International Political Economy: An Introduction to Approaches, Regimes, and Issues

 

 

 

Timothy C. Lim, Ph.D.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


 

International Political Economy: An Introduction to Approaches, Regimes, and Issues © 2014 Timothy C. Lim, is licensed under a Creative Commons Attribution (CC BY) license made possible by funding from The Saylor Foundation's Open Textbook Challenge in order to be incorporated into Saylor.org's collection of open courses available at http://www.saylor.org. Full license terms may be viewed at: http://creativecommons.org/licenses/by/3.0/legalcode

.


Table of Contents

 

Chapter 1: Demystifying the Complex World of International Political Economy

Introduction

What Is Globalization (and Why Is It Important)?

Why Do Scholars Disagree?

The Social World as an Open System

Defining International Political Economy: The First Step

Putting the Global in International Political Economy

Globalization: A Reprise

Political Economy and the State-Market Dichotomy

International/Global Political Economy Defined (Finally!)

The Significance of Power

Conclusion

 

Chapter 2: Foundational Theories of IPE: An Unconventional Introduction to Mercantilism, Liberalism, and Marxism

The Three Major Perspectives of IPE: Still Going Strong?

 

Chapter 3: Contemporary Theories of International Political Economy

Introduction

Hegemonic Stability Theory

Post-Hegemonic Theories

Two-Level Games and IPE

Constructivist Approaches to IPE

Conclusion

 

Chapter 4: Politics, Economics, and Cross-Border Trade

The Long History of Cross-Border Trade

Basic Concepts and Data on Cross-Border Trade

Cross-Border Trade: A Still Contentious Debate

The Rise of “Free” Trade in the 20th Century, Part I

The Rise of “Free” Trade in the 20th Century, Part II

Regional Trade Agreements

A Quick Conclusion

 

Chapter 5: The Global Financial System

Introduction

The Global Financial System: The Basics

Constructing the U.S.-Led Postwar Global Financial System

Crises and the Global Financial System

Conclusion

 

Chapter 6: Transnational Production, Foreign Direct Investment, and Economic Development

Introduction

Transnational Production: Definitions, Concepts, and Basic Data

Explaining the Transnational Production Structure

Transnational Production, FDI, and Economic Development

Transnational Production and State-Firm Interactions

Conclusion

 

Chapter 7: Inequality, Poverty, and Exploitation in the Global Economy

Inequality, Poverty, and Exploitation: An Overview

Basic Concepts and Data on Inequality, Poverty, and Exploitation

Why Does Poverty Exist?

How Can Poverty Be Defeated? Beyond Conditionality

Poverty and Capitalism: An Unbreakable Bond?

Hyper-Exploitation: The Resurgence of Slavery in Africa

Summing Up: Marxist and Liberal Views of Capitalism, Exploitation, and Poverty

Capitalism Is What People Make It

Conclusion

 

Chapter 8: Governance in the Global Economy

The Need for Governance

What Is Global Governance?

The Significance and Relevance of Global Governance

Transnational Organizations and Global Governance

The Main Points of Contention in the Global Political Economy

A Very Brief Conclusion

 

Bibliography

 

Glossary

 

Chapter Answers:

Chapter 1

Chapter 2

Chapter 3

Chapter 4

Chapter 5

Chapter 6

Chapter 7

Chapter 8


Chapter 1

 

Demystifying the Complex World of International Political Economy

 

 

Introduction

Text Box:  

Figure 1.1. On the left is Friedrich Hayek and on the right, John Maynard Keynes. The debate between these two influential economists is still unresolved. 

Image sources: The photo of Keynes is in the public domain, and is free to use for publication purposes. The photo of F. Hayek is licensed under the Creative Commons-Share Alike 3.0 unported license, and was released by the Mises Institute. The mash-up of the two photos was done by the author.
















D

            There are many books on international political economy, or IPE for short. Not surprisingly, each contains its own assumptions and views about the key concepts, issues, and concerns of IPE. Sometimes the authors of these various books hold the same assumptions and share the same, or at least very similar, views about how the world works. Sometimes they don’t. In fact, as we will see in a few of the chapters that follow, the perspectives of the people who write and think about IPE are often dramatically, if not fundamentally, different. You may already have an inkling that mainstream economists and radical economists (e.g., Marxists) do not agree on many central issues and concepts. But even among those who seem to share basic ideas, there can be sharp disagreements. Within the broad school of neoclassical economics, for example, there is an intense and still-unresolved debate between those who believe that markets must be left alone and those who believe that government intervention in markets is sometimes necessary. This debate is encapsulated in the ideas of, and debates between, two famous economists—John Maynard Keynes and Friedrich Hayek. Keynes, who died in 1946, is best known for his ideas about the importance of “pump priming,” which refers to deficit spending by governments in times of recession or depression. The goal is to increase demand and create a virtuous circle: higher demand means more need for workers, more workers keeps demand strong, and strong demand keeps the economy going. Keynes’s ideas, it is important to note, are far from dead: the global recession that began around 2008 spurred the United States government to engage in stimulus spending—a type of pump priming—and other policies (including maintaining historically low interest rates and quantitative easing). These are all Keynesian policy prescriptions. Hayek, by contrast, expressed profound confidence in the ability of markets to take care of themselves, and saw only a very limited role for governments at the national level, one based on ensuring a relatively stable supply of money. Hayek is most famous for his classic book, The Road to Serfdom, first published in 1944. In The Road to Serfdom—which has become one of the bibles of libertarianism, along with Ayn Rand’s Atlas Shrugged (1957)—Hayek argued strongly that government control of economic planning inevitably leads to the loss of individual freedom. While his ideas were marginalized in the 1940s and 1950s, they found a much more receptive audience beginning in the 1970s; since then, Hayek’s writings have developed a very strong and even fervent following, especially among policymakers in the United States and Great Britain.

Significantly, the debate between followers of Keynes and followers of Hayek has been going on since the late 1930s. Think about this for a moment: in seventy-odd years, mainstream economists have yet to reach consensus on a fairly basic issue (i.e., Does stimulus spending work or doesn’t it?). Indeed, in an important respect, the disagreement today is even stronger than in the past, when there were long periods in which one or the other view held sway. While neoclassical economists continue to debate a range of issues, it is important to emphasize from the outset that neoclassical economics is not the same as international political economy. As I will discuss in detail below, IPE is a distinct field of inquiry. There is, to be sure, some overlap between the two fields—neoclassical economics and IPE—but there are also areas of very strong divergence. One of the most salient differences is embedded in the terminology itself. International political economy considers politics and economics to be inextricably intertwined, while neoclassical economics asserts that economics and politics are—and should be—two essentially separate areas or processes.

We will consider this issue in much more detail below. For now, it is also important to emphasize that, as a field of study, IPE is much more strongly connected to the discipline of political science than it is to economics. The reason for this is clear: IPE is an outgrowth of international relations, or IR for short. IR, a major subfield within political science, has traditionally focused on the struggle for power between and among states. Although a diverse and heterogeneous field in its own right, IR has long been dominated by a particular theoretical perspective known as realism. Realism, in turn, has long held a heavy bias toward “high politics,” which refers to all matters considered vital to the survival of the state. In practical terms, this entails a near exclusive focus on military-strategic issues. Economic concerns, therefore, are relegated to the domain of “low politics” and, as the term implies, are considered relatively unimportant. For many scholars, however, both the dismissal of economic concerns as unimportant and the implicit separation of politics from economics is unwarranted: in a nutshell, this is what led to the emergence of international political economy as a distinct field of study (beginning in the 1970s).

 

What Is Globalization (and Why Is It Important)?

            Interestingly, in moving away from IR, IPE scholars continued to use the word international to describe the field. Yet, as most of us recognize, the world is increasingly characterized by the phenomenon referred to as globalization. There are, unfortunately, not only many ways to define the term globalization, but there is also no general consensus on how it should be defined. We cannot resolve the debate here; for now, then, suffice it to say that globalization is a complex and multidimensional (economic, political, social, technological, and cultural) process that involves a compression of time and space (Harvey 1989). The time-space compression, most simply, is a situation in which geographic distance has become less and less an obstacle to communication and information flows, to production, and to the movement of goods, people, ideas, and capital around the world. Time-space compression is represented in many developments, but nowhere is it more evident than in the Internet and other forms of information technology. Today ordinary citizens can instantaneously communicate with thousands, even millions of people across the globe at the press of a button via Twitter (or the Chinese version, Weibo), Facebook, or Pinterest; the cost of this communication, moreover, is practically nil for the individual user. Somewhat slower, but still significant, are video-sharing sites such as YouTube, which are used by regular people, influential organizations, governments, and powerful corporations alike. In 2012, for example, the group Invisible Children posted a video called Kony 2012, which appealed for Joseph Kony’s capture and arrest for his role in the commission of crimes against humanity and war crimes against civilian populations in Uganda. The video has generated almost 100 million hits (as of June 2013).

Globalization, as the foregoing discussion suggests, also means increasing interconnectedness, through which the actions and activities of states, societies, organizations, and peoples in one place can have significant reverberations in many other places, virtually anywhere on the planet. Such descriptions of globalization have become trite, but nonetheless, the implications of globalization remain immense, especially for the field of international political economy. Indeed, as we have already seen, in the era of globalization, the label international may well have become anachronistic. The term era, it should be noted, is generally defined as “a long and distinct period of history with a particular feature or characteristic.”[1] The era of globalization, therefore, necessarily implies that what exists today is meaningfully different from the past. This does not mean that globalization is an entirely novel phenomenon. It is not. But as one scholar put it, “globalization did not figure continually, comprehensively, intensely, and with rapidly increasing frequency in the lives of a large proportion of humanity until around the 1960s” (Scholte 2001, p. 17).

A key implication of globalization is in the de-linking of specific processes, relations, and activities from specific territories. Territorial and political space—as in a country such as the United States, Brazil, or China—still matters a great deal, but globalization is challenging the still-prevailing idea of a single territorial state as the exclusive site for organizing political, economic, and social relations. This means, in turn, that the state is losing its unrivaled status. We will return to this issue below (in the section “Putting the Global in International Political Economy”), so for now it is enough to say that globalization has expanded the influence and power of a range of nonstate or transnational entities. Corporations, of course, are among the most important of these entities, but so too are nongovernmental organizations (NGOs), regulatory agencies, associations, social movements, and the like—many of which “treat the whole planet as their actual or potential clients” (Scholte 2001, p. 16).

 

Why Do Scholars Disagree?

            Earlier I discussed the disagreements that exist among economists, between various disciplines and fields of study—such as neoclassical economics and IPE, or IPE and IR—and among scholars and analysts more generally. Left unaddressed, however, is the question of why such disagreements exist in the first place. That is, why don’t scholars and others who think about economics, politics, or international political economy agree on how things work? Why can’t they even seem to agree on what factors or processes are most important in the world (political) economy? There are many answers, but perhaps the most fundamental reason has to do with the subject itself: as with all social sciences, IPE ultimately studies the behavior and actions of human beings, which means that, unlike the physical world of the natural sciences, the social world is populated by subjects with the capacity to think, learn, and make willful choices. To understand the difference, just imagine if atoms, planets, and chemicals had minds and wills of their own. Certainly, the task of physicists, astronomers, and chemists would be far more difficult than it is already. But it is not only individual consciousness that separates the social from the natural sciences. The social world is also composed of historically contingent structures, institutions, and systems of belief (i.e., cultures). The term historically contingent means that major elements of our social world are the product of specific and sometimes unique processes and circumstances that make, say, the United States meaningfully different from Japan, or Japan different from France, and so on. In every country, too, there are often dramatic differences—including differences in national identity and culture—between different time periods, such as Japan in 1850 compared to Japan in 2013. To fully understand or explain the social world, then, it is not enough that we find the “universal key” to individual behavior (which some social scientists claim to have done with the concept of rationality); we must also try to understand how the broader social, economic, and cultural contexts in which individuals live alter, shape, and constrain—in both subtle and dramatic ways—the behavior of people and the types of societies, polities, and economies they produce. Given this, a grand totalizing theory (a single theory that explains everything) is exceedingly hard to imagine.

 

The Social World as an Open System

            From a different perspective, we can say that the social world is, by nature, an open system, which basically means that the subjects or objects (e.g., forces or factors) we want to study cannot be isolated from other subjects or objects in the environment. In many of the natural sciences, by contrast, objects of study can be isolated, albeit not always completely. The whole point of many experiments in the natural sciences, in fact, is to create closed or quasi-closed systems so that regular sequences of events can be observed under carefully controlled conditions (Sayer 1992). It is this ability to create closed systems that has led to the precision and predictive success of certain sciences like physics and astronomy. On the other hand, there are some natural sciences—e.g., meteorology, geology, and the environmental sciences more generally—that do not have the ability to directly experiment with closed systems, although they can sometimes borrow from closed-system sciences to establish rough predictions and explanations (ibid.). It is the relative lack of precision and predictive power in these more open-system natural sciences that has led to some intense and even fundamental disagreements, which may or may not be amenable to resolution in the long run. One of the most prominent examples of this is the continuing debate over global warming. After decades of intense research, for example, there is near-universal scientific consensus on theories related to the greenhouse effects of global warming, yet because the global environment is inherently open, there continues to be room for debate.

In general, the difficulties posed by open systems in the environmental sciences pale in comparison to those in the social sciences, where subjects also have the added capacity for learning and self-change; this means that the subjects of study are, in principle, never exactly the same from one time period to the next (or even from one minute to the next). Thus, it should be even less surprising that sharp theoretical disagreements in IPE not only exist, but also show no sign of ever disappearing. The basic reason, to repeat, is clear: the inherent openness of social systems means that there will always be strict limits on what we can know (although there are many stalwart social scientists—including, no doubt, several of your past or current professors—who never have accepted this, and probably never will). Given these limitations, the goal of this book is not to provide a definitive, much less objective exposition on international political economy. I especially do not want to tell you the “proper” or “correct” way to think. Rather, I want to provide you with knowledge that will enable you to develop your own ideas and frameworks of analysis. In this regard, I have two other related goals. First, I wish to get you to think more clearly and explicitly about your own (theoretical) assumptions, values, and beliefs. This is a critical step, since many students often do not understand the basis and/or implications of their own beliefs about the world. It is this lack of self-understanding that leads to inconsistent, sloppy, and sometimes contradictory thinking. Second, I want to introduce you to a variety of ways of understanding, explaining, and interpreting the world political economy. In the process, however, I also want to help you make much better sense of the conflicting perspectives and approaches in the field of IPE.

