An overview
André de Palma and Alexandre Guimard
1.1 Foreword
Historically, the concentration of people has always shifted. Individuals concentrated at some specific places for short periods of time and moved their habitats from one place to another. Several reasons may lead to âephemeral citiesâ: events and effects of seasonality, or nomadism. First, the âephemeral cityâ can be a place where a group of people meet from time to time. For example, individuals can periodically meet at some particular places (markets) to exchange goods, but this does not necessarily lead to the emergence of cities. The reason for temporary concentrations could also be religious. As an example, people met in Mayan sites (with no permanent residents) for ceremonial settings (Kostof, 1993). According to Mumford (1961), when people were still nomads, sepultures were a point of concentration. Second, communities moved their habitat over time for reasons related to climate, natural resources (soil erosion, water resources, etc.) and livelihood. Before the Neolithic Age, for example, individuals hunted and fished to feed themselves and frequently moved their habitat.
As a matter of fact, the concentration of a large population size is not the prerogative of man. Indeed, a number of animals are referred to as social because they live in a community. These communities may include thousands of insects, as in ant colonies, or even several millions of insects, as in communities of termites. These systems are extremely well regulated. For example, workers in the nests of termites manage to keep the temperature of the nest close to the ideal temperature (Deneubourg and Franks, 1995). These regulations are very different from those implemented in many other animal societies. These phenomena of self-organization involve large-scale systems, which could provide fruitful sources of inspiration for researchers in social sciences.
At around 6000 BC, only a few thousands of people lived in cities, which is not many compared to the millions of people who are currently living in metropolitan cities such as Tokyo, Delhi, Mexico or New York City (NYC). The way cities emerge and grow depends on the historical context. We can roughly distinguish two types of cities, the pre-industrial cities and the post-industrial ones (Sjoberg, 1960): âPreindustrial cities depend for the production of goods and services upon animate (human or animal) sources of energy â applied either directly or indirectly through such mechanical devices as hammers, pulleys, and wheels.â By contrast, post-industrial cities depend on âinanimate sources of power [âŠ] used to multiply human effortâ and allowed by the invention of steam machinery. After the Industrial Revolution, production processes and modes of transportation stayed roughly the same over time. The city was dense (Newman and Kenworthy, 1996), with mixed land use, and âcity dwellings often serve as workshopsâ (Sjoberg, 1960). After the Industrial Revolution, plants and new modes of transportation changed cities. The city became more spatially specialized and spread out. Urban sprawl increased even more after World War II and found its apogee with the automobile city during the 1970s, due to the democratization of the private car.
Cities are defined quite differently across countries and over time, and this makes comparisons very difficult: âBecause of national differences in the characteristics which distinguish urban from rural areas, the distinction between urban and rural population is not amenable to a single definition applicable to all countriesâ (Chen et al., 2013). In China, for example, urban administrative areas, âoften include large stretches of farmland and sizeable rural population, thus inflating the urban population figuresâ (Chan et al., 2008). The hukou system, which âacted as an internal passport arrangement regulating mobility and granting people citizenship in the locality,â makes the task of counting urban population in China even more difficult since a lot of migrants are not treated as permanent migrants in census data. However, it is necessary to have a common definition of what is a city in order to compare them around the world.
According to Mumford (1938), the city is âa point of maximum concentration for the power and culture of a community.â Marshall (2007) recalls the definition of urban areas by the US Census Bureau:
The US Census delineates âurban areasâ as areas based on the population density of a census block or block group being greater than 1000 people per square mile and the population density of surrounding census blocks being greater than 500 people per square mile. Urban areas (UAs) must have a population size of 50 000 or greater. UAs are delineated at the start of each decade.
