Third World Cities In Global Perspective

The Political Economy Of Uneven Urbanization

David O Smith, David Smith

• Publisher: Routledge
• Year: 2019
• eBook ISBN: 9780429975295
• Topic: Politics & International Relations
• Subtopic: Politics
• Index: Politics & International Relations

In this innovative book, David Smith ultimately links what happens on the ground in the neighbourhoods where people live to the larger political and economic forces at work, putting these connections in a historical framework and using a case study approach. The societies of the world's underdeveloped countries are now undergoing an urban revolutio

---

1

Developing an International Political Economy Approach

Dominance and Dependency: A Theoretical Reprise

Notions about the emergence of hierarchy and dominance within urban systems are very old in U.S. social science. Early writings by Norman S.B. Gras (1922) and Roderick McKenzie (1933) on “metropolitan dominance” pointed to the linkage between the ascension of a city to a position of economic and political importance and the concomitant subordination of surrounding secondary towns and rural areas to this center. In fact, some of McKenzie’s observations were particularly prescient in terms of the new global political economy perspectives (see D. Smith, 1995, for a complete discussion). In papers addressing “industrial expansion” and the emergence of “dominance in world-organization,” he highlighted the “predatory and exploitative” history of European expansion, discussed economic “penetration” of less developed countries, and suggested a core/periphery structure and a cyclic pattern of uneven development in which cities and urbanization play key roles (McKenzie, 1927, 1933). More recently, in a major work on twentieth-century urban patterns in the United States, Otis Dudley Duncan and his collaborators (1960: 133) emphasized the importance of studying “the structural geographic patterns of flows” between U. S. cities in shaping the emerging national “urban hierarchy.” The emergence of dominance relations within large-scale social systems is a basic principle of the human ecology approach (Hawley, 1968a: 333).

Dependency theory initially emerged in the form of a vitriolic attack on the culturalist and developmentalist assumptions of modernization theory by a group of Latin American social scientists.¹ Like human ecologists, some early proponents of this approach were also interested in the development of an exchange system with asymmetrical relationships between the parts. In this case, however, the boundaries of the system have been set by the imperialism and neo-imperialism of the advanced nations. In the language of urban systems, Andre Gunder Frank (1969: 6–9) described the result as “a chain of constellations of metropolis and satellites” that relate

all parts of the whole system from its metropolitan center in Europe or the United States to the farthest outpost in the Latin American countryside …. When we examine this metropolis-satellite structure, we find that each of the satellites … serves as an instrument to suck capital or economic surpluses to the world metropolis of which all are satellites … within this world embracing structure, the metropolis tends to develop, and the satellites to underdevelop.²

Dependency theorists argued that as a result of the penetration of foreign political and economic influence, many formerly undeveloped areas became incorporated as “peripheral” or “dependent” underdeveloping appendages to the colonial core states. This type of perspective sensitized the researcher to the importance of international exchanges such as trade and foreign aid and investment. It suggested that “unequal exchange” between colonial cities and core metropoli benefits the latter and leads to the stunted or distorted growth of the former.

In its earliest formulations, dependency theory was intended to be a rather narrow framework for dealing with the problems of contemporary underdevelopment in peripheral areas, particularly in Latin America. One problem, especially in the least sophisticated versions of this approach, was the rather limited attention given to the two-way relationship between the imperial (or neo-imperial) core nation and the colonized (or neocolonized) peripheral country. Additionally, the terminology of dependency theory seemed to imply an either-or situation—dependent or autonomous, metropolis or satellite, core or periphery. These deficiencies are at least partially corrected when a more comprehensive conception of the world economic system, like that proposed by Immanuel Wallerstein, is accepted (Evans, 1979a: 16). While this “world-system” perspective still focuses on external conditions affecting the development of peripheral areas, the emphasis is on “the consequences of occupying a given structural position within the world system as a whole” (Evans, 1979a: 15). This reconceptualization further allows for (1) an intermediary stratum of semiperipheral countries as “a necessary structural element in the world economy” (Wallerstein, 1974a: 349) and (2) the possibility of mobility in the international system through dependent development (see Cardoso, 1973; Evans, 1979b).