 

Defining International Political Economy: The First Step

Text Box:  
Figure 1.3. In the traditional view, the relationship between the state and the market is seen as adversarial, like two boxers fighting. However, this is not the only way—nor is it necessarily the best way—to conceptualize the relationship between the state and the market.
Source: Library of Congress, New York World-Telegram & Sun Collection. 
http://www.loc.gov/pictures/resource/cph.3c14335/ 
“No copyright restriction known”




            So far we have seen the term international political economy numerous times, but we have not yet discussed its meaning. Before we get to the nitty-gritty, though, a few words of warning are in order. First, there is no universal agreement on how IPE should be defined—although this is definitely changing. This means the discussion that follows will not be as simple or straightforward as you might expect (or want). At the same time, a careful reading of this section will provide you with a better foundation for understanding and interpreting the concepts, issues, and problems that are examined throughout the remainder of this book. Second, it is also important to emphasize, at the outset, a key point: definitions are important. A big reason for this is that they tell us what to include in our analysis and what to leave out. Put in slightly different terms, definitions within a given area of inquiry tell us what is considered legitimate—what matters, or what is relevant—within that field, as well as how it is supposed to be studied. We can certainly see this in the more common definitions of international political economy. Consider this somewhat dated definition from a once-popular textbook, which tells us that international political economy “is the study of the tension between the market, where individuals engage in self-interested activities, and the state, where those same individuals undertake collective action . . .” (emphasis added; Balaam and Veseth 1996, p. 6). This seemingly innocuous definition is based on several important, but unstated assumptions. First, it suggests that there are only two significant subjects of international political economy: (a) markets, which are composed of self-interested individuals (and the firms that they operate), and (b) states, which are the primary political institutions of the modern international system. Further, it suggests that a clear-cut distinction exists between economic or market-based activities and political or state-centered ones. Second, this definition tells us that the most important aspect of the relationship between markets and states is based on tension, which is “a strained state or condition resulting from forces acting in opposition to each other.”[2] In other words, the definition presupposes that markets and states relate to one another in fundamentally adversarial ways.

Text Box:  
Figure 1.4. Marx’s theory is often misunderstood, but it arguably still has much to tell us about the world economy. The class-analytical basis of Marxist analysis is particularly important. Chapter 2 will discuss his theory in more depth.
Source: Unknown. This image is in the public domain because its copyright has expired. 

On the surface, there is nothing terribly objectionable about this definition of international political economy. Markets and states are obviously important, and it seems apparent that a strong degree of antagonism can exist between them. However, in looking below the surface, problems begin to arise. The exclusive focus on states and markets, in particular, is exceedingly narrow. In the definition, for example, states represent the political world, but if political society is defined solely in terms of the state, then whole categories of other actors, issues, and activities are essentially eliminated from view, or at least relegated to the outer margins of the field. According to the foregoing definition, in other words, we don’t even get to ask the question, “Who are the most important actors in world politics?” Yet this would seem to be a crucial question. What if states are not as all-powerful as the definition suggests? What if there are other powerful actors out there in the world? What would this mean to our understanding of how the world works? In addition, defining state-market relations as “tense,” or adversarial, rules out other possible aspects of that relationship. Can the state-market relationship, for instance, be reciprocal or mutually constitutive? (Mutually constitutive means, most simply, that two entities cannot exist apart from one another, or that each part exists on account of and for the other part.) Again, using the foregoing definition, we cannot even ask these types of questions. Yet such questions—and the answers to them—are vital to the study of international political economy.        

Let us take a closer look at the issue of the assumed “tension” or antagonism between the state and the market. A number of scholars have convincingly argued that states and markets are inextricably bound together. Karl Polanyi, to cite one scholar whose work we will focus on later in this chapter, provided a convincing argument that the emergence and subsequent development of a “market society” was made possible by the enormous and continuous intervention of the state. Similarly, Charles Tilly’s now-classic study, Coercion, Capital, and European States, AD 990–1990, showed that the development of the modern state in countries such as Britain and France paved the way for capitalist development. At a more basic level, another prominent scholar, Albert O. Hirschman (1977), convincingly argued that the “invention of capitalism depended on the creation of a new type of political actor—an individual liberal subject who was the product of a liberal state” (Blyth 2009).

In vivid contrast to the state-centered definition of international political economy is the Marxist view, which, generally speaking, focuses on the social relations of production. From a Marxist perspective, the key aspect of political economy (note that Marxism does not distinguish between IPE and political economy more generally) is the inescapable conflict between opposing class interests—that is, between the owners of the means of production (i.e., modern capitalists) and wage laborers (i.e., workers). The questions and answers that result from a definition that focuses on the tension between states and markets versus one that focuses on opposing class interests are, needless to say, likely to be quite different. Yet, as with the more conventional or mainstream definition, Marxists also tell us on whom we should focus all, or the bulk of, our attention—i.e., social classes—and how we should conceptualize the relationship between the key actors: as conflict-ridden. There is no middle ground here, either. This definition, then, can be just as problematic as the first if we assume that the world is more complex than the definition indicates, which I think it is. Let us just consider one problem. In the Marxist definition, no mention is made of the state. The reason for this, at least to Marxists, is clear: (classical) Marxists considered the state to be an appendage, or tool, of the dominant class. That is, the state existed to serve the interests of capitalists. Period. It is difficult, however, to sustain this argument, since we can, with relative ease, find evidence that states can and do act against the interests of dominant economic actors—not always, but more than just occasionally. Recognition of this fact, by the way, has compelled contemporary Marxists to offer up the notion of relative autonomy, which acknowledges that states are, indeed, actors with their own interests, but that they are “relatively autonomous” because they can never be totally independent of dominant class interests.

As I suggested above, things are changing. David Balaam (one of the authors cited in the first definition), for example, later amended his conceptualization of IPE by adding societies into the mix. Specifically, he (and a different co-author) wrote, “as a subject area or field of inquiry … [international political economy] involves tensions amongst a variety of state, market, and societal actors and institutions” (emphasis in the original; Balaam and Dillman 2011, p. 7). This amended definition clearly expands the range of relevant actors (and could easily include social classes); it also explicitly introduces another type of actor— namely, institutions. We will learn more about institutions later; for now, suffice it to say that many scholars, and perhaps most, agree on broadening the domain of IPE (although the amended definition still insists on restricting IPE to the “tensions amongst” these actors). More generally, then, we are seeing a shift to an inclusive definition of IPE. This has had significant implications. Most importantly, it has allowed hitherto excluded or ignored actors, activities, and issues to finally be fully incorporated into the mainstream. Over the years, for example, we have seen more emphasis on a range of societal or nonstate actors: corporations, labor unions, social movements, criminal organizations, nongovernmental organizations (NGOs), religious institutions, epistemic communities, and so on. We have also seen an expansion of issue areas. Twenty years ago, most IPE textbooks would focus on a limited number of topics, especially international trade and monetary relations, international finance, and international debt and development. Today, such issues as the global environment, cross-border migration, social movements (including indigenous peoples’ movements), poverty and hunger, nationalism, gender, and race/ethnicity are considered appropriate topics of study in international political economy.

 

Putting the Global in International Political Economy

            In the introduction to this chapter, I briefly discussed a significant limitation of IPE, a limitation that stems from the use of the term international. Strictly speaking, international applies only to relations between and among sovereign states. The term also implies a clear distinction between the national and the international—between what goes on inside states and what goes on outside states. With just a little reflection, though, it is clear that a great deal of, and likely most, economic activity that occurs in the world today is conducted—and sometimes controlled—by nonstate actors in ways that transcend national boundaries. Most of us know, for example, that large corporations engage in all sorts of economic transactions and activities that cut across borders: from buying, selling, and trading products and services, to building and investing in global chains of production (whereby a single product is designed, manufactured, assembled, distributed, and marketed in various locations throughout the world), to forging strategic alliances with other corporations based in a range of different countries. We even have a special name for these types of firms: transnational corporations, or TNCs for short.

Text Box: Table 1.1. Revenue of World’s Largest Corporations (by Sales) Compared to GDP of Selected Countries, 2011 (in U.S. $ Billions)

Country	GDP	Corporation	Sales
Austria	419.2	Wal-Mart Stores	421.8
UAE	360.1	Shell (Netherlands)	369.1
Thailand	345.6	ExxonMobil (U.S.)	341.6
Greece	303.1	BP (UK)	297.1
Chile	248.4	Sinopec-China	284.8
Ireland	217.7	PetroChina	222.3
Philippines	213.1	Toyota (Japan)	202.8
Algeria	190.7	Chevron (U.S.)	189.6
Romania	189.8	Total (France)	188.1
Kuwait	176.7	ConocoPhillips (U.S.)	175.6

Sources: Statistics for corporations comes from Forbes.com (“World’s Biggest Public Companies 2011”), available at http://www.forbes.com/lists/2012/18/global2000_2011.html; data for GDP comes from the International Monetary Fund, “Report for Selected Countries and Subjects,” in the World Economic Outlook Database, April 2012, available at http://www.imf.org/external/pubs/ft/weo/2012/01/weodata/index.aspx. 
The ability of TNCs to quickly and (relatively) cheaply move operations and assets—physical, financial, technological, et cetera—across borders is a fairly recent phenomenon. To be sure, for a very long time corporations have had operations in multiple territories, but establishing and maintaining a presence across the globe was a slow, arduous, and expensive undertaking. As technological, financial, and political barriers have begun to fall away—as the world, according to a popular saying, has become “smaller” (this refers to the notion of time-space compression, which I discussed earlier)—the costs associated with operating on a transnational basis have decreased rapidly. Today, according to UNCTAD (United Nations Conference on Trade and Development), there are over 82,000 TNCs with as many as 810,000 foreign affiliates (UNCTAD 2009, p. 222). Consider just one well-known example: Toyota Motor Corporation. Toyota has operations and facilities in 27 countries and regions, and its products are sold in 160 countries. Toyota’s revenues, moreover, totaled almost $203 billion in 2011, which was more than the GDP of 150 countries. Indeed, as a group, corporations—not states—directly control most of the world’s productive, financial, and technological resources. The combined sales of the top ten corporations in the world in 2011 were $2.69 trillion, which is larger than the GDP of all but four countries (the United States, China, Japan, and Germany). While a comparison of corporate revenue to GDP is admittedly simplistic, it nonetheless gives a general sense of the economic size of corporations relative to most states. Where, when, and how TNCs decide to invest, manufacture, and/or distribute their products is therefore of considerable importance to the world political economy. The rise of the transnational corporation, in sum, means that we can no longer just talk about states (actually, this has been true for quite some time). This does not mean, however, that corporations have surpassed states as the primary sources of power in the global economy. They have not. Yet, it does mean, to repeat, that we can no longer analyze the international Text Box: Figure 1.5. An Example of Regulatory Arbitrage
 
In an era of globalized production, states have more difficulty regulating and taxing TNCs, since corporations can easily relocate some or all of their operations to countries with minimal regulations and taxes. Mobility and their productive capabilities give TNCs significant power in the world economy.
Source: Created by author.
political economy as if only states have power.

Indeed, many scholars argue that TNCs are now able to directly challenge states’ authority to regulate their activity. Consider this simple, but oft-cited example: by threatening to limit or close down their operations in a given location, corporations can compel governments to modify local regulations or standards for health, safety, wage levels, and/or the environment—a phenomenon dubbed regulatory arbitrage. In essence, TNCs are telling states, including the most powerful ones: “If you want our business, then you have to play by our rules.” Another example can be found in the area of corporate tax arbitrage—an issue that became particularly salient in 2013, when Apple Computer was criticized for “offshore profit shifting” in order to cut dramatically the taxes it pays in the United States. Again, this is not to say that TNCs have necessarily become the equals of the largest and most powerful states; instead, it means that the relationship between states and large corporations is not as clear-cut or unilateral as it once appeared to be. In fact, even the most powerful state in the world—the United States—is not immune from corporate power. In the area of international trade, for instance, it is well understood that U.S. policy is influenced, and even sometimes dictated, by the interests of corporations. A noteworthy example of this was the decision, by the Bush administration in 2002, to impose tariffs on imported (foreign) steel. Many observers have argued that it was pressure from the U.S. steel industry that drove the administration’s policy decision, rather than the interests of the country as a whole. This may seem an obvious point, but it is a very important one to keep in mind: if state action is even partially determined by nonstate actors, this tells us again that we cannot focus exclusively on what states themselves do. (As we will see, too, such policy decisions are complicated: Bush’s actions may have pleased steel-producing companies, but they hurt steel-consuming companies, as well as consumers.)

Globalization: A Reprise

            The increasingly important role that TNCs (and other transnational actors) are playing in the world today can be attributed, in large part, to globalization. Again, globalization is one of those terms about which there is no broad-based consensus. While some see it as an over-hyped myth, others argue that it has already brought about fundamental changes to the world. I am more sympathetic to the latter view. That is, I believe that globalization is a critical phenomenon that must be accounted for in any examination of international political economy. To repeat, globalization is a complex, multidimensional, and ongoing process. The most salient aspects of globalization—that is, those aspects most people think about when they hear the term globalization—are economic. Economic globalization is more than the simple extension of economic activities across borders, which has been going on for centuries. Instead, it refers to the functional integration of economic activities across borders (Dicken 1998). Imagine a network with connections crisscrossing the globe; each point (or node) has a different, sometimes very specialized function, whether in manufacturing, finance, transport, marketing, sales, or something else. Each point of activity relies on other nodes to do their part in creating or sustaining a larger whole. This is, in very simple terms, functional integration. One important implication of this condition is a de-nationalization of corporations. For the most part, we still tend to think of corporations as essentially American, German, Chinese, Mexican, and so on. Yet, when corporate operations are part of a globalized network, nationality matters less and less—or, perhaps more accurately, it matters in different ways. In the past, to paraphrase a famous quote by the former head of GM, what was good for General Motors (or any other U.S. company) was good for the country. In the era of globalization, this is not necessarily the case. What is good for GM might be good for the United States, but it also might be good for Russia, China, the European Union, Brazil, India, and a slew of other countries where GM invests and sells its products. At the same time, the de-nationalization of TNCs does not mean that political borders have, or will necessarily, become irrelevant. Political space still matters! At the most basic level, we know this because of a point we already covered— namely, that states and markets are mutually constitutive. Doremus et al. put it this way: “Without stable political foundations, markets collapse” (1998, p. 3).

