Shrinking cities, the hidden challenge

Contents

1. Introduction

2. Growth and decline of cities
2.1. The conditions for city growth and decline
2.2. Time, space and city development
2.3. Shrinkage and the death of public sphere

3. Cities with a past but no future?
3.1. Shrinking industries - Detroit is not alone
3.2. The East German case

4. What can be done?

5. Concluding Remarks

References

Annex
A1: Annual growth rate of global population
A2: Percent aged 65 and over
A3: The World’s 25 oldest countries 2000
A4: Fertility rate in the EU-25
A5: Eurostat: Annual average population change 1996-2001
A6: Population dynamics of broad age groups in East Germany

1. Introduction

The story of world demographics is a growth story and it is very closely linked to urbanization. Since the early 19th century population growth has accelerated dramatically. 118 years, this is the time it took to increase world population from one billion in 1807 to two billion. For the third billion reached in 1922 only 37 years were necessary and the jump from 5 to over 6 billion world population was done in an un-preceded 12 years^[1]. The increase in world population has influenced many observers to use terms such as ‘explosion’ or ‘over-population’ and ‘mega-cities’. Indeed the growth patterns are very closely linked to a sustainable trend towards higher levels of urbanization. As UN data shows 3 billion people are already living in cities. This trend is continuing leading to 5 billion city dwellers by 2030 and the percentage of world population living in cities is expected to pass 50 per cent in 2007^[2].

In the public and also academic discussion of urbanization and world demographic trends, cities such as Lagos, Nairobi, Karachi, Dhaka or Mumbai are often the case studies for the enormous challenges growing cities have to deal with. These challenges have to do with growing scarcity of resources, collapse of infrastructure, poverty and environmental problems.

Although population growth and the challenges it presents for cities, mostly in the developing world, is a very serious issue, there is another topic deserving attention. Overall city populations are increasing but there is no single trend as there is no single city. Growth patterns vary very significantly between cities, countries and even continents. African cities such as Lagos and several Asian cities such as Bejing or Shanghai are expected to grow steadily at least for the next decades, whereas several cities in the Western world already show signs of decline and shrinkage. Former industrial centers such as Detroit, Pittsburgh or Manchester suffering from the economic shift towards a modern service industry are not alone. According to the Berkeley Institute of Urban and Regional Development: “Today, every 6th city in the world can be defined as a “shrinking city”.”^[3].

This paper aims at casting light on the hidden challenge of shrinking cities. Its main hypothesis is that in the current debate on the effects of demographic change and city management shrinking cities are widely neglected but will be a major urbanization issue in the near future.

The first part ’Growth and decline of cities’ presents and discusses world urbanization trends. Hereby the idea is to contrast trends of growing urbanization and population increase with the spreading phenomenon of shrinking cities. Furthermore the conditions for the rise and decline of cities are identified. Based on this more introductory part, ‘Cities with a past but no future?’ focuses on case studies of city shrinkage. Among the most often found cases in the literature are cities such as Detroit, Manchester and East German cities.

2. Growth and decline of cities

As already pointed out 2007 will most probably be the first year in human history that more people will be living in cities than in rural areas. As the graphic on the left shows this trend is going to continue. Nevertheless it shouldn’t be overseen that the already high number of shrinking cities will increase too. Besides very clear cases of growing (mega-) cities (above 5 million inhabitants) such as Lagos growing from 12 million in 2003 to about 24 million 2015 and declining cities such as Hoyerswerda in East Germany there will be several mixed types where growth and shrinking happens simultaneously in different areas of the same city: “Until today we have planned cities for growth, but we must also plan for their decline.”^[4].

illustration not visible in this excerpt

Mega-cities get the highest public attention in this growth process: “Yet, despite their size and importance, mega-cities still account for only a small share of the world’s urban population and of course of the world’s total population.”^[5]. Even though the number of mega-cities is expected to increase significantly by 2015 they will account ‘only’ for 5% of world population compared to 4,1% in 2000^[6].

