Urban shrinkage refers to a concomitant process of demographic and economic decline with a structural impact on two constitutive elements of the city, the density of the population and its economic functions, thus generating considerable social effects (Martinez-Fernandez, Audirac, Fol, Cunningham-Sabot 2012). The notion of urban shrinkage puts emphasis not so much on the spatial dimension and the irreversibility of the process as on the idea of urban decline, sometimes used as a synonym. In contrast to urban growth, it does however stress the attendant loss of population, jobs and wealth. Urban shrinkage is a concept that appeared in English and German, where the phrases Shrinking cities and Stadtschrumpfung have become common. In French, the phrase Décroissance urbaine (in other words the reverse of croissance meaning growth) is used, in an attempt to reflect the ideas of decrease and shrinkage (Florentin, Fol & Roth, 2009). The usage appeared at the end of the 1970s in the USA and at the end of the 1980s in Germany in reference to the processes underway in the large cites in the American Middle-West or the Ruhr. They came into use across the world from 2000, particularly in Europe, to describe the regressive dynamics that were beginning to characterise a number of cities of various sizes and with different economic specialisations. The popularity of the concept can be explained by its (re)discovery of the non-universal nature of urban growth, and the challenges raised by shrinkage for urban developers within a cultural model dominated by the paradigm of growth.
Depending on the country, urban shrinkage has been explored in different fields of research. In the USA, studies on urban shrinkage or decline concerned analyses of the joint effects of deindustrialisation and suburbanisation. In the 1970s, the crisis of the Fordist model was particularly marked in cities whose growth was based on heavy industry. As the cities lost their industries, and an number of the new tertiary activities set up in the suburbs, suburbanisation processes, which accentuated in that period, emptied them of a large part of their wealthier, white population. This process of urban decline (Beauregard, 2003) particularly affected the large cities in the north and east of the USA. Some of them are still in a critical situation (Detroit, Flint, Youngstown etc), which was aggravated by the sub-primes crisis of 2008.
In Germany, the concept of Shrumpfung developed in spectacular manner from the observation in the years immediately after 2000 of the demographic collapse of cities in the former GDR. This demographic phenomenon has been attributed both to the massive emigration of the youngest and best qualified populations to the west and to the combined effects of economic restructuring and the “second demographic transition” (Oswalt, 2006).
In Japan, it was via the theme of the ageing of the population that urban shrinkage was approached. Although the issues of demographic change have long been the subject of study in Japan, the effects on urban growth have received attention only recently in the fields of geography and urban studies. In fact, the spatial effects of the ageing of the population were first of all felt in rural areas and in the former industrial and mining cities, before they gradually affected smaller cities and towns, and today the suburbs distant from metropolises, whose centres have thus far been little affected by urban shrinkage (Buhnik 2010).
In the field of geography, one of the main scientific challenges is the measurement of urban shrinkage in its spatial and temporal definitions, and the choice of indicators. Apprehending the scale, the intensity and sometimes the very reality of urban shrinkage depends very much on the time-scale considered and the area and scale on which observations are made. Thus indicators can suggest a shrinkage process over short periods, while this is not evidenced over a longer period, and the reverse can also occur. The importance of the spatial-temporal definition and of the appropriate indicators is not purely formal. The fact that this was not taken into consideration in the early studies led to the classification as “shrinking cities” of spaces in widely different situations (ranging from a district in a growing metropolis to a declining city in a declining region), and the issues in terms of territorial development were so varied that it questioned the scientific relevance of the very notion of shrinkage.
Urban shrinkage cannot be explored in isolation. Using different approaches, it is analysed in its relationships with other entities. It belongs to the processes of evolution of city systems, where the effects of size and specialisation are unfavourable at the bottom of the urban hierarchy (Paulus, Pumain, 2000). It therefore has links with the debate on spatial inequalities in the setting of globalisation (Fol & Cunningham-Sabot, 2010). It is also analysed for its place in systems of power relationships, so that in Germany the academic debate has recently turned to the question of the scientific relevance of the notion of shrinkage, with constructivist currents of thought favouring the more relationship-based concept of “peripherisation” (Bernt & Liebermann, 2014).
Finally, scientific debate has also focused on urban policies in shrinking cities. One of the main criticisms of these policies concerns the fact that they focus on the attractiveness and competitiveness of cities (Rousseau, 2010; Miot, 2912). Most local strategies do indeed aim to attract investment and also new types of population (such as the “creative classes” identified by Richard Florida), via more or less overt policies of «gentrification». Another debate, very active in seriously declining cities like Detroit, concerns the allocation of sparse financial resources to rehabilitation policies: should action be taken everywhere, or should in be concentrated on a few districts liable to react more favourably to investments, with the risk of abandoning the others? In these two types of debate, it is invariably the outcome for the most underprivileged that is in the balance, and this dimension is often not envisaged in redevelopment policies.
Emmanuèle Cunningham-Sabot, Hélène Roth, Sylvie Fol