Globalization, I must emphasize, is not only about deepening economic integration and interconnectedness. Another key aspect of globalization is occurring in the realm of ideology, values, and beliefs. This means, in part, that people throughout the world are beginning to communicate—albeit as a product, to a significant extent, of advances in information technology (IT)—in terms of a common discourse centering around human and political rights (especially democracy), social, economic, and environmental justice, and global governance. Just how meaningful this “globalization of ideas” might be is still open to debate, but we can see evidence of its impact with increasing frequency: from the peasant rebellion in Chiapas, Mexico, to the Arab Spring, from the protests against sweatshop labor in the garment districts of New York, Honduras, Haiti, and Los Angeles to the Occupy Wall Street (OWS) movement in the United States and Europe. What is significant about—and common to—all these cases is that, to varying extents, each is based on an appeal to a set of rights that were once almost entirely within the domain of the state. Even more significant is the fact that social, political, and civil rights are now becoming part of a relatively autonomous and transnational source of authority that is increasingly being used to delegitimize state actions (Sassen 1995), as well as the activities of TNCs. The protests against Nike, Reebok, and other corporations that “exploit” workers in poor countries are a good example of the latter. In short, the globalization of ideas, like the globalization of production and finance, is showing the potential to challenge the authority of states and other powerful global actors both within and outside national borders.

On the surface, the changes that globalization are bringing about appear to be moving us in a generally positive direction: toward a weakening of central control (by states, for example) and toward a greater dispersion of power among a plethora of institutions, groups, and individuals—that is, toward a more democratized world. On this point, some argue that the Internet and other advances in information technology are helping to empower and give voice to people whose concerns may not otherwise be heard—primarily because governments can no longer unilaterally control or even mediate the flow of information. While there is certainly some truth to this argument, we need to understand that the processes of globalization can be complex, highly uneven, and contradictory. For example, while the Internet has undoubtedly opened new possibilities for global democracy, it also provides an opportunity for the state to keep tabs on its citizens more effectively and less obviously than ever before, and at a much lower cost to boot. Just think, for example, how much easier it has become for states to gather, store, analyze, and instantly access information about millions of individual citizens. This point was made crystal clear when it was revealed, in June 2013, that the National Security Agency (NSA) has been engaged in a vast surveillance program that scooped up every call placed on the Verizon network (over a set period of time), and then subjected this metadata to patterned analysis. Granted, the targets of this operation were “foreign terrorists,” but the targets could have just as easily been U.S. citizens. The same applies to corporations, the most powerful of which are able to exert tremendous influence on the development of the Internet and information technology more generally. Indeed, as corporations become more geographically dispersed, the need for centralized, top-level control becomes even stronger (Sassen 1995). This means, in many cases, not less concentration of power, but more—and generally in fewer and fewer places. On this point, consider that, of the 82,000 TNCs I mentioned earlier, a mere 147 wield control over 40 percent of the economic value of all TNCs through a complicated web of ownership relations (Vitali, et al. 2011, p. 6).

 

 

Table 1.2. List of the Top 20 Corporate Power Holders

 

Rank

Economic Actor Name

Country (HQ location)

1

Barclays PLC

GB

2

Capital Group Companies

USA

3

FRM Corp

USA

4

AXA

FR

5

State Street Corporation

USA

6

JPMorgan Chase & Co

USA

7

Legal & General Group PLC

GB

8

Vanguard Group, Inc.

USA

9

UBS AG

CHINA

10

Merrill Lynch & Co

USA

11

Wellington Management LLP

USA

12

Deutsche Bank AG

GER

13

Franklin Resources, Inc.

USA

14

Credit Suisse Group

CHINA

15

Walton Enterprises LLC

USA

16

Bank of New York Mellon Corp

USA

17

Natixis

FR

18

Goldman Sachs Group

USA

19

T. Rowe Price Group

USA

20

Legg Mason

USA

 

Source: Glattfelder (2010, p. 200)

 

 

 

According to critics, it is not only the “usual suspects” that we need to worry about. In the era of globalization, another relatively new set of actors—international institutions and organizations—are beginning to exercise more and more influence over more and more peoples and societies, but without any of the responsibility typically attached to policymaking entities in democratic societies. On this point, Barnett and Finnemore, in their book Rules for the World: International Organizations in World Politics (2004), assert that international organizations, such as the International Monetary Fund, engage in an ironic process of spreading liberal norms around the world without any democratic oversight. They refer to this as “undemocratic liberalism.” Others have leveled the same charges against another increasingly important international organization, the World Trade Organization (WTO). Kapoor (2004), for example, claims the WTO primarily acts as a vehicle for forcing liberal economic practices on countries and peoples around the world, but without a meaningful degree of legitimacy and democratic accountability. These are complicated issues, which we will return to in later chapters. The key point to remember is simply this: there is nothing inherently democratizing in the process of globalization.

It is the complex and contradictory nature of the changes being brought about by globalization that compels us to go beyond the territorial and substantive boundaries of mainstream IPE. A simple, but not necessarily trivial step in this regard is to abandon the concept of international political economy and replace it with global political economy (GPE). The move from IPE to GPE does not mean, however, that we must also abandon the traditional concerns of international political economy. States still matter (a great deal), but so do societal actors. Increasingly, then, the activities and concerns of states constitute only part of a much larger, more complex picture. One prescient scholar, writing 15 years ago, summed up the issue nicely: “The structure of global political economy contains the ‘old’ international economy within a new framework which is based in the territory of states, but not necessarily ‘national’ in terms of purpose, organization, and benefit” (Tooze 1997, p. 221). Placing the international economy within the framework of the broader global economy should also move us beyond a second restrictive boundary of mainstream IPE: the state-market dichotomy.

 

Political Economy and the State-Market Dichotomy

            Earlier I pointed out that conventional IPE textbooks tend to treat the market (economics) and state (politics) as separate entities, each operating according to its own, largely independent logic. While not entirely unjustified, the separation of the market and the state into mutually exclusive zones has always been problematic. One reason for this is clear: a market economy cannot exist, much less operate, without some kind of political order. This is not a new observation, nor is it one with which many (political) economists, even neoclassical economists, would disagree. There is, however, a great deal of disagreement over exactly what kind of political order is needed. Some take a minimalist view: the best political order is one in which the state only provides the legal-institutional framework for enforcing contracts and protecting private property (this is a view with which most neoclassical economists would agree). Others are convinced that the most appropriate political order is one in which the state plays an active and direct role in a much wider range of economic activity. Rather than discuss, in detail, the full range of different perspectives on this issue, it might be better to concentrate on just one. In this regard, I would like to introduce you to Karl Polanyi, who I mentioned earlier and whose work offers a useful perspective for understanding the state-market relationship.

 

Polanyi and the State-Market Dichotomy

            Almost 70 years ago, Polanyi wrote about the inextricable connection between the emergence and subsequent development of the market economy and the modern state. In one of his most important works, The Great Transformation, Polanyi explains how the “road to the free market was opened and kept open by an enormous increase in continuous, centrally organized and controlled interventionism [on the part of the state]” (1944, p. 140). This was necessary because the market economy, as we know it, required a very unnatural action—i.e., turning land, money, and especially labor into commodities, or things to be bought and sold. To accomplish this required a great deal of political power, exercised primarily through the state. For Polanyi, it is not difficult to see why this was so. Prior to the advent of the market economy, labor and land were decidedly not commodities; rather, they were “no other than the human beings themselves of which every society consists and the natural surroundings in which it exists” (ibid., p. 71). To turn them into commodities required an intentional and sometimes highly coercive effort, which only the state was capable of carrying out.

Once created, the so-called free market continued to rely on the exercise of state power. As Polanyi put it, “Just as, contrary to expectation, the invention of labor-saving machinery had not diminished but actually increased the uses of human labor, the introduction of free markets, far from doing away with the need for control, regulation, and intervention, enormously increased their range … even those … whose whole philosophy demanded the restriction of state activities, could not but entrust the self-same state with new powers, organs, and instruments required for the establishment of laissez-faire” (ibid., p. 140). In other words, the expansion of control, regulation, and intervention is an inevitable outcome of a “free” market. This is a crucial insight, and one that allows us an even stronger understanding of the mutually constitutive—or dialectical—relationship of the state and the market. Nor is it, in retrospect, a particularly controversial view. Nonetheless, it was not a generally accepted proposition when Polanyi first wrote about it, and still today, among many neoclassical economists and popular pundits, there is a strong conviction that markets can and should be kept isolated from the meddling of states. This conviction, it is important to point out, is based on the assumption that, for the market to operate efficiently, it must stand apart from the state. As I have already discussed, though, even in the mainstream, there are some economists who implicitly recognize that this separation is problematic. Among scholars of international political economy, more importantly, it is fair to say that Polanyi’s views are generally accepted (although there certainly may be disagreement on specific points). In other words, there is an understanding that markets and states are interdependent, or mutually constitutive: each depends on the other, and therefore they cannot be analytically separated. This is, in an important way, a starting point of IPE.

Polanyi also examined the relationship among the market, state, and society. He argued that the state continuously plays a critical role of mediating between the market and society (on this point, it is important to note that Polanyi indirectly challenged the older Marxist view that saw states as tools of the dominant class; indeed, while Polanyi drew from Marxist analysis, his own work opposed Marxism on important issues [see Block 2003]). That is, the state helps to establish and maintain the framework within which market activity takes place, but it also provides social protection to society from the inevitable “destruction” wrought by market forces. The concrete result of state intervention between the market and society has been a range of political orders or arrangements. Some of these we call capitalist, others we call socialist or communist. And, of course, there are versions in between, including the modern welfare state, of which the United States is just one example. No one of these arrangements, I should emphasize, can be said to be necessarily superior to another. This is because various political orders represent, at base, a mixture of different values, such as efficiency, equity, security, justice, and so on. Which value is given the highest priority and which is given the lowest reflects decisions made by individual societies, albeit within the context of a broader international/transnational system. Thus, to say, for example, that economic efficiency is more important than any other value is not to make an objective or scientific statement that must apply to all cultures and societies. Rather, it is to make a subjective or normative judgment, which typically reflects how the tension between the market and society has played out over time. Overall, then, Polanyi’s framework allows us to see that even a very narrow conception of politics as state action cannot be seen as standing in opposition to the market, or economics. Politics, including the exercise of state power, is fundamental to the market. In this sense, politics is economics. Students of IPE should keep this point firmly in mind.

 

International/Global Political Economy Defined (Finally!)

            In light of the rather lengthy discussion above, let us now return to the basic question: what is international, or global, political economy? In the end, the definition I use is simple. International political economy is an area of study. As an area of study, it is concerned with, as Susan Strange (a prominent IPE scholar) puts it, “the social, political, and economic arrangements affecting the global systems of production, exchange and distribution, and the mix of values reflected therein” (emphasis added; 1994, p. 18). This definition has the advantage of expanding—rather than limiting—the range of questions, concerns, and issues considered relevant to the study of international political economy, whether at the local, national, international, or global level (although not everyone would consider this an advantage). Moreover, it does not lead us to think that any one arrangement or set of values is superior to another; nor does it suggest that certain relationships or dynamics, such as tension or conflict between states and markets, or between opposing social classes, should or must be the focus of study. Similarly, it does not force us to view the world through a particular set of (theoretical) lenses. In short, Susan Strange’s definition encourages us to look at the complex reality of international political economy in an open manner. (Note: to avoid confusion, I will use the term IPE throughout this book, even though I prefer the alternative, GPE, or global political economy.)

            Another important advantage of Professor Strange’s definition is that it encourages us to think critically about the global or world economy. On first thought, it may not be apparent to you why this is so. If anything, you might feel just the opposite. The reason, however, is fairly simple: the definition we’ve chosen forces us to ask questions about what and who matters in the world economy, and why. It also pushes us to question many of the basic assumptions and values that underlie dominant and alternative perspectives of IPE. This occurs whenever we ask questions such as: are states still the dominant players in the world economy? To what extent have states lost control of the economic and political activity within their borders? What impact, if any, is the globalization of production, finance, and ideology likely to have on the world? How has globalization transformed relationships of power in the world? Where does the line between the domestic and the international, or between the economic and the political, lie? What is the relationship between democracy and capitalism? Are social justice, political equality, and human rights compatible with the “free” market? Not only do these and many other important questions flow from the definition given to us by Susan Strange, but also, the answers are far from obvious. By asking such questions and developing our answers to them we are, of course, engaging in a highly critical and evaluative process.

 

The Significance of Power

            Thus far, we have covered a number of important definitional issues and posed some key questions. But there is another question we must address in any study of international political economy. In political science, the study of politics revolves around the issue of power. Power is also a central concept in IPE. Yet, in many introductory textbooks on IPE there is, curiously, very little discussion about just what power is. Many writers seem to take for granted that power is an unproblematic, even self-evident concept. If pressed for a more formal conceptualization, however, most might agree with Robert Dahl’s oft-quoted definition, which asserts that power is the ability of actor A to get actor B to do something he or she would not otherwise do (Dahl 1957). Certainly, this way of looking at power has merit. Sending in thousands of heavily armed troops to keep workers from blocking access to a factory, for instance, is an exercise of (coercive) power whereby A (the state) gets B (workers) to do something they don’t want to do. Conversely, when workers are successful, they can force their company to increase wages, provide more benefits, or otherwise improve the conditions of work—all actions that the company would otherwise not have taken. You can probably think of dozens of similar examples that occur on a regular basis. This type of coercive, or interventional, power is clearly important. But it is hardly the case that most—or even a significant fraction—of what happens everyday in the political economy can be attributed to such direct applications of force or coercion by one actor against another. Most activity in the world political economy, instead, occurs as part of a process wherein power is exercised in a far less direct or interventional manner.

Text Box: Figure 1.7. A quote from Chairman Mao Zedong (The Little Red Book, 1964)

 
“Every Communist must grasp the truth: Political power grows out of the barrel of a gun.”
Pictured: From ARMOR magazine: The Chinese Type 98 Main Battle Tank
This image is a work of a U.S. military or Department of Defense employee, taken or made during the course of an employee's official duties. As a work of the U.S. federal government, the image is in the public domain.