In the long run- according to U.S. Census Bureau data “the level of fertility for the world as a whole will drop below replacement level before 2050.”^[7]. But already today and in the foreseeable future shrinking populations and in turn shrinking cities play an important role. The continuous population growth (at least until 2050) is very unevenly distributed. Africa and Asia are the present and future growth centers, also with the highest increase in urbanization levels, whereas for example Europe, Argentina, Australia and to a lower degree the US will have stagnating or declining populations^[8]. These patterns will be discussed in more detail in part 2.1. Time, space and city development.

The following two points present the conditions that favor city growth or lead to its decline. There is no single variable that explains the rise and shrinkage of cities; nevertheless it should be possible to determine the most important influence factors of city development.

2.1. The conditions of city growth and decline

The rise and fall of Rome is an expression everybody knows. Lost cities have been found in Middle-America centuries after their inhabitants left them, volcanism wiped out Pompeii and war destroyed cities from ancient Cartage to Nagasaki in World War Two. Natural catastrophes such as recently the Tsunami in East Asia, hurricanes Rita and Katrina flooding New Orleans, earth quakes in Kobe, Japan or Pakistan will always severely affect cities and are hard to calculate.

Besides natural catastrophes, war destruction and other relatively seldom, single events there are less spectacular factors that lead to city shrinkage and general decline. Economic or demographic changes are powerful but very silent factors that usually need decades to be significantly felt by city inhabitants. Usually these factors impact cities in whole regions such as for example the formerly industrialized rust belt in the US or the “Ruhrgebiet” in West Germany. Scarcity of resources, of labor and especially capital might lead to a vicious circle. When city economies focus on a few goods and industries they might grow at a very rapid rate. This was the case for Manchester focusing on textiles and for Detroit relying heavily on the automotive sector. A supply industry develops, whole areas of the city become neighborhoods for workers of very few factories and the sector continues to specialize. As long as the industry performs well, the city as a whole, its tax base, infrastructure and cultural life is prosperous. When the national economy changes in an unfavorable way for the city’s economy the reason for its rapid growth becomes the main factor for its rapid decline. Technological change, high competition or the collapse of (overseas) markets or supply can be seen as at the core of the shrinkage, based on the assumption of competition between cities.

In very general terms resources enable city growth but also limit its scope. The size of ancient cities of Mesopotamia and other early settlements was determined by the availability and quality of its surrounding agriculture. Therefore historical cities can often be found at trade ways and rivers to secure sustainable food supply. As Lewis Mumford remarks: ” These early cities bore many marks of their village origins, for they were still in essence agricultural towns: […] they could not grow beyond the limit of their local water supply and their local food sources.”^[9].

Another characteristic of the early city was its high degree of autonomy and low interconnection with other cities. They were often religious and administrative centers, market places, protected by courts and law enforcement mechanisms, often surrounded by high walls against enemies. Over the course of history natural restrictions such as the quality and quantity of soil, became less and less relevant and finally vanished. On the one hand improvements in agricultural methods and technology increased efficiency and made cities more autonomous. On the other hand improvements in sea- and water transportation and communication systems dramatically improved the interconnection between cities and sustainable food and resource supply.