Even with the advantages that the world-system reformulation enjoys, this approach has some important weaknesses. One major problem for the entire dependency/world system school involves a tendency to place a disproportionate emphasis on the exchange aspect of the world-economy and to give inadequate attention to the internal dynamics of areas that are labeled “dependent” (Portes, 1979: 5–6). This issue of “internal” versus “external” determination has become a central concern for scholars adopting the dependency/world-system approach (Frank, 1979). A major thrust of this perspective is that the exclusive intrasocietal perspective of modernization theory (or, for that matter, orthodox Marxism) is incorrect. But too much emphasis on the exchange relations between nations has led to charges of “economic reductionism” (Skocpol, 1977) and “neo-Smithian Marxism” (Brenner, 1977).

One remedy for the problem is to also analyze the relationship between exchange and the local modes of production and class structures (Evans, 1979a: 19; Cardoso and Faletto, 1979: xvi). Linking class structure and local policymaking with variables that the human ecological approach emphasizes, particularly changes in transportation and communications systems and technology (see Hawley 1950, 1971), offers a particularly promising strategy. The tie to human ecological theory is particularly germane for comparative urban research since cities and urban systems are major foci of this school. In light of these suggestions, a useful way to conceptualize a more adequate, synthetic theory of macrosocial change views the external world-system interrelationships as setting the broad parameters for processes such as urbanization and development, while the internal political, economic, and ecological variables help to explain more specific outcomes.

Another way to move away from economic reductionism is to recognize that the world-system is also an “interstate system” (Chase-Dunn, 1981b). Two of the most damaging critiques of the world-system perspective during the late 1970s focused on the economistic nature of exchange-based “neo-Smithian Marxism” with its lack of attention to political and class forces (Brenner, 1977; Skocpol, 1977). These attacks, directed at the early writings of A. G. Frank and Wallerstein’s (1974a) explanation of the European transition to capitalism, had considerable merit. Further theoretical elaboration and mounting empirical research now clearly demonstrate the inadequacies of economic reductionist arguments that consign politics to a derivative or epiphenomenal status. This led to a major effort to “bring the state back in” to a theoretical synthesis labeled the “new comparative political economy” (Evans, Rueschmeyer, and Skocpol, 1985; Evans and Stephens, 1988). In the case studies in this volume, the role of the state and contending classes in the struggle for dependent development are assumed to be central to the political economy of urbanization in the contemporary Third World.

A major emphasis of the dependency/world-system literature is on the importance of the historical context of social change. Because world capitalism has been an expanding system since its emergence in Europe in the sixteenth century, a full understanding of its operations must take the changing needs of the system and the shifting mechanisms of capital accumulation into account. Therefore, dependency/world-system theorists have found it useful to delineate stages or phases of capitalist development (Wallerstein and Hopkins, 1977: 125; Valenzuela and Valenzuela, 1978: 546; Frank, 1979: 9).³ Because of the importance given to the changing nature of capitalism and the need to blend an analysis of changes in the world-economy with an analysis of local dynamics, researchers must adopt a “historical-structural” approach (Cardoso and Faletto, 1979). The critical aspect of this methodology is that it attempts to examine the history of specific societies in such a way that the linkages between the external and internal factors shaping change are both illuminated.

Cities in the World-System: Previous Research

Research directed specifically at urbanization from the dependency/world-system perspective started to appear in the late 1970s. This work began with the premise that the contemporary patterns and processes of urbanization can only be fully understood as part of the expansion of the capitalist world economy. The patterns of urban growth are shaped by the historical contexts of a region’s initial incorporation into and changing role within this world-system (Walton, 1977: 12–13; Slater, 1978: 27). “The process of urbanization becomes, therefore, the expression of this social dynamic at the level of space, that is to say, of the penetration by the capitalist mode of production, historically formed in the western countries, of the remainder of the social formations at different technological, economic and social levels (Castells, 1977: 44).