Thus, to understand power we need to begin by ridding ourselves of the idea that power is the same as brute force, or, as Mao Zedong put it, that it only “grows out of the barrel of a gun.” Before we consider other ways to look at power, however, let us return to the claim I made above—namely, that excluding power from analyses of the international political economy is a fundamental problem. Why is this the case? That is, why is a firm understanding of power essential to the study of IPE (or GPE)? Part of the reason for this is, I hope, already apparent to you: to the extent that markets play (or do not play) a dominant role in the economic life of a country or system of countries, they do so as a consequence of a political process. In this process, it is the distribution of power in society that determines, to a very large extent, the rules and values that govern economic and social relations. Power (or a particular structure of power), in this sense, is required to create and sustain the framework within which economic activity takes place. An efficient and productive market system, in particular, cannot exist where private property rights are not respected, where contracts cannot be enforced, or where domestic security is weak or nonexistent. Yet, protecting property rights, enforcing contracts, and providing security require a great deal of power, which—it is important to emphasize—must be exercised by a nonmarket actor like the state. To better appreciate this point, consider what happens to societies racked by social and political upheaval. In Somalia, to cite one of the most disturbing examples, orderly market activity is hardly possible when there is no centralized and legitimate political authority capable of governing the entire country.

The rather sorry condition of Russian capitalism in the decade following the collapse of Soviet communism presents a less extreme example. In that case, the financial and political power of the once-feared Russian (Soviet) state proved insufficient to create an orderly and effective framework for a smooth transition to capitalism. Power, instead, rested in the hands of a corrupt oligarchy, who essentially wrote their own rules—rules that were designed to funnel huge sums of money into their hands at the expense of the larger economy and the rest of Russian society. When Vladimir Putin took office in 1999, however, things began to change very quickly, as he reasserted state control over important aspects of the Russian economy. Indeed, the Russian economy got back on relatively firm footing and did exceptionally well during his eight years as president, from 2000 to 2008 (see figure 1.8). This is not to say that Putin single-handedly solved Russia’s economic problems (he did not); rather, it is to emphasize the political aspects of capitalist development—unless, of course, the resurgence of Russia’s economy and the rise of Putin were entirely coincidental, which is a possibility.

 

Figure 1.8. Russian GDP Growth Rates, 1990–2011

 

 

Chart generated by Google based on data from the World Bank: http://www.google.com/publicdata/.

 

 

Most economists would accept the fact that power is required to create and sustain the general framework within which economic activity takes place. Yet they might also argue that, for capitalist markets in particular, this power must be exercised in a neutral manner. Once the framework for market activity is created, in other words, all actors should have an equal chance to compete and flourish—if, that is, the highest level of efficiency is to be achieved. Thus, power becomes largely irrelevant in terms of understanding what goes on within well-functioning markets, since everyone is equally empowered. This view, however, ignores two critical issues. First, power is not just needed to create a general framework, but (as I have emphasized several times already) is also needed to sustain that framework. Second, power is never equally distributed. In a political/economic system, power is typically distributed in a skewed—often extremely skewed—manner. To ignore power, then, is to ignore a particularly important aspect of reality.


Sources of Power

Text Box:  
Figure 1.9. Winston Churchill in 1916 holding a giant pen: “After all, some say the pen is mightier than the sword.”
Source: F. H. Townsend in Punch. This image is in the public domain because its copyright has expired.
            Before moving on, let us make one more point related to the Russian case, which reinforces a key argument in this section—that power is not a simple matter of who has the most guns. If it were, the Russian state (with control of the military) should easily have been able to put a stop to the corruption that, in the eyes of one prominent American expert, had “poisoned the Russian political process … [and] undermined the Russian fiscal system” (Sachs 1999, p. 31) prior to 2000. The Russian case suggests, in other words, that power has multiple sources, of which the control over the means of violence (or force) in society is but one. This raises an obvious question: what are other sources of power and how significant are they in relation to one another? Think about this question before you continue reading: again, what are the sources of power in an economy and a society? On this question, most of us would concede that wealth is clearly another source of power in society. But is wealth always trumped by military force? If not, under what conditions is wealth a more significant source of power? Many of us have also heard the saying, “The pen is mightier than the sword,” which encapsulates the rather bold claim that ideas are stronger than armies. Can this really be true? Do ideas—ideology or knowledge—constitute a source of power equal, or at least comparable, to military force (or wealth)? Consider, for example, the idea of nationalism or national identity. This idea, which Lind (1994) described as the “world’s most powerful force” (p. 87), should not be underestimated. Many have argued that its binding power is largely responsible for both the stability and instability of the modern state system, and is the force that makes large-scale war possible. After all, why else would ordinary citizens risk their lives to fight wars from which they have little to gain and everything to lose?

Still, there are no easy answers to any of these questions, or to the questions I posed above. But one thing is clear: power is not one-dimensional. This is a simple yet crucial point, because the study of political economy must not only pay serious attention to the importance of power itself, but to the many different aspects or kinds of power as well. My intention in the last part of this chapter (and throughout this book) is to do just that. The primary focus, however, will be on the distinction between coercive power and structural power. Both, as we will see, are important, but structural power (as I suggested above) has far more day-to-day relevance to the world.

 

Structural and Coercive Power

            When someone holds a gun to your head and demands you give him all your money, this is coercive power. The United States under George Bush undertaking a massive military campaign to overthrow Saddam Hussein in 2003 is also an example of coercive power. But when Iranians, Cubans, or North Koreans (the people or the governments) conduct financial transactions in U.S. dollars (both domestically and internationally), or when workers agree to do dangerous and difficult jobs for little pay, are these also purely reflections of coercive power? My view is that they are not. The latter two examples reflect structural power. One of the key differences between coercive and structural power may already be apparent: the exercise of coercive power reflects an interventional, clear-cut cause-and-effect relationship, wherein the intervention of the more powerful agent directly causes the weaker agent to do something he or she would not have otherwise done. Structural power, by contrast, is not easily reduced to such a simple equation of force. For example, workers who agree to do dangerous and difficult jobs for little pay do so because they have few other options, not because they are directly forced to do so (at least in democratic countries). Iran, Cuba, and North Korea, to use our other example, do not use U.S. dollars because the American government (or anyone else) forces them to; rather, these ostensibly anti-American countries use dollars because the U.S. dollar is the primary global currency. The structure of the international financial system, in other words, creates a framework (and a financial hierarchy) that strongly influences and/or limits the choices available to most actors. In particular, those who occupy peripheral positions within this structure are subject to rules, values, and practices over which they have little to no control. They agree to abide by the rules of the system because failing to do so is very costly. Occasionally, dominant actors in this structure will attempt to use their advantageous positions in an interventional manner, but this is not common.

Those who occupy central positions within the structure of international finance, on the other hand, have the power to write the rules (to some extent) or to define the framework itself—usually in ways that put them in an advantageous position. This has certainly been true in the case of the United States, which played a key role in shaping the international financial system following World War II via the Bretton Woods system. (I will discuss the Bretton Woods system in much greater depth in subsequent chapters.) It is important to understand, however, that structural power is not just a broader, more generalized version of coercive power. To see why this is so, consider the following three related points. First, we need to recognize that once a framework or structure is created, all actors—from the most powerful to the weakest—become subject to the same system of constraints and opportunities (albeit on different terms). In the international financial system, this might mean that a “weak” currency can become a source of strength. Consider, in this regard, China: throughout the 1990s and into the first decade of the 2000s, Chinese authorities intentionally worked to keep their currency, the renminbi, weak relative to the dollar. This helped to spur China’s extraordinarily fast growth in exports (a weaker currency means that Chinese products have a competitive advantage in world trade, since they are cheaper than would otherwise be the case). In other situations, monetary policy can be transformed into an exercise in political symbolism, and support “of the national currency may be promoted as a glorious stand on behalf of the imagined community—the ultimate expression of amor patriae” (Cohen 1998, p. 121). Conversely, a strong currency may become a source of weakness for governments, particularly if authorities attempt to preserve an international role for a currency whose popularity has begun to fade (ibid., p. 122). This happened to Britain after World War II. In the future, it may well happen to the United States.

Second, we need to understand that structural power is based on a network of (historically constituted) relationships that extend well beyond the interaction of two individual actors. This may sound abstruse, but it is really not. Consider, for example, the relationship between a student and a teacher. Teachers have power over students not because teachers are necessarily smarter, stronger, or wiser than their students. And it is certainly not because teachers are richer. Rather, a teacher has power over a student because he or she can assign a grade that others—such as a parent, an honor society, a law or medical school Text Box: Figure 1.10. The Dimensions of Structural Power
 
This figure shows the four dimensions of structural power discussed by Susan Strange. Different actors will have varying degrees of power, from very little to preponderant, in the different dimensions, but it is unlikely that a single actor can be dominant in all four. The four dimensions are separate, but inter-related.
Source. Image created by author, but is based on illustration in Strange (1994), p. 27. 
admissions committee, a potential employer—will use as a basis for determining the “quality” (and sometimes fate) of the student. A student’s well-being, in other words, “is affected by the grade only through the mediation of human beings [or institutions] situated outside the classroom, who use the grade as a sign that results in their administering ‘harm’ [or benefit] to the student—for example, by denying him access to the opportunity to further his education” (Wartenberg 1990, p. 145). In this situation, power is exercised not by the teacher per se, but by a range of external actors, who, in turn, are also part of a larger framework of action. This aspect of structural power also helps us understand its context-dependent nature. In the case of the student-teacher relationship, we can easily see the significance of the broader context: the power of the teacher, for example, will necessarily erode if the grade no longer functions as a means of access to a decent job (which may help to explain why discipline is a major problem in many of the poorest urban schools). In the international political economy, to put this issue very simply, this means that an exclusive focus on dyadic relations (e.g., the United States–China relationship) will not tell us all we need to know. We need to evaluate the relationship in terms of the broader structures and institutions of the global economy.

Third, we must recognize that structural power is reciprocal. This means that, in any relationship of power, both parties have a degree of power no matter how wide the disparity may seem. Again, this may seem an abstruse or perhaps trite point. But it is a crucial one, for it tells us that power is never absolute—that there are always structural limits to power. This suggests, in turn, that power relationships are rarely, if ever fixed. We can see this in the constantly shifting relationships between capital and the state, between capitalists and workers, and between rich and poor countries. Understanding the structural limits of power is important if we want to understand, first, how and why things change in the international political economy, and second, what the possibilities for change are. Indeed, without understanding the reciprocal nature of structural power, it would be hard to explain how or why change in the political economy ever takes place. After all, if those who lack power also lack the capacity to challenge those with power, how can unequal relations of power change, once established?

Taken together, these three aspects of structural power can help us develop a deeper, more realistic understanding of international or global political economy. But these are not the only aspects of structural power with which we should be concerned. Susan Strange argues that structural power should be separated into four distinguishable but integrally related structures: security, production, finance, and knowledge. Each of these structures, Strange notes, highlights critical aspects of power, which are generally ignored or glossed over in more conventional analyses (especially those that focus on coercive power). Yet, according to Strange, each is no more than a statement of common sense. The security structure, for instance, is simply the framework of power that provides protection to human beings from both natural and man-made threats. Those who provide this protection (or security) acquire a certain kind of power that lets them determine, and perhaps limit, the range of choices or options available to others (Strange 1994, p. 45). It is here that states tend to dominate, especially in terms of providing security against external threats.

The production structure includes all the arrangements that determine what is produced, by whom, by what methods, and on what terms. Those who control or dominate the production structure clearly occupy a position of power in any society, in part because the production structure is the primary means of creating value and wealth. The finance structure determines who has access to money, how, and on what terms. Money itself, however, is not critical; rather, it is the ability to control and create credit that really counts. As Strange puts it, “whoever can so gain the confidence of others in their ability to create credit will control a capitalist—or indeed a socialist—economy” (p. 30). The knowledge structure—perhaps the most overlooked and underrated source of power—“determines what knowledge is discovered, how it is stored, and who communicates it by what means to whom and on what terms” (p. 121). To appreciate the significance of the knowledge structure, Strange points to the example of the Catholic Church in medieval Christendom: the extraordinary power of the church was, first and foremost, a reflection of its ability to dominate the knowledge structure—to establish itself as the only legitimate source of moral and spiritual knowledge. A contemporary example is the debate surrounding climate change and global warming. The scientific community is clearly responsible for generating knowledge about climate change, but there has been an intense struggle among a variety of both state and nonstate actors over how that knowledge is interpreted, communicated, and constructed. How this struggle plays out will have immense implications for arguably one of the most important issues facing the world today.

 

Text Box:  
Figure 1.11. Jamie Dimon, CEO of JPMorgan Chase
Source: Financial Times. Licensed under the Creative Commons Attribution.

These four structures of power, according to Strange, are inextricably connected, and no one dimension is inherently or necessarily more important than any of the others. “Each is supported, joined to and held up by the other three” (1994, p. 26). This reinforces a point I made earlier—namely, that power grows not merely out of the barrel of a gun, but also (for example) from the factories that manufacture guns, from the technical knowledge needed to produce weapons, from a belief system that legitimizes mass violence against others, from a financial system that provides credit to create the weapons industry, and so on. Understanding this multifaceted nature of power is critical for anyone who wants to make sense of the international or global political economy. Consider, on this point, the power of transnational corporations: one cannot explain the increasing significance and autonomy of TNCs—and the concomitant erosion of state sovereignty—without paying serious attention to changes in the underlying structures of power. For example, it is now almost undeniable that changes in the system of global finance have seriously eroded the capacity of states to control credit—the lifeblood of any capitalist enterprise. As this power erodes, state authority diminishes, which leaves the door open for those who are better positioned to deal with transnationally mobile funds—e.g., TNCs, international banks, fund managers, and even a few wealthy asset holders (such as George Soros)—to exert greater control, not just on how funds are allocated, but also over government policy that deals with finance. On this point, consider the 2012 Senate Banking Committee hearing involving Jamie Dimon, CEO of JPMorgan Chase. The hearing was meant to examine the reasons behind a multibillion-dollar loss tied to the bank’s trading of credit derivatives. As one observer put it, “The senators should have interrogated Dimon about his role in moving toward that reckless gambling strategy,” which posed a threat to the entire financial system. “Instead, they mostly cowered and cringed and sat mute with thumbs in their mouths, while Dimon evaded, patted himself on the back, and blew the whole derivative losses episode off as an irrelevant accident caused by moron subordinates” (Tiabbi 2012). An even more telling example of financial power took place that same year, when banking giant HSBC admitted to violations covering $200 trillion worth of transactions involving Mexican and Columbian drug cartels—groups that were allegedly tied to terrorist organizations. Surely, this would warrant a major sanction by the world’s most powerful country, the United States. Instead, HSBC was fined a paltry $4.2 billion. No HSBC executives were even charged with criminal wrongdoing, because, in the view of U.S. Attorney General Eric Holder, the bank and its executives were simply “too big to jail.”

