level, would it be better to work on the basis of the urban zone defined by commuter flows? Not really, because the kind of statistical information needed for such calculations only exists in a few developed countries. Although the statistical services of the United Nations provide a recommended definition, only a few countries take this into account. Individual States have total authority as to how they formulate their own statistical definitions, which of course is a reminder that the origin of this word is the Italian statistico, meaning "what concerns the State".