Figure 3 The evolution of population growth rates according to the components of ‘aires urbaines’ 1968-99 Jobs Spread Out Less Rapidly than Residences Job locations still remain highly concentrated in the central sections of the agglomerations (Lainé, 2000): in 1999, more than 41% of jobs were located in the central communes, and 30% in suburban communes. Admittedly, the number of jobs located in the central communes dropped by 1.3% per year between 1990 and 1999, but it continued to rise in the suburbs much more significantly than in the peri-urban ring (1.3% as opposed to 0.4%). Today fewer than 10% of jobs are established in the peri-urban rings, while the number of jobs in the rural communes (16% of the total) continues to fall (those of the communes multipolarisées remain proportionately stable, that is, at 3%).
Admittedly, in the periphery of the largest cities, centers for secondary jobs have emerged, either by the absorption of preexisting urban centers, or by the implantation of new activities near transportation infrastructures (airports, motorway intersections), or in new zones of urbanization (La Défense, west of Paris, the new towns, or again the technopolitan zones in most of the regional metropolises). The growth of these new job zones has taken place very rapidly, without however posing a threat to the preeminence of the principal urban centers. The profile of their activities is often more specialized (for example in logistical activities, or in large commercial complexes) and less diversified than the profile of the city centers (Guérois, Le Goix, 2000). It can therefore be seen that the spatial structures of cities have become more complex, involving in particular new patterns of movement, from outskirts to outskirts, but that, up until now, they have not obliterated organizations of the center-periphery type.
As a result of a greater spreading out in residences than in jobs, the length of daily travel has been increasing : since 1975, the average distance between the home and the workplace has multiplied by two (in 1999 it was 15 km for active people who did not work in their residential commune). The time spent in these trips, however, has remained constant, on the order of 30 minutes per trip. This is explained by an increase in the average speed of circulation, which, according to national surveys, rose in cities from 26 to 31 km/h between 1982 and 1994 (Orfeuil, 2001). This increase, which is linked with the intensive use of the private car and the improvement in transportation routes, is especially felt in the outlying zones (from 40 to 43 km/h from suburbs to outskirts, from 22 to 29 km/h from outskirts to outskirts), but it remains stable for trips made in the city centers. In its general investigation of transportation, INSEE showed that for the whole of France, an average door-to-door trip in a private car took 16 minutes, as against 36 minutes in public transportation.
Interpreting Recent Trends The process of urban sprawl and reduction in population density at the local level began in 1968 for Paris, and from 1975 on for the other French cities, which then saw their outlying areas grow twice as rapidly as before. The awareness of peri-urbanization was thus delayed by the lack of an appropriate definition when it first appeared. Bauer and Roux’s study (1976) and the Mayoux report (1979) had already warned of the magnitude of the phenomenon, but it was the results of the census of 1982 in particular that provided a more complete picture. “Renaissance des communes rurales ou nouvelle forme d’urbanisation?” asked the statisticians Boudoul and Faur (1982). The changes in the population of the cities between 1975 and 1982 were experienced by many as a break in the process of urbanization, marking a renewal of rural communes and perhaps the “end of the cities” (after the title of the work published by the sociologist Chombart de Lauwe in 1982). The higher rate of growth of the small towns in this period, the decline of the central population densities in most of the urban agglomérations, as well as the migration of populations from the city centers towards rural communes, were often interpreted as ‘counter-urbanization’, following the description of B. Berry for North America (1976), which was taken up again in connection with Europe by T. Champion (1989).
A number of publications offered a different interpretation, however, supported by a longer-term analysis of the evolution of the spatial distribution of urban growth (Pumain, 1982 and 1983). According to this interpretation, the time-honored process of the concentration of the population in cities at the national level continued during the whole of this period (confirmations of this theory came in the censuses of the 1990s, which attested in particular to a return to growth in the metropolises). In addition, the process of urban sprawl, at the local level, certainly marked a reversal in the tendency towards increased density (a reversal that had been begun almost two centuries earlier in the central quarters of the largest cities), but which can also be interpreted, in a certain historic continuity, as an expansion of the cities into an accessible space enlarged by the use of the automobile, but relatively stable in distance-time (Bretagnolle, Paulus. Pumain, 2001). This tendency seems likely to continue, but at what rate?
At the level of the aires urbaines, the de-concentration of the population of the city centers and suburbs towards the peri-urban zones will probably continue. The differences in demographic evolution of the city center, the suburbs and the peri-urban ring that were observed between 1990 and 1999 definitely confirm the continuation of the tendency for the populations to become spaced out in the city centers: the farther one goes from the center, the greater the average demographic growth, even if on the periphery this growth is taking place among smaller groupings of people (Table 3). The increase in the activity of women, and, as a result, the number of two-job households, contributes, in theory, to swelling the flow of inter-communal home-to-work travel. If the present tendencies in the development of urban mobility continue, the number of kilometers driven in cars in the French city-suburbs units could increase by 30% between now and 2010 (GART, 2001).
A comparison with the evolution of mobility in North America leads nevertheless to some more nuanced conclusions. While the rate of motorisation was already very high in North America in 1960, mobility progressed relatively little there between 1960 and 1990: it ‘only’ doubled, at a time when it was almost multiplied by three in Europe. By analogy we might expect that the evolution in the rate of motorisation and the use of the car in France could be much slower than in the past. This being said, mobility continues to increase significantly in the United States, when the country is much more motorised than the European countries. In addition, there are still an important reservoir of non-motorised population among older people, women, and especially the young (Orfeuil, 2000).
One of the principal stakes for urban mobility in the years to come is the greater role of trips from outskirts to outskirts. The growing difficulty in managing these movements is increased, furthermore, by the impact of the law establishing the 35-hour workweek on the de-synchronization of urban time. Public transportation, conceived for mass movement in dense zones, is not well adapted to the spatial and temporal scattering of urban mobility.
However, although the process of peri-urbanization continued in the 1990s, the results of the last census showed clearly that the slowing in this process, which had been detected in 1990, was being confirmed. The contraction in the rhythm of urban sprawl is perhaps not unrelated to the drop in French population growth. It can also be partly explained by a return to the rise in demographic growth in the city centers. The relative recovery of certain city centers had been noted in 1990, but became even more apparent in 1999, since the great majority of central communes ceased to lose inhabitants between 1990 and 1999.
Nevertheless it is not enough merely to extend the quantitative tendencies in order to draw up the future contours of urban sprawl in the cities. The factors that explain this movement have been modified because of the change in the social and political contexts, and also from analyses of the consequences of urban distribution. These transformations must be taken into account if we are to make a correct assessment of possible future situations.