Changes in corporate power, in sum, cannot properly be understood without considering the financial structure of power. The issue, of course, is far more complicated than I have suggested here. Certainly, we need to examine all the structures of power, and the different aspects of state-business relations in the global political economy, to achieve an adequate understanding. The basic lesson, however, should be clear: questions of structural power must be addressed.

A second example I would like to discuss is perhaps the most problematic. Many of you probably agree that large corporations are capable of exercising structural power. But what about ordinary citizens working together in grassroots organizations, unions, or broad-based social movements? What is the source, if any, of their power? Can such groups even hope to have a meaningful impact on the world economy? For example, do the groups who challenge neoliberal globalization—such as those who participated in protests against the World Trade Organization in Seattle in December 1999, the 2011 G20 Summit in Cannes (France), or those in the 2011–12 Occupy Wall Street movement—have any chance of succeeding? Or are they simply wasting their time? The easy answer, of course, is that they are wasting their time. States, despite an erosion of sovereignty, are still extremely powerful and coercive institutions. Corporations, if anything, are getting stronger. Ordinary citizens, on the other hand, have neither armies nor wealth; they seem powerless. We have learned, however, that social power is invariably reciprocal and context dependent. The reciprocal nature of social power, for instance, tells us that states and corporations, to a significant

Text Box: Figure 1.12. A Lone Protestor—The WTO Protests in 1999 (Seattle)

 
This shot was taken by J. Narrin during the protests in Seattle, Washington, in November 1999. 
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Documentation License and under the Creative Commons Attribution 3.0.
degree, depend upon the perception or belief that their activities are legitimate. In this sense, global capitalism survives because a large majority of the world’s population believes it serves their interests. Take away or undermine this belief and the system itself is threatened. This helps us understand, I might add, why capitalism is not just an economic system, but (according to some critics) a cultural system as well. Capitalism (or neoliberalism), in other words, is an ideology that convinces people that it is the only rational—even conceivable—way of organizing an economy and society. This is the basic message underlying the writings of Antonio Gramsci, who coined the phrase cultural hegemony to describe how the ruling class is able to manipulate a society’s culture so that the values, norms, and interests of the ruling class become the values, norms, and interests of the society as a whole. At a more concrete level, however, it is important to recognize that societal actors have many tools at their disposal—tools that have become more effective in the era of globalization. These include the type of mass protests and demonstrations I alluded to above (i.e., the 1999 protests against the WTO in Seattle, etc.), as well as internet-based campaigns, consumer boycotts, and the like.

In a similar vein, the context-dependent nature of social power tells us to look at broader forces that exist outside, say, the relationship between citizens and corporations. What effect, for example, will the spread of political liberalism (e.g., democracy) and human rights across the globe have on the capacity of citizens to exercise power? What of the seeming dispersion of control over access to information via the Internet? What role can transnational institutions, such as the Catholic Church, or transnational religions, such as Islam, play? I do not want to try answering any of these questions now; rather, I would like you just to think about the ways in which a structural analysis of power compels us to address hitherto hidden or obscured aspects of important issues. This is perhaps the best way to learn about the complex and sometimes confusing world of international or global political economy.

 

Chapter 1: Conclusion

            We have covered a lot of ground in this chapter, which has likely left you more confused than enlightened. To a certain extent, this was my intention. That is, one of the purposes of this introductory chapter was to introduce you to the increasingly complex world of international or global political economy. At the same time, there are a number of basic points that I want you to keep firmly in mind as you read the remainder of this book. First, in IPE, there is no definitive approach or theory. If anything, disagreements and debates define the field. This does not mean, however, that IPE is a chaotic mess. It is not. There are, as we will see in the following two chapters, a number of extremely well-developed, coherent, and insightful perspectives around which the field as a whole revolves. Second, just as there is no definitive theory, there is no common agreement on how to define what IPE includes—at least beyond the traditional concerns of international trade, finance, and production. Third, globalization has introduced important, even centrally important, elements of novelty into the international-global political economy. It is, in particular, changing relations of power, bringing in a range of additional actors (societal or nonstate), and altering global dynamics through technological innovation. Fourth, regardless of theoretical differences and debates, the study of IPE requires that we take questions of power—what power is, who or what has power, how power is distributed and exercised—extremely seriously. Power is not external to the world economy; it is part and parcel of the world economy, and of the social world as a whole.


Chapter 2

 

 

 

 

Foundational Theories of IPE: An Unconventional Introduction to
Mercantilism, Liberalism, and Marxism

 

 

 

The Three Major Perspectives of IPE: Still Going Strong?

            IPE, as we have already seen, is a contentious field. This does not mean, however, that there is a complete lack of agreement among IPE scholars. In fact, for a long time, research in IPE has been broadly divided into three major schools, or perspectives, which we can classify as mercantilist, liberal, and Marxist. Each of these perspectives has been around for a long time. Mercantilism is the oldest of the three, dating back as early as the 16th century (perhaps even earlier). As a coherent politico-economic theory, however, many scholars point to Friedrich List (1789–1846) as the intellectual father of mercantilist thought. The National System of Political Economy (first published in 1841) is List’s best-known work on the subject. List, it is important to note, mounted his defense of mercantilism as a response to classical economics and, more specifically, to the writings of Adam Smith (1723–1790), whose An Inquiry into the Nature and Causes of the Wealth of Nations (or more simply, Wealth of Nations), published in 1776, quickly became one of the basic treatises of the liberal perspective. Marxism, then, is the youngest of the three. Karl Marx published his most famous work, Das Capital, in 1867 (later, his colleague Friedrich Engels used Marx’s notes to publish two additional volumes, in 1885 and 1894). Marx, too, wrote Das Capital partly as a critique of classical economics, but also as a larger examination of the social and historical forces that shape human society.

Text Box: Figure 2.1. The “Big Three”: Friedrich List, Adam Smith, and Karl Marx

 

All images are in the public domain.

 

The original mercantilists believed that a country’s economic prosperity came from its stocks of precious metals, and that the best way to increase these stocks was to limit imports through tariffs and other protectionist policies, while maximizing exports, thus creating a trade surplus. Despite its relatively old beginnings, mercantilism is far from a moribund tradition. Indeed, mercantilism enjoyed a strong revival in the latter part of the 20th century, due in no small measure to Japan’s rapid ascent to the status of economic superpower in the few decades following World War II. In fact, Japan’s rate of economic growth in the early postwar period was unprecedented: no country had ever achieved so much economic progress in so little time. Consider, on this point, that Japan’s per capita income in 1950 was less than half of the Western European average, and yet, by 1970 Japan had virtually caught up. It did this by quintupling per capita income, while Western Europe only doubled its income in the same period of time (see table 2.1, “Comparative Per Capita GDP Figures”). The key point is this: many scholars argue that the Japanese state—practicing an updated form of mercantilism or, most simply, neo-mercantilism—was primarily responsible for the country’s stunning economic success.

The neo in neo-mercantilism highlights a number of distinctions from the older version. First, the emphasis on holding precious metals was replaced by holding foreign exchange reserves (usually in the form of U.S. dollars). Second, the newer form of mercantilism is much more strongly concerned with developing a country’s domestic manufacturing capacity; this led to a strong emphasis on infant industry protection. Third, in neo-mercantilism, especially as it developed in the 20th century, states were expected to play a much more sophisticated and interventionist role in the national economy. For example, instead of just engaging in protectionism, states were charged with identifying and helping to develop strategic and targeted industries (i.e., industries considered vital to long-term economic growth) through a variety of means, including tax policy, subsidization, banking regulation, labor control, and interest-rate management. States also had to fulfill a disciplinary role in the domestic economy—to essentially take the place of the invisible hand of the market by ensuring adequate levels of competition. The Japanese state fulfilled these roles, some argue, almost perfectly. Japan, moreover, was not alone: South Korea, Taiwan, Singapore, and most recently China, have closely followed Japan’s state-directed lead to achieve remarkable economic growth. The case of China is particularly instructive in this view, since the Chinese Communist Party (CCP) still governs the country: in China, in short, we have a seeming paradox whereby a highly interventionist and authoritarian political party is presiding over one of the most dynamic capitalist economies of the past 20 years (I will have much more to say about the case of China later in this book). The proof of the continued relevancy of mercantilism, therefore, is in the pudding. (Few researchers actually use the terms mercantilism or neo-mercantilism to describe their work. Instead, they have adopted less politically loaded terms, one of the most prominent of which is the statist perspective; a more specific term, typically used to refer to the East Asian economies (especially Japan, South Korea, and Taiwan), is the developmental state approach.)

Table 2.1. Comparative Per Capita GDP Figures (in international dollars*), Selected Countries and Years

 

1950

1970

1980

1990

2000

2008

Mexico

2,365

4,320

6,320

6,085

7,275

7,979

Japan

1,921

9,714

13,428

18,789

20,738

22,816

Sri Lanka

1,253

1,499

1,830

2,424

3,597

4,895

Ghana

1,122

1,424

1,157

1,062

1,265

1,650

Philippines

1,070

1,764

2,376

2,197

2,377

2,926

Taiwan

916

2,537

5,260

9,938

16,872

20,926

Egypt

910

1,254

2,069

2,523

2,936

3,725

South Korea

854

2,167

4,114

8,704

14,375

19,614

Western Europe (avg.)

4,569

10,169

13,154

15,908

19,176

21,672

Key Points. This table shows the rapid economic ascendance of three East Asian economies, Japan, Taiwan, and South Korea, since 1950. Note that Taiwan and South Korea started as two of the poorest countries in the world, but by the end of 2008 had essentially caught up with Western Europe in terms of per capita income. Japan accelerated more quickly: by 1970, Japan had already achieved the same basic income level as Western Europe. At the same time, other countries have tended to fall further behind, relatively speaking. Consider Mexico: in 1950, its per capita income was about half of per capita income in Western Europe, but in 2008, it had dropped to about one-third. In addition, Mexico, which was richer than all the East Asian countries in 1950, was substantially poorer than all three as early as 1990.

 

* The international dollar, also known as the Geary-Khamis dollar, is a hypothetical unit of currency that has the same purchasing power that the U.S. dollar had in the United States at a given point in time. In the data above, 1990 is used as the benchmark year for comparisons that run from 1950 to 2008. While the international dollar is not widely used, for per capita GDP comparisons across a range of countries over a relatively long period of time, it is a useful measure. The World Bank and the International Monetary Fund do use the international dollar in some of their published statistics.

 

Source: Maddison (2008)

 

 

Text Box: Figure 2.2. What Is Communism?
 Many people—including scholars—assume that “communism” in the former Soviet Union was the same as Marxism. Yet, in classical Marxist thought, this is decidedly not the case. According to Marx, communism was a historical stage, but to reach this stage, it was necessary to first pass through capitalism. Why? Because communism required a firm economic or material foundation that only capitalism could provide. In this view, then, no society could skip the capitalist stage. Yet this is precisely what the Soviet Union (and China, Vietnam, Cuba, North Korea, etc.) tried to do. In so doing, however, they were doomed to failure. In this sense, we might say that Marx predicted the collapse of the Soviet Union before it was ever created, just as he predicted China’s turn toward capitalism. 

Image: The work (image) is not an object of copyright according to Part IV of Civil Code No. 230-FZ, Article 1258 of the Russian Federation.

            Marxism, too, is far from a dead tradition. Admittedly, this might sound strange given the collapse of the Soviet Union, and the apparent embrace of the market by most remaining socialist countries, including mainland China and Vietnam. Even North Korea has been forced to take tentative steps toward market reform. “Communism,” in this sense, has clearly failed, and this failure is supposed to have completely discredited Marxism and all its variants. (I put quotation marks around communism because, in classical Marxist theory, what existed in the former Soviet Union and other countries was not, and could not be, communism—a point that is discussed further in figure 2.2.) However, while it is certainly true that central planning in command economies has proven to be an utter disaster, it is not necessarily true that all or even most of the Marxist critique of capitalism has been negated by historical and contemporary realities. In fact,

just the opposite is the case, at least according to advocates of Marxism. Global and national income inequality, for example, remains extreme: according to an analysis by Ortiz and Cummins (2011), in 2007, the richest 20 percent of the world’s population controlled 83 percent of the world’s income, while the poorest 20 percent controlled just 1.0 percent. This differential actually represents an improvement from 1990 (when the respective figures were 87 percent and 0.8 percent respectively), but it is nonetheless indicative of, as the authors put it, “an incredibly unequal planet” (p. 11). Exploitation of labor, moreover, shows no sign of lessening; in fact, in parts of the world—e.g., Ghana and Côte d’Ivoire (also known as the Ivory Coast)—the problem of child labor and even child slave labor has become endemic. In 2011, for instance, it was estimated that more than 1.8 million children throughout West Africa were working in the cocoa industry (Hawksley 2011), and tens of thousands of them were forced to work without payment (Manzo 2009)—the very definition of slavery. At the same time, the global candy and chocolate industry, located primarily in wealthy Western countries, had revenues of $118 billion the same year, and it was predicted that with the “high level of value addition during the production process … the industry’s major players [are expected] to realize high profit margins and perform well in 2012” (IBIS World 2012).

Figure 2.3. Global Income Distribution by Population Quintiles in 1990, 2000, and 2007 (in constant 2000 U.S. dollars)

 

Each quintile (e.g., Q1) represents 20 percent of the observable world population. The figure includes all individuals for which data is available, from the poorest quintile in the Democratic Republic of Congo to the richest quintile in Luxembourg.

Source: Ortiz and Cummins (2011), p. 11.