One can stress religion, centralized administration, the symbolic value of capitals and many other reasons for growth and decline of cities. Since not only but also through globalization cities are nowadays more interconnected than ever before in history I like to see current cities foremost as market places. The view of cities as marketplaces that secured economic exchange is of course not new and goes back to the earliest (urban) sociologists such as Georg Simmel (1905), Max Weber (1905) and others^[10]. Two developments were and are still at work. The transaction costs of relocating capital have decreased and the opportunity costs for inhabitants to stay in a declining city are increasing. Investors nowadays have a higher opportunity to move (their capital, investments) to more productive places and so have city inhabitants. To explain the shrinkage of cities, (economic) utility maximization of city inhabitants, who choose between several competing cities is seen as the main condition for city growth. There is not only a growing mobility of inhabitants between cities but also within the city. Continuously declining commuter costs are among the main factors causing the growing trend of sub-urbanization and therefore declining density. Analyzing empirical data from a study done by Kim (2005) on the rise and decline of U.S. urban densities^[11] helps to identify and to distinct the two trends of sub-urbanization and of shrinkage. The population density of cities increased since 1890 (7,203 inhabitants per square mile), peaked in 1950 with 8876 inhabitants per square mile and is declining since reaching 5647 in 1990 the most recent data entry of the study. This decline of around 36% since 1950 does not necessarily mean shrinkage of the overall city since the size of cities also grew and people might have moved to the suburbs. Therefore one would have to control for the population size of the whole metropolitan region. Putting together the two data tables of Kim for city and metropolitan region development shows that both trends were at work. City density declines by 36%-an estimate for sub-urbanization but also the whole metropolitan region lost inhabitants-an estimate for people leaving to other cities (shrinkage). Here the density peaked in 1960 with 589,4 inhabitants per square mile and declined steadily reaching 288 in 1990, which means a 51% decrease over the observed time period^[12].

2.2. Time, space and city development

This paragraph is about to show how city develop over time and how (population) growth centers and regions where cities shrink are distributed geographically. The following graphic^[13] is taken from the ECOSOC Population Division World Urbanization Prospects Report and gives a first impression how urban and rural population growth varies over time.

One striking pattern is that the urban population in less developed countries will increase steadily and will make up the very biggest share of population growth in the future.

illustration not visible in this excerpt

In the early 1950s the contribution of urban growth from less developed regions relatively to urban population growth from developed regions was 1,3 times bigger. Nowadays urban population growth from less developed countries is 12,7 times bigger than the shrinking population growth in developed countries^[14]. This huge gap between growing urban populations in less developed countries such as in Africa and very low growing or shrinking urban populations in developed nations such as in Europe is going to increase further. The gap widened in the mid 1970s and will steadily enlarge until 2020. Afterwards the gap will decrease just slightly due to increased aging processes in todays so called youth bulge countries of the third world. Given this data it comes therefore at no surprise that the phenomenon of shrinking cities is foremost a Western phenomenon. Dramatic economic changes in the Chinese economy, combined with the one-child policy leading to aging at high pace, will also bring the European and US phenomenon of shrinking cities to the Asian continent. Today shrinking cities are more related to economic changes and areas in Europe and the US than to Asia. But this is going to change when China and today’s very young populations will be aging, partly declining in population size. It is therefore probable that demography will replace economic changes as the key variable influencing the size of cities. But besides vague estimates and hypotheses, how dramatic is the phenomenon of shrinking cities already?

[...]

^[1] U.S. Census Bureau (2004): 1. As annex 1 shows the speed of population growth slowed down after 1999 so that world population will be growing but it will also be aging at a higher rate.

^[2] UN Press Release (2004)

^[3] Berkeley (2005): http://www-iurd.ced.berkeley.edu/scg/index.htm. Since empirical comparative data on shrinking cities is very seldom these results/statements should be interpreted carefully.

^[4] Holcim Foundation (2005): 158

^[5] ECOSOC (2004): 84, see also: Cohen, B. (2004)

^[6] See: ECOSOC (2004): 84

^[7] U.S. Census (2004): 2. Such an estimate should of course be interpreted carefully, since there are numerous influence factors of population growth ranging from economic prosperity over the AIDS pandemic to natural catastrophes and warfare.

^[8] ECOSOC (2004): 1-3

^[9] Mumford, L. (1956)

^[10] For a detailed overview of Webers’ conception see: Flanagan, W.G. (1990): 50

^[11] Kim, S. (2005)

^[12] Se : Kim, S. (2005) : 15-16

^[13] ECOSOC Population Division (2004): 10

^[14] Own calculation based on ECOSOC data. In: ECOSOC Population Division (2004): 10