The earliest articles developing this “international political economy of cities” approach were primarily intended as attempts at theoretical reconceptualization. They pointed to a number of potentially fruitful avenues of research, but the actual analyses performed were more suggestive than conclusive.

Recent research has attempted to apply the perspective more rigorously to empirical situations. In an introduction to a major compilation of studies, Michael Timberlake (1985: 10) succinctly summarized the shared premise of this research:

Urbanization must be studied holistically—part of the logic of a larger process of socioeconomic development that encompasses it, and that entails systematic unevenness across regions of the world. The dependence relation is an important theoretical concept used to pry into ways in which the processes embodied in the world-system produce various manifestations of this unevenness, including divergent patterns of urbanization.

Within the developing literature devoted to this approach, two distinct modes of analysis are apparent—arising from a methodological division among dependency/world-system analysts in general (for a more detailed analysis, see D. Smith, 1991).

Quantitative Cross-National Analysis

One strain of research on cities in the world-system particularly popular with U.S. trained sociologists stresses a comparative statistical analysis of urbanization and development. Scholars in this school are attempting to formulate hypothesis-testing procedures for the effects of dependency on patterns of urbanization. Specifically, dependency is expected to affect overall levels of urbanization, urban primacy, or “overurbanization,” through intermediary variables such as tertiary sector employment, state power, and so forth. In attempting to perform quantitative cross-national “tests” of dependency/world-system formulations, researchers are expanding on a groundswell of similar studies in the late 1970s.⁴ These studies tried to assay the effect of international dependence, operationalized using measures of trade concentration or direct foreign investment and/or aid, on various indicators of national development.

The first real attempt to relate dependency to patterns of urbanization using such a rigorous cross-national approach was offered by Timberlake (1979). He hypothesized that overall urbanization, urban primacy, and tertiary sector employment are critical intermediate variables that help explain the effects of dependence on economic development, income inequality, and political violence. Using multiple regression and causal analysis, Timberlake found only modest support for most of his models. The mixed results of this study may be attributable to several factors. First, the author himself admitted “that dependency is a concept in need of greater specification” (Timberlake, 1979: 141). He argued that one must differentiate between various types of dependency—and the mechanisms through which these forms are likely to affect processes like urbanization. Second, the study faced a familiar problem in cross-national research: a tremendous amount of missing data. Only a handful of nations had complete information on the variables of interest. As a result, Timberlake’s regression equations covered between 14 and 27 cases—which leads to great difficulty in interpreting results. Finally, the study used a fairly easy to calculate but widely criticized measure of urban primacy employing data on only the four largest cities, called the Davis Index (see Davis, 1970). This indicator has been criticized as a very crude operationalization of the concept of an even/skewed city-size distribution (C. Smith, 1982b; Walters, 1985). Yet despite its flaws and lack of conclusive results, Timberlake’s work was a pioneering attempt to systematically apply the dependency/world-system approach to global urbanization.

Following Timberlake’s lead, Christopher Chase-Dunn and his associates systematically developed the quantitative approach to urbanization in the world-system (see Chase-Dunn, 1979, 1982). They sought to redress problems involving lack of data and inadequacy of measures by expending considerable time and energy collecting and compiling time-series measures of these variables. Concurrently, they developed new summary indicators for describing the shapes of city-size distributions (Chase-Dunn, 1982; Walters, 1985).