 

The planet and its environment, Marxists will also point out, are being destroyed by rapacious corporations, and the capitalist system overall is becoming more and more unstable: the global financial crisis of 2008—a crisis that lasted four or five years—is certainly testament to this. This crisis, however, is simply another in a long line of increasingly more frequent financial and economic crises, including the Latin American debt crisis of the 1980s, the U.S. savings-and-loan crisis (1985), the U.S. stock market crash of 1987, Japan’s asset bubble collapse (1990), Black Wednesday in Europe (1992), the Mexican peso crisis (1994), the Asian financial crisis (1997), the Russian financial crisis (1998), the Argentine financial crisis (1999), and the dot-com crash (2000), among many others. Marxists tell us that all of these crises are cut from the same cloth. In particular, they all reflect the inherent instability and volatility of a global capitalist system that has become increasingly reliant on financial speculation for profitmaking. To be sure, some actors will always make huge sums of money from the speculative bubbles that finance capitalism produces, and this can create the illusion that everything still works. “But”, as Wallerstein (2008) succinctly put it, “speculative bubbles always burst, sooner or later.” In fact, they not only burst with unnerving regularity, they also emerge time and time again. The reason is clear: the traditional avenue for generating large-scale corporate profits is choked off by excessive production, investment, and competition; thus, financial speculation serves as one of the few roads, if not the only major road, still open to capital accumulation on a sufficient scale. Indeed, we can expect an acceleration of this trend, since the world’s productive capacity will continue to outpace its consumptive capacity. We will come back to this basic point later; suffice it to say for now that reports of Marxism’s death, to paraphrase Mark Twain, have been greatly exaggerated.

            Of the three perspectives, liberalism has had the strongest and most sustained following: it is the mainstream approach in economics, although not necessarily in IPE, and has been for quite some time. (An important note: the word liberalism here does not refer to, say, the progressive political orientation of the Democratic Party in the United States; instead, it refers to the classical principles of individual liberty and limited government.) As I suggested in chapter 1, there is no single liberal economic theory. Instead, there are a variety of theoretical positions, some of which can differ quite significantly from others, even with regard to some fairly basic assumptions. That said, there are core features on which most liberal economic analysts agree. In this regard, a good place to start is with the market, and more specifically the free market. A market, in the most general sense, is any place where the sellers of a particular good or service can meet with the buyers of that good or service to conduct an exchange or transaction. A free market refers to exactly the same arrangement, but is conditioned on voluntary and unrestricted exchanges. This includes trade between and among countries—i.e., international trade. This leads to a key assumption in the liberal view: (voluntary) exchanges in free markets, whether between individuals or between countries, generally result in mutual benefit. Rothbard (2006) put it this way: “Trade, or exchange, is engaged in precisely because both parties benefit; if they did not expect to gain, they would not agree to the exchange” (n.p.).          

            The beauty of the free market—and another basic area of agreement—is that it tends to operate in a largely self-regulating fashion. This means, in part, that while the free market can and does experience problems, the market process will automatically resolve these problems. Consider, on this point, a recurring phenomenon: throughout history, free markets have experienced regular ups and downs (i.e., periods of strong economic growth followed by periods of economic slowdown or recession). In the simplest terms, we can say that these so-called boom-and-bust cycles are caused by a temporary imbalance between supply and demand. One reason for this imbalance is overinvestment. The logic here is easy to understand: in a growing market, market actors will take advantage of new opportunities for profitmaking by ramping up investment to meet still-rising consumer demand. At some point, though, demand becomes sated (through the entry of more and more firms, through a change in consumer tastes, etc.), and a large number of companies will find themselves unable to sell their products. They go bankrupt, workers lose their jobs, and demand may be further weakened (as incomes decline). If this happens on a nationwide basis, a country’s entire economy may go into recession. If it happens on an international basis, the world economy may go into recession. For many liberals, the prescription for how to solve this problem is clear: do nothing. Or, rather, leave the market alone to self-correct. In the foregoing example, the process of self-correction is readily apparent: as companies go bankrupt and leave the market, supply dwindles and supply and demand come back into balance, or equilibrium. The strongest and most efficient producers continue to thrive, while others seek out new, more profitable opportunities elsewhere. Investment soon recovers, workers are hired back, overall demand increases, and the economy starts to grow again. The result is an economic boom. These boom periods more than make up for the economic loses suffered in the bust periods. Self-regulation, in short, keeps markets operating smoothly and at maximum efficiency over the long run.

To liberals, the proof of their point of view is also in the pudding. To see this, liberal economists and their supporters tell us to simply look around the world. The breakdown of communist rule in the Soviet Union and Eastern Europe, for example, clearly confirmed what many people already took for granted: the self-evident superiority of liberal economic principles,

 

Figure 2.4. Trade as a Percentage of World GDP

 

 

This chart shows a steady increase in world trade (measured as a percentage of gross domestic product) between 1960 and 2010. The dip in 2008–2009 is due to the global recession; by 2010 a strong recovery was already in evidence.

 

Chart generated by Google based on data from the World Bank: http://www.google.com/publicdata/.

 

 

meaning laissez-faire, comparative advantage, free trade, and competition. In addition, China’s embrace of market reform, beginning in 1979, followed by that country’s economic takeoff and ascendance—which we are witnessing right now—is just another example of the power of unleashed market forces (prior to 1979, China’s centrally planned economy had only succeeded in creating a slow-moving, industrially backward leviathan.) Moreover, while mercantilists (and neo-mercantilists) point to the success of Japan and other East Asian economies, liberals tell us that success was more mirage than reality. To see this, just look at Japan for the past twenty-plus years (since 1990). Its economy has been mired in a prolonged slump, with high levels of inefficiency, low productivity, and corruption—a wicked combination that has been summed up by the term crony capitalism. Even some individuals who once lauded Japan and the rest of East Asia’s neo-mercantilist success now argue that, for those countries to continue to prosper, their governments must learn to get out


Figure 2.5. Number of People Living on Less Than $1.25 (PPP*) a Day

 

        This figure shows that the total number of people living on less than $1.25 a day (the amount typically used to indicate severe poverty) has been slowly, but steadily declining since 1981. While a direct connection between increasing world trade and decreasing global poverty cannot be made, a strong case can be made that trade does contribute to poverty reduction (see, for example, Wiig, Tøndel, and Villanger 2007).

 

        * PPP stands for purchasing power parity. It is both a theory about exchange-rate determination and a tool used to make more accurate comparisons among countries. For further discussion of PPP, see Suranovic (2010) [His article is available on the Saylor.org site at http://www.saylor.org/site/wp-content/uploads/2012/10/Chapter-17.pdf]

 

            Source: Figure is copied directly from UN Department of Economic and Social Affairs (2009, p. 16).

 

 

of the way of private enterprise; what worked in the past, in other words, will no longer work today, since global competition has made it impossible for government bureaucrats to keep up with increasingly rapid changes in the world economy. More generally, we can easily see that the opening of borders to freer trade, despite recent economic problems, has increased economic growth and likely helped to decrease poverty on a global scale. There is much more evidence that liberals can point to; suffice it to say, for now, that the free market has not only proven, time and time again, to be the only rational basis for countries, both individually and collectively, to prosper, but it has also proven to be tremendously resilient and adaptable.

The Continuing Debate

How is it that all three of these contending perspectives continue to be relevant today? After all, one would think that after more than a century of debate, one perspective would have proven clearly superior to the others. But this has not happened. We already covered one of the major reasons for this in the previous chapter—namely, the nature of the subject itself. We cannot, therefore, expect to resolve the issue here. There are also other, less obvious, reasons, which will be addressed later. For now, though, it will be useful to adopt an admittedly unconventional approach, not only to help you get a better grasp of the differences among mercantilists, liberals, and Marxists, but also to give you a sense of the obstacles to finding the one “true” account of IPE. Specifically, I would like to engage you in a conversation involving three scholars representing the three main perspectives of IPE.

vvv

We catch the action in a dingy office as three middle-aged and somewhat gruff scholars debate how to organize their new think tank on international political economy. After hours of discussion, however, they can’t even agree on whether what they do is political or economic. In any case, the members of this contentious trio are: Friedrich the Mercantilist, Joanna the Liberal, and Karl the Marxist. In the midst of their discussion, in walks a bright-eyed but somewhat confused looking student.*

 

Karl: If you’re looking for the cafeteria, it’s outside and to the right. Now, get out, we’re busy.

Student: Oh, I’m sorry. I just had lunch. I’m really looking for the “Group of Three.” The student newspaper has been reporting that the G-3 is starting up a new think tank here on campus, and I need to ask them about some very important matters. You see, I’m planning to major in international political economy, but to be perfectly honest, I’m not even sure what that is. All I know is that it’s supposed to be important.

Karl: Yes, yes, we’re the G-3. But if you’re unsure of yourself, this isn’t the place for you. Except for me, my colleagues all suffer from a bit of befuddlement themselves. None of them seems to know what really matters.

Student: Well, if that’s the case, maybe I should leave. After all, if you can’t even agree among yourselves …

Friedrich: No, stay. Karl here tends to exaggerate and he’s a little too serious to boot. We may not see eye to eye on everything, but I can certainly agree with some of what he says, especially how we should be wary of the rich and powerful—who, we all know, won’t hesitate to crush the weak in order to protect their own self-interests. Even my esteemed liberal (or should I say neoliberal?) colleague agrees that self-interest is vitally important. Isn’t that right, Joanna?

Joanna: Don’t mock me, Fred. You know that I believe self-interest to be of paramount importance. But you also know that we cannot let the self-interest of government officials interfere with individual choice. Anyway, we’re wasting time here, and none of us really has time to waste, so let’s get down to brass tacks. Young man, what do you think we can do for you?

Student: Well, I was hoping you could answer a few questions about IPE. You know, help me get a better grasp of what the field is all about.

Friedrich: Well, well! That’s the reason we decided to create our new think tank to begin with. Go ahead—give us a shot.

(Before Karl and Joanna can disagree, the student begins to speak.)

 Student: Okay, here goes nothing. It seems that in international political economy there are a bunch of different theories out there. In fact, there seem to be at least three distinct schools of thought, which all of you obviously already know. Aside from the three major schools, there also seems to be a figurative avalanche of sub-schools, splinter groups, varying interpretations, and confusing debates that I guess make sense to those doing the talking, but which seem completely irrelevant to the average person.

Friedrich: Quite impressive, young man. You seem to know a little more than you first let on. Now, let’s see what I can do to set you straight …

Student (interrupting): Hold on a minute! That’s not all I have to say. I’ve also been told, or at least led to believe, that scholarship and research are supposed to lead to the truth of something. Well, here’s my million-dollar question: Where’s the truth? Or, should I ask, is there a truth in IPE? That is, is there one school of thought that I can look to in order to truly understand the relationship between politics and economics in world affairs? Or is it possible that all the schools simply provide small portions of a larger truth that exists out in the world someplace? In fact, perhaps all of these competing schools of thought are slowly moving toward the same version of the truth? Wasn’t it Aristotle who said, “Truth … is the work of everything intellectual” (cited in Ackrill 1987, p. 418)? You three don’t seem to work in the traditional sense of the word, so you must be intellectuals. Ergo, your work ought to be producing truth.

Karl: Of course, you’re right young man. At least I’m in the business of discovering the truth; as for the others, they purport to be working on the truth, but I’m not at all convinced. They seem much more concerned about justifying their own privileged positions in the world—keeping themselves in power with an ideology disguised as theory, as it were.

Joanna: Karl, you really must temper your comments. We’re certainly no less concerned with the truth than you. In fact, economic liberals, especially my neoclassical friends, have even developed a rigorous—and by rigorous, I mean highly mathematical—methodology practically guaranteed to find the truth … sooner or later.

Student: Excuse me, but this is exactly what I mean. Both of you claim to be searching for the truth, but at least from my rather rudimentary knowledge of your theories, what you say is completely different. My point is this: if one of your theories is true or correct, then the others must be incorrect. Or, if your theories are somehow parts of a larger puzzle, they are all still incorrect unless you can show the possibility of a synthesis or unification of the theories, which seems unlikely. 

Joanna (looking annoyed): So what is it that you want from us? It seems you already have things figured out. Besides, I’m getting tired of all this talk about the truth. As I said, in my work, I’m always searching for the truth, as you mean it. But, I’m much more interested in results. As I like to say: the proof is in the pudding, and it’s clear that free markets lead to prosperity, choice, and liberty. Why, just look at the world around you—the collapse of communism, the triumph of Anglo-American capitalism, the spread of democracy …

Student (interrupting): Wait a second, now. Before you go on, I’d like to answer your question. You asked what I want. What I want is simple: answers.

Joanna: What sort of answers, young man? And, mind you, don’t be impertinent.

Student: Sorry, ma’am. Basically, I want to know what these three theories, or perspectives, of IPE really are. Also how, and on what basis, do they differ? Moreover, I want to know how deep these differences go.

Friedrich: Ah, that will be easy. Mercantilism is …

Student (impertinently): Not so fast, my friend. After you cover the basic points, I want you to answer the bigger questions I mentioned earlier—namely, how each of your respective schools of thought relates to the truth. After that, I want you to tell me how and why the study of any of your theories might be important to me. More generally, I want you to tell me what relevance your work has to anything in this, or any, world.

[A long silence ensues as Joanna, Karl, and Friedrich, a little stupefied, exchange uneasy glances. Friedrich breaks the silence.]

Friedrich: Listen, my fellow IPErs, we cannot let this outsider—this whippersnapper—come here and scare us. In fact, if we back down, it would only embolden him more. He might even try to move in on our operation here—take us over, if you will.

Joanna: For once, Fred’s right. Besides, this sort of competitive challenge will only make us more efficient and productive thinkers in the long run.

Friedrich (standing up): Why, thank you, Joanna. That’s the first time you’ve agreed with me for as long as I can remember. Shall I be the first to speak?

Karl: Be our guest, Fred. You are, after all, the senior member here.

Text Box: Figure 2.6. A Dog-Eat-Dog World

 

For mercantilists, the world is a tough, unforgiving place, where only the strong survive. In this world, all states must use whatever tools they have to help them not only survive, but also prosper. Liberals forget that, in the real world, “free trade” is a luxury. 

Source. Permission is granted to copy this image under the terms of the GNU Free Documentation License.