(US Census, 2004)
In France, the continuity of settlements and the concentration of individuals are the key statistics of cities. The definition of cities changed with the introduction of new concepts over time, such as urban areas. In France, according to INSEE, âan urban unit is a municipality or a set of municipalities presenting a zone of built up continuous (no cut of more than 200 meters between two constructions) ⊠and with at least 2,000 inhabitants.â Since 1997, INSEE introduced the new concept of urban areas by allowing remote settlements to be part of a larger urban core. However, the core must have a large labour market (10,000 employees for large urban areas). Thus, many remote settlements are excluded because they are not included in an urban core that is big enough in terms of total employees. This biases the analysis of cities over time and may lead researchers to think that rural areas resist the urbanization process, whereas actually, according to LĂ©vy (2013), urbanization in France has been achieved. The proportion of urban areas is at its maximum level since even the more remote âruralâ areas are connected to urban cores (peri-urbanization).
Worldwide, just a few years ago (by 2007), there were as many people living in cities as in rural areas. Urbanization is increasing in such a manner that, in 2050, 70 percent of the total population will live in cities. In less developed countries, the growth of cities is substantial. In developing countries, between 2010 and 2015, 183,000 individuals will move to cities each day, leading to 91 percent of urban growth worldwide (UN Habitat, 2013). According to the United Nations (2012), in less developed countries, urban populations will grow from 2.7 billion in 2011 to 5.1 billion in 2050. As a result, control of urbanization is often difficult to achieve.
This large gain is due to population growth and to rural exodus. From the date of writing to 2050, the total population worldwide will increase by 30 percent and the urban population by about 70 percent, implying a large increase in the urban proportion. Total population figures worldwide will grow from 7.130 million residents in 2013, to 9.306 million residents in 2050 (INED).1 The urban population will grow from 3.6 billion in 2009 to 6.4 billion in 2050 (UN, 2012). Between 2011 and 2030, both in the developed and the less developed countries, the urban population will increase more than the total population (0.52 percent vs. 0.23 percent and 2.02 percent vs. 1.07 percent respectively), but this trend will be evident in less developed countries. In developed countries, the increase of the urban proportion will be moderate: it will increase by 100 million residents between 2011 and 2050, whereas the total population will increase by only 70 million. In China, from 2035, the total population will decrease (by â0.14 percent between 2035 and 2040) while the urban proportion will keep growing (from 47 percent in 2010 to 61.9 percent in 2030) because the rural exodus will remain in place.
Fast urban growth will mainly occur in medium and large cities. Mega-cities, defined as cities exceeding ten million inhabitants, are responsible for only 9.9 percent of total urban growth in 2011 and will increase by only 13.6 percent in 2025. Between 2000 and 2010, the number of cities above ten million inhabitants increased by 35 percent; it grew from 17 mega-cities in 2000 to 23 in 2010.
Urban growth is not homogeneous among regions. For example, the population of Cleveland decreased by 27.4 percent in only 12 years. The number of inhabitants decreased from 501,662 in 1999 to 393,806 in 2011 because of the industrial crisis. Urban growth is thus a function of historical time periods, but may differ across countries and cities. Cities do not grow in the same way, and there exist distinct urban growth patterns between cities around the world. Some cities grow faster than others and some are spreading out more than others. Large investments in road infrastructures and a high car detention can explain these disparities.
Despite the uniqueness of each situation, several regularities have been documented concerning urban development. A scaling law informs by how much in percentage terms citiesâ characteristics increase or decrease, with mostly a 1 percent increase of city size being noted. There exists a scaling law between energy consumption and city size, a scaling law between levels of crime and city size, a scaling law between revenues and city size, and a scaling law between the city size and urban sprawl. The literature indicates by how much, on average, urban sprawl increases in percentage terms for a one percent increase of city size (Fuller and Gaston, 2009). At a lower-scale level we also find some regularities, such as the constant share of travel expenditures in total revenues (Schafer, 2000). In addition, it is not surprising to find regularities at the scale level of cities, in particular to find a rank-size rule, i.e. a law that relates the population size of a city to its rank in the countryâs or worldâs ranking of cities by population.
The Zipfâs law (rank-size rule) postulates that city growth rate is inde...