Several researchers associated with these efforts published empirical studies. Jeffrey Kentor’s (1981) article on the structural determinants of peripheral urbanization concluded that a dependency theory approach provides a more accurate description of this process than a modernization approach does. He reported differential effects for various linear measures of dependency on urban outcomes. Timberlake and Kentor (1983) focused on the issue of “overurbanization”⁵ and found that (1) this condition is significantly related to dependence upon foreign capital, and (2) increases in overurbanization are systematically related to relative economic stagnation in less developed countries. Chase-Dunn (1985a) used the new historical database to examine differential primacy patterns in Latin America during the twentieth century. Although primacy has been a generic characteristic associated with the external dependence of these countries, Chase-Dunn noted that changes in these imbalanced urban hierarchies often do not conform with hypotheses linking primacy to phases of export-led development and even urban growth to periods of loosened core control. He (1985b) also used these data to argue that the world urban hierarchy over long periods of time has cyclically fluctuated in sync with changing phases of single-nation hegemony versus political multicentricity in the world economy. Directly inspired by the efforts of the Chase-Dunn team (and using their data), Robert Fiala and David Kamens used simple descriptive statistics to compare national urban primacy ratios in the various “zones” of the world economy from 1800 to 1970. In addition to showing long-term trends keyed to long cycles of global competition, these data clearly revealed very high (and growing) lead-city primacy in noncore areas late in the twentieth century and differentiated between peripheral and semiperipheral urban patterns.

In the mid-1980s, a number of researchers followed the quantitative cross-national path using the hypothesis-testing methodology of multiple regression. David Meyer (1986) blended the human ecology and world-system perspectives in research examining the international urban hierarchy of the Western hemisphere. Using a “world-system of cities” notion very similar to Chase-Dunn’s, he documented the pattern of core domination and control of the financial and economic activity of South American cities by empirically examining the headquarter—branch office links of major global banks. Bruce London (1987) also used both the human ecology and world-system approaches to formulate hypotheses about the determinants of Third World urbanization for a quantitative cross-national regression analysis. He concluded that both investment dependence and categorical measures of world-system position contribute to the level of urbanization, overurbanization, and urban primacy in underdeveloped countries but that eight “internal” factors, such as “rural adversity,” also have an independent causal effect. Using similar techniques, York Bradshaw (1985, 1987) illustrated the generic influence of economic dependency on overurbanization and urban bias in Africa and throughout the Third World. He also concluded that investment dependence is a significant but not singular cause of deleterious urban patterns in underdeveloped countries. Operating in the same methodological fashion, London and I (1988) also found a similar dependence effect on urban patterns and further demonstrated that both urban bias and multinational penetration contribute to economic stagnation in noncore nations. A consistent thrust in this body of research is the need to move toward synthetic theoretical models that acknowledge the crucial role of world-system forces but incorporate “internal” ecological factors as well.

Comparative/Historical Case Studies

A very different direction in studies of urbanization in the world-system has involved the intensive examination of historical cases of dependent or peripheral urbanization. While Frank’s early writings are quite suggestive in this regard, the first scholar to systematically explore the relationship between dependency and urban patterns was Manuel Castells in his classic monograph on Marxism and urban sociology, The Urban Question. In this volume (Castells, 1977), a chapter was devoted to an empirical and theoretical exploration of differences between urbanization in advanced and underdeveloped areas (chapter 3: 39–63). Castells argued that, like other processes of macrostructural change, urban growth is closely associated with the historical process of capitalist penetration and development. He also set out some working hypotheses about “dependent urbanization,” arguing that city growth, while differing in form and content in various parts of the Third World, nevertheless must be understood as the expression of the imperialist/ neo-imperialist ...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Dedication
• Table of Contents
• List of Tables and Figures
• Preface
• Introduction
• 1 Developing an International Political Economy Approach
• 2 Global Patterns: A Cross-National Analysis of Urbanization
• 3 The Logic of Comparative Historical-Structural Analysis
• 4 Dependency, Development, and Urbanization in West Africa
• 5 Nigerian Urbanization: A Semiperipheral Case?
• 6 Urban Diversity in East Asia: Toward a Political Economy Approach
• 7 South Korean Urbanization and Semiperipheral Development
• 8 Urbanization in Global Perspective: Summary, Synthesis, and Policy Recommendations
• Notes
• References
• About the Book and Author
• Index