Friedrich (now facing the student): Well, then. Let me start by saying that mercantilism is not so much a theory as it is a politico-economic strategy. Perhaps this is why our critics accuse of us of being unscientific and even amoral. Mercantilism is neither. We simply recognize that economic processes do not take place in a vacuum, but in an inherently unequal and power-ridden world. That said, let me highlight an important assumption of mercantilism. Namely, mercantilists, and I am not ashamed to apply that label to myself, view the modern state as the main player (or actor) in world affairs. This is primarily because we live in a dog-eat-dog, everyone-for-himself world; in such a world, a strong state is necessary to provide protection against external threats, be they economic, political, or military. Simply put, it’s a matter of self-help: in the real world, you can’t rely on anyone else to protect your interests, so you must do it yourself—if you don’t, you will almost certainly be subject to the whims and desires of your richer and more powerful neighbors, and nobody wants that, do they? From an economic standpoint more specifically, this means that states must also pursue a set of economic policies designed to maximize their own wealth, for, as we all know, wealth and power go together like bread and butter. I think there is no denying the truth of this statement. What do you say, my dear boy?

Student: Well, so far, it’s hard to disagree.

Text Box: Figure 2.7. Alexander Hamilton: An American Mercantilist
 

Hamilton, a seminal figure in early American history, was a strong advocate of mercantilist policies.

The image of Hamilton is in the public domain.
  


Friedrich (smiling): Exactly. But let me continue. Mercantilism, as should already be apparent, is a very practical—or, as I like to say, realistic—perspective. Indeed, our most important forebears were, first and foremost, men of action rather than cloistered intellectuals. Take for example one of the most famous of our ilk: the great American statesman Alexander Hamilton [1755–1804]. Hamilton, as you should know, was a steadfast advocate of a strong and active central government, one that had a duty to promote and protect the nation’s manufacturing industries. To promote American industry, Hamilton wanted to establish a healthy credit system in the United States—but he knew this would require strong support on the part of the federal government. Thus, Hamilton urged, among other things, the creation of a national bank to facilitate the expansion of both public and private credit, which he correctly understood as more useful than gold and silver. To protect American industry, Hamilton argued that the federal government must grant subsidies to domestic industries, promote internal improvements, and impose tariffs or duties on foreign products. In this sense, Hamilton was a “protectionist,” but this is not the nasty word that our liberal friends would have you believe. For what is a protectionist but someone who wishes to defend and promote the interests of his country and by extension, his people, his community, and his family? [Pointing to the student.] What can be wrong with this?

Student: Nothing that I can see.

Friedrich: My, my. You are an apt pupil, are you not? In any case, I should point out that Hamilton’s ideas not only won the day—particularly in his debates against Thomas Jefferson—but also gradually created the basis for American economic policy, both foreign and domestic, for the entire nineteenth century. Today, of course, the United States is unarguably the world’s economic powerhouse, and has been so for quite some time now. Would the U.S. have achieved that position were it not for the mercantile policies expounded upon by Hamilton? I think not! Nor is America the only example: Germany, Japan, South Korea, Taiwan, China. Mercantilists all, I say! [Friedrich begins pounding his fists on the table.] What more could you ask for in terms of relevance!

Joanna and Karl (in unison): Take it easy, Fred.

Friedrich (a bit chastened): Quite sorry. I will try to restrain my passion. Now where was I? Oh, yes. Let me say a few words about another famous mercantilist, Friedrich List. List is a somewhat disparaged soul, as are all mercantilists. Perhaps it is because he was not only unafraid to take on the intellectual establishment of his time, he was also an unabashed nationalist. Yet, it is the writings of men like List that often help us see the truth of the real world. This is most apparent in List’s ideas vis-à-vis liberalism, or “cosmopolitical” economy, as he liked to call it. For example, he argued that the so-called advantages of international trade and comparative advantage [see below for additional discussion], while fine in the abstract, neglect the cold realities of unequal power and wealth in the world. In this respect, it’s no accident that the Brits, back in the nineteenth century, were in favor of free trade—their industries were the envy of the world. Recall that in the late 19th century, Britain was known as the “workshop of the world.” The British could, at the time, outcompete anyone, so to them “free trade” meant British dominated trade and, more important, a British dominated world. Besides, the British only encouraged free trade when it suited their interests, unless you argue that their colonial empire was free, which quite obviously was not the case. We can, moreover, say the same thing about the United States, especially since 1945, when American leaders became the leading advocates of free trade. It was after 1945, of course, that the United States emerged not only as the leading military power in the world, but also as the unchallenged economic power. Think about this: Why was Britain only an advocate of free trade once it established its economic dominance? Why was the United States a mercantile state for most of its history, but then suddenly a cheerleader for free trade after it emerged as the world’s top economy? Is this sheer coincidence? I think not!

                            

Table 2.2. Relative Shares of World Manufacturing Output, 1750–1880

Country/Region

1750

1800

1830

1860

1880

United Kingdom

1.9

4.3

9.5

19.9

22.9

Hapsburg Empire

2.9

3.2

3.2

4.2

4.4

France

4.0

4.2

5.2

7.9

7.8

German States/Germany

2.9

3.5

3.5

4.9

8.5

Italian States/Italy

2.4

2.5

2.3

2.5

2.5

Russia

5.0

5.6

5.6

7.0

7.6

United States

0.1

0.8

2.4

7.2

14.7

Japan

3.8

3.5

2.8

2.6

2.4

 

These figures show the relative share of world trade from 1750 to 1880, a period in which the UK rose from being a relatively minor player in the 18th century, to the strongest economic player by the late 1880s. Not coincidentally, a mercantilist will point out, it was only around the middle of the 19th century, by which time the UK was the top economy in the world, that the country abandoned protectionism and advocated free trade. For example, in 1846, Britain repealed the Corn Laws that had protected agricultural producers, and began to lower tariffs. In 1860, “Britain eliminated all remaining protectionist duties and maintained a tariff only to raise fiscal revenue on a few imported consumption items that were either not produced at home or were already subject to domestic excise tax” (Irwin 1993, p. 147).

 

Source: Kennedy (1989), p. 149.

 

Text Box: Figure 2.8. Pat Buchanan and Ross Perot: American Mercantilists

 

Both Buchanan and Perot ran for president several times. Of the two, Buchanan was and still is an unapologetic economic “patriot.” In a 2012 editorial entitled “We Need More Economic Nationalists,” Buchanan writes, “[Global] free trade makes suckers and fools out of patriots.” Perot, in general, was more moderate, but one of his famous lines comes from his criticism of NAFTA, the free-trade agreement among the U.S., Canada, and Mexico. As he put it, NAFTA would lead to a “giant sucking sound,” as he argued that lower wages in Mexico would suck up hundreds of thousands of American jobs. 

Picture of Ross Perot is licensed under Creative Commons Attribution-Share Alike 3.0 Unported; picture of Pat Buchanan is licensed under the GNU Free Documentation License.

My main point is this: the principles of liberal IPE are most strongly advocated by countries—or groups of people, regardless of their nationality—that have the most to gain from a more open economic system. But those who would suffer from so-called free trade and free markets oppose these principles. It should come as no surprise, then, that there are always intense debates, even within the wealthiest countries, about the advisability of open markets. This was clearly the case in the United States, where battles over protectionist trade policies started in the early 1800s between Northern industrialists and Western farmers, on the one hand, and Southern planters and Northeastern merchants on the other hand. And it has continued right up to today. Witness, for example, the relative success of political figures like Pat Buchanan and Ross Perot (yes, I know my references are a bit dated, young man, but I’m rather old myself), who fought the Republican establishment tooth and nail over the issues of free trade and NAFTA (North American Free Trade Agreement). At heart, both men are mercantilists—although they would prefer to call themselves “economic patriots”—who are committed, first and foremost, to the welfare of the United States. By contrast, we should also not be surprised that many political leaders in the so-called Third World are advocates of liberal economic policies. Occupying privileged positions, such leaders have much to gain from implementing liberal policies. This is true even if most of their compatriots suffer from the ravages of free markets. But even the Third World leaders who aren’t willing to open their economies are eventually forced to do so by U.S.-dominated institutions like the IMF and World Bank.

Student: What ­do you mean “forced”? How can international banking institutions, such as the IMF, force countries to open their economies?

Friedrich: Actually, this is something Karl is much more interested in. However, the short answer is this: countries in the Third World—perhaps a better term might be developing world—do not have well-developed and competitive industries, and therefore are unable to generate the foreign exchange they need to import necessary goods, such as oil, food, capital goods, medicine, and so forth. Almost all have also borrowed money from international sources to finance the little industry and infrastructure they do have. These loans must be paid back in hard currencies, such as U.S. dollars. If, for any reason, these countries cannot pay for essential imports or make payments on their loans, they need to borrow more money, but usually the only organization willing to lend that money under such circumstances is the IMF. The richest countries, though, largely control the IMF. Naturally, they want something in return for their “generosity,” and what they typically ask for—demand, really—is “market liberalization.” That is, through the IMF, the rich countries demand that poor countries open their markets in return for access to a short-term “bailout” loan. This is called conditionality

Student: I can see your point. It seems Hamilton and List really knew what they were talking about.

Friedrich: Precisely, my dear boy. However, lest you think that Mr. List’s only contribution was to show the problems with the liberal perspective, I must point out that he made other important intellectual contributions. In his most important book, The Natural System of Political Economy, for example, List helped us understand that it is not wealth per se that is important to a nation, but “productive power.” By this he simply meant the capacity to make or manufacture a good, rather than the mere possession of that good, or of the specie—i.e., gold and silver—to buy it. To List, the power to produce was important because it developed the necessary foundation of human skills, technological know-how, and industrial expertise necessary for long-term prosperity. To develop this capacity, however, is not a simple thing. For it requires a nation to “sacrifice and give up a measure of material prosperity in order to gain culture, skill, and powers of united production” in the long run. But will people, on their own, sacrifice material prosperity for the sake of the nation at large? Certainly not. But, even if they do, who will help channel the extra resources into the right areas? The answer, my friend, is obvious: the state. My dear boy, can you think of any contemporary examples that confirm what List said? I’ve already given you a hint, or rather the answer, so just think back to what I said earlier.

Student (scratching his head): Well …

 

Table 2.3. Top Ten Largest Companies in China and Global Rank (by revenue), 2012

 

Rank*

Company Name

State-owned

Revenues

(U.S.$ millions)

Industry

5

Sinopec Group

Yes

375,214

Oil

6

China Petroleum

Yes

352,338

Oil

7

State Grid

Yes

259,142

Utilities

54

Industrial & Commercial Bank of China

Yes

109,040

Banking

77

China Construction Bank

Yes

89,648

Banking

81

China Mobile Communications

Yes

87,544

Telecommunications

84

Agricultural Bank of China

Yes

84,803

Banking

91

Noble Group

No

80,732

Energy

93

Bank of China

Yes

80,230

Banking

100

China State Construction Engineering

Yes

76,024

Construction

 

*Rank is from Fortune magazine’s annual list of the world’s 500 largest corporations, or the “Global 500.”

 

Source: http://money.cnn.com/magazines/fortune/global500/2012/full_list/

 

 

Friedrich [visibly impatient]: Back when I was a younger man, I would have said Japan or South Korea, but the best example right now is China. Just think about this: China is ruled by the Chinese Communist Party, or CCP. The CCP still runs things in China, including the economy. In fact, some of the largest companies in the world today, such as Sinopec and China National Petroleum, are owned and controlled by the Chinese government (see table 2.3). And there’s little doubt that the CCP protects and subsidizes Chinese companies, whether state-owned or private. And guess what else? Over the past two decades, China has transformed itself into the second largest economy in the world, and while it still earns a lot of money through the export of cheap, low quality goods, it is also moving into value-added, high-technology sectors—at a very rapid pace.

Student: Wow, I didn’t realize that. So what you say must be the truth! I can clearly see that there is a great deal of competition among states today. I can also see how important power is. The strong, mainly Western, states can dictate to other states what they can and cannot do. I mean, during the 1990s, Iraq couldn’t even sell its own oil without getting permission from the West! The strong states even have their own exclusive clubs, like the OECD, NATO, the G8, the G8+5, the BIS, and the Davos Forum, and a few others. I’m not sure what these clubs do, but they must be designed to promote the interests of only their members, or else why would they be exclusive in the first place? Why, now that I think about it, I can even see that globalization is nothing more than an effort by the strong and wealthy states to force everyone to open their markets, not because the whole world will benefit, but because open markets mean more profit, more wealth, and more power for those who already have it! Surely, the rest of you must agree?

Joanna: Well, I agree that Fred tells a good story, which is adequate as bedtime reading, but it’s not at all satisfactory as science.

Friedrich: Now, be nice, Joanna.

Joanna: Young man, before you take to heart what Friedrich here has said, you might want to hear me out first. Then you can decide where the truth really lies. In fact, the liberal perspective of IPE is a thorough critique of all that you heard above, and more. [Note: as we saw in chapter 1, there are several liberal variants, such as classical and neoclassical economics, Keynesian economics, libertarianism, and the like.]

                        To begin, let me state a principle that is common to most, if not all liberal perspectives: namely, that free-functioning markets, based on a division of labor (i.e., specialization) and mutually advantageous trade, will ultimately lead to the

Figure 2.9. Map Showing Members of the OECD

OECD stands for Organization for Economic Cooperation and Development, which has a current membership of 34 countries. The stated mission of the OECD is to promote policies that will improve the economic and social well-being of people around the world. Until recently, however, membership in the OECD was restricted to the major capitalist economies tied to the U.S. and Western Europe. Note how much of the world is still left out of the OECD.

Source: The image is licensed under the Creative Commons Attribution-Share Alike 2.0 Generic license.

 

greatest good for all—for individuals, societies, countries, even the entire world. Moreover, achieving this “greater good,” by which I mean general prosperity, does not require force or coercion of any kind; nor does it require planning on the part of any authority, centralized or not. Indeed, this is the beauty of a liberal economy: it is an essentially spontaneous, voluntary, and self-regulating system guided only by the “invisible hand” of which Adam Smith spoke so eloquently. The invisible hand, by the way, is the metaphor used by Smith to describe the principle by which a benevolent society emerged from the unintended consequences of individuals acting in their own interests—or as Smith put it, on the basis of “self-love.” All of this is encapsulated in the famous phrase that Smith borrowed from the French: laissez-faire, laissez-passer, which simply means, “Let things proceed without interference.”

There are several other liberal assumptions that you should know. The most important of these, perhaps, is that human beings—or larger collectivities of individuals like the nation-state— pursue or act in their own self-interests. This concept is embodied in Smith’s idea of self-love and has been extremely important to the development of liberal economics, which is something I’ll talk about in a second. Before I do this, though, I should say that the idea that people act in their own self-interests—in other words, that we are rational actors—has been criticized as highly unrealistic. To be sure, the concept of rationality is an abstraction or simplification of a much more complex reality. We liberals understand this. But abstractions are necessary for theorizing: if one doesn’t abstract, one can’t hope to make sense of the world. We just happen to believe—and with good reason I might add—that much can be learned by trying to reduce human behavior to its essence, while recognizing that culture, religion, history, and yes, even power, as well as a whole host of other factors, are likely to affect and shape the actions of human beings in ways that cannot be wholly predicted. Trying to identify the essence of human behavior does not mean, moreover, that liberals are heartless monsters—something we have also been accused of by our critics. In fact, liberals, almost by definition, are fundamentally concerned with improving the human condition. That is, we are committed to understanding how all societies can achieve democracy, freedom, and prosperity!

I realize all of this may seem old hat to you, young man. But, I really cannot emphasize enough the importance of these principles, which are so basic to most liberals.

Student: Oh, no. It’s all quite interesting, although I’m not yet sure how your story is better than the one told by Friedrich.

Joanna: I will get to that in a moment. First, though, let me go back to a point I raised earlier, namely, that the development of political economy, as a science, owes a great debt to Adam Smith. Smith’s brilliant concept of the invisible hand, for instance, allowed us to see that there can be “spontaneous order” in the seeming chaos of human activity. This was an extremely important insight because it essentially made social or economic science possible (Vaughn 1998). Although the reason for this is not readily apparent, it is not particularly complicated either. To put it simply, in order to develop a scientific understanding of the social world, it is first necessary to develop a concept that allows us to speak of regularities, patterns, or general tendencies existing in human society—Smith’s invisible hand did just this. Before Smith came up with the idea of the invisible hand there was no real alternative to conceiving of all social institutions and practices, on the one hand, as products of carefully conceived and fully orchestrated social engineering, or, on the other hand, as the result of natural or supernatural phenomena beyond the ken of human understanding. 

There is, however, a fundamental problem with these two choices: the former is obviously wrong, since human beings are clearly not infallible, God-like creatures, while the latter means that the concepts of explanation and understanding must be confined to religious dogma, superstition, or mere speculation/opinion. The idea that the unplanned and uncoordinated actions of myriad individuals acting in their own self-interest could lead to the creation of orderly (and highly efficient) social institutions, like the market, by contrast, literally set our minds and our intellects free. I’m sure you can see the importance of Smith’s insights, can’t you?

Student: Yes, I think so. Essentially, Smith’s ideas provided a basis for applying the methodology of the natural sciences to the social sciences. Is that it?

Joanna: Precisely! Although Smith’s framework did not allow for the use of sophisticated mathematics, he clearly laid the first bricks in the foundation of neoclassical economics, which is the epitome of social-scientific reasoning and analysis. The concept of self-love, for example, may sound fuzzy and unscientific, but it was translated by later economists into the concept of rationality, which allowed the development of more sophisticated mathematical models.

Student: You keep talking about “sophisticated mathematics,” but why is this so important?

Joanna: That’s a good question. The reason is simple: mathematics is the language of science. It is precise, logical, and objective. And it is only through precision, logic, and objectivity that we might discover the truth, which seems to be one of your main requirements.

 Student: Are you saying, then, that those who don’t speak the language of mathematics aren’t capable of speaking the truth?

 Joanna: In an important sense, the answer is yes. Even if some or most of what they say is valid and useful, unless we can translate their stories, or narratives, into science, we can never know if what they say is the truth. But I am beginning to digress here. I started off talking about liberal perspectives of IPE, and not all liberals speak the language of science; a few even have different views of such central concepts as laissez-faire and the invisible hand. Some “revisionist” liberals, for example, believe that markets sometimes need a little outside help to operate smoothly and efficiently. In this regard, you probably already know something about John Maynard Keynes, who convinced a lot of people—scholars, politicians, and bureaucrats alike—that government intervention is occasionally necessary to maintain full employment, control inflation, and encourage economic growth. Keynes’s ideas started to die out in the 1980s and 1990s, but he certainly had a huge impact on all of our lives. Indeed, his ideas were resurrected in the U.S. in order to deal with the financial crisis of 2007. So, if you want relevance all you have to do is think about Keynes and his legacy. As Keynes himself said: “The ideas of economists and political philosophers, both when they are right and when they are wrong, are more powerful than is commonly understood. Indeed the world is ruled by little else. Practical men, who believe themselves to be quite exempt from any intellectual influences, are usually the slaves of some defunct economist” (1936, pp. 383–84)—or, as I might add, some defunct liberal economist.

But Keynes is not the only revisionist out there who sees an expanded role for the government. One more recent argument revolves around the idea of endogenous growth—sorry for the use of jargon here—which holds that investments in human capital, innovation, and knowledge are main contributors to economic growth. A key implication of this view is that the long-term growth of a national economy is dependent on certain policy measures, such as subsidies for research and development and education. Needless to say, this contrasts with most other models of economic growth in neoclassical economics, and provides even more space for the government than Keynes envisioned. In my view, this is a slippery slope, but I mention it to give you a sense of the variety of revisionist perspectives within liberal economics. [Note: For further discussion of endogenous growth, see Romer (1986) and Lucas (1988).]

Significantly, even when you consider certain arguments originally put forth by mercantilists, we can see that an understanding of liberal principles changes things. Consider, on this point, an argument normally associated with mercantilism called hegemonic stability theory. The standard version of hegemonic stability, which is the phrase most typically used to describe this perspective, portrays the hegemon (which is defined as the predominant military-economic-political power) as sort of an interventionist kingpin: as the kingpin, it maintains a degree of order or stability in the international economy by underwriting and, when necessary, enforcing the rules that govern the system (the different sets of rules are also referred to as international regimes). Without the stabilization provided by the hegemon, in other words, the world markets would not only operate inefficiently, but could fall apart completely. From a liberal perspective, this is true because, like domestic markets, international markets operate best when certain public goods are present. These include such things as a sound financial system, free trade, peace, and security. In a domestic market, the state provides these goods and pays for them by collecting taxes from its citizens. This doesn’t work at the international level, however, since all states are ostensibly sovereign. The only way around this problem is if one state can afford to bear most of the costs of providing international public goods. This is where the hegemon comes back in: because it is, by definition, the most efficient and productive economic power in the world, it can recover the costs of underwriting the system through a general increase in world trade (because it will end up trading more). In this sense, I might note, we can still see that the principle of rationality is operating, since a hegemon is ultimately acting in its own interest. In this liberal view of hegemony, then, the hegemon is clearly benefiting, but so too is everyone else! In other words, it’s not the my-gain-is-your-loss world of mercantilism, which, I should emphasize, is a key (but usually hidden) assumption of that approach. Do you understand what I’m saying here? [Note: We will learn more about HST in chapter 3.]

Student: Why, yes … I think I do. Mercantilists start off with the premise that we live in a zero-sum world. Based on that assumption, Friedrich’s argument is logically sound. But, liberals like you, Joanna, start off with the premise that we live in a positive-sum world—your gain is my gain and vice versa—which not only means the beggar-thy-neighbor strategies of mercantilists are logically flawed, but actually make us all worse off. It’s pretty darn clear, too, that we don’t live in a zero-sum world—why, just look at how much the world economy has grown in the last few decades, not to mention the last few centuries. It sure does appear that we’re all getting rich at the same time, which is what you seem to be implying, Joanna. Wow, I didn’t realize assumptions could be so important.

 

Figure 2.10. Growth in World Adjusted Net National Income (Constant $U.S.), 1970–2010

 

 

Between 1970 and 2010, world net national income, in constant terms, increased from just under $11 trillion to almost $34.7 trillion, an increase of more than 300 percent. This helps to illustrate the liberal emphasis on the positive-sum nature of capitalist development. In this view, one country’s gain is not another country’s loss; instead, as long as there is free and fair trade, both benefit because the economic “pie” is getting bigger.

 

Chart generated by Google based on data from the World Bank: http://www.google.com/publicdata/.

 

 

Joanna: You’re finally seeing the light—or should I say the truth. And you’re exactly right: assumptions are key. You’ve got to understand the assumptions that underlie a theory before you can properly evaluate what that theory is saying. The assumption that free-functioning markets lead to positive-sum results, for example, underlies the theory of comparative advantage, which is another key element of most liberal perspectives.

Fred, you’ll remember, argued that international trade only benefits the most economically advanced nations—but you’ve already seen the logical flaw in the mercantilist position. What you also need to know, though, is that David Ricardo [1772–1823] provided the original insights as to why the liberal perspective is superior.

Basically, Ricardo showed that even when one country (country A) can produce each of two products at less cost than another country (country B), it will still be worthwhile for them to trade. For example, if country A can produce both wine and semiconductors cheaper (and better) than country B, it would still make economic sense for country A to specialize in that product for which is has a relative advantage. That is, if country A is a very efficient producer of semiconductors, but not quite as efficient as a wine-maker, it will be better off concentrating on semiconductors, which it can then trade for wine from country B. Country A is clearly better off, but so is country B, since it can now import semiconductors at a much lower cost than would have otherwise been the case. So, you see, they both win! In the long run there only winners and losers—and this is the problem with mercantilism—when one country attempts to alter the “natural” basis for mutual gain by constructing protectionist barriers or by giving domestic industries other unfair advantages. Friedrich’s argument, then, sounds persuasive because he and other mercantilists are narrowly focused on short-term and highly exclusionary gains at the expense of long-term and general prosperity. I hope, dear student, that you are now fairly convinced that the mercantilist position is fundamentally flawed.

 

Table 2.4. Trade and Comparative Advantage: Opportunity Costs and Efficiency

 

 

 

This is a very simple example of how trade and comparative advantage might work. In the example, a U.S. worker can produce more rice than a Chinese worker, but fewer iPhones over 10 hours. The Chinese worker, by contrast, produces only half the rice an American worker can produce in the same 10 hours, but can produce two more phones. If workers specialize—U.S. workers focusing on producing rice and Chinese workers on phones—and then trade, workers in both countries are better off. In the example above, neither country possesses an absolute advantage—meaning one country is better at producing both rice and iPhones—but even if one of them did, it would still make sense to specialize and trade, with American workers focusing on what they are best at producing (and vice versa).

 

Source: Tables are adapted from Ralph Byrns, “Comparative Advantage and Absolute Advantage,” http://web.archive.org/web/20130615222305/http://www.unc.edu/depts/econ/byrns_web/Economicae/Essays/
ABS_Comp_Adv.htm
 • The author (Byrns) has granted provisional permission to use his materials for educational purposes only.

 

 

Student: You do make a convincing case; in fact, it’s pretty difficult to disagree with anything you’ve said. If I think about great capitalists like Henry Ford or Bill Gates, I can easily see that when individuals are left to their own devices and ingenuity, they are able to come up with a product or system that makes them immensely wealthy, but that also benefits countless other people. If I remember correctly, Henry Ford—and his ideas about the assembly line, for example—made his workers and their communities better off, his investors better off, and his country better off. But, he also—I think—made the world better off, since his innovations obviously didn’t prevent the Germans, British, Italians, Japanese, Swedes, and others from following suit. And what Henry Ford did for the automobile, Bill Gates has done for the computer. It’s certainly hard to argue that his innovations in software and programming haven’t made the world a more productive, prosperous, and just plain better place. How can we begrudge Gates his fabulous wealth when he’s obviously done so much good? It all makes perfect sense, now that you’ve explained things to me. Certainly, if governments around the world would just stop meddling in markets, rely on their comparative advantages, and give their citizens the freedom to make a profit, everyone would be better off. I can even see how this laissez-faire approach is fundamentally democratic.

Joanna: You’ve obviously been listening well. I might add, too, that when you combine liberal ideas with the language of science and centuries of historical evidence (like the rise of the liberal West and the collapse of Soviet and Eastern European communism), you have a powerful combination.

Karl: Powerful, yes. But only in its ability to fool most of the people most of the time. For if liberalism was really all that you say it is, the world should be on the cusp of Utopia. Just think about the logic of the free-market ideal, which tells us that power is irrelevant, greed leads to prosperity for all, individual freedom is assured, and peace and democracy will spread throughout the world if only governments let people pursue their own self-interests unfettered by all but the most minimal controls, laws, and regulations. Sheer poppycock! Why, just look around you. Billions of people still live in poverty, misery and degradation—and not just in the developing world, but right in the midst of the richest country the world has ever seen. Just the other day, I came across an astounding statistic: according to a study by scholars at the University of Michigan and Harvard University, the number of families living on $2 or less a day in the United States is almost 1.5 million—more than double the figure 15 years earlier (cited in Bello 2012).

Student: You mean you disagree? To me, liberalism sounds so logical, so objective, so scientific, so convincing. How could what Joanna says be wrong?

Karl: Unfortunately, my young friend, she couldn’t be more wrong. Why don’t you sit back while I tell you a different story? Before I begin, though, I must ask you to listen carefully and with an open mind, for you have no doubt been bombarded by propaganda about the supposed evils of Marxism. Also, I am equally sure that what you think you know about Marxism is colored by the now more than two-decades-old collapse of the Soviet Union and Eastern Europe, which liberals have gleefully pointed to as evidence of its utter failure as an ideology and as a real-world program of action. Needless to say, the liberals have it all wrong. Before I tell you why, though, let me start off by emphasizing that Marxist IPE, like liberal IPE, is a fairly diverse body of work with many different strands. There are, however, a few concepts that clearly distinguish Marxist perspectives from the others you have heard about thus far. Can you guess what these concepts might be?

 

Table 2.5. People Living on Less Than $2 Day, 1981–2008 (Selected Years)

 

 

This table shows the number of people in different regions living on less than $2 day (in PPP terms), based on calculations by the World Bank. Currently, the World Bank uses $1.25 as the cut-off point for severe poverty—and based on that figure, poverty has declined significantly since 1981. However, at less than $2 a day, many people cannot afford basics such as food, shelter, and access to clean water. It is worth noting, too, that as many as 1.4 million families in the U.S. live on less than $2 a day, per person, a number that more than doubled between 1996 and 2012 (Bello 2012).

 

Source: World Bank (2013) at http://iresearch.worldbank.org/PovcalNet/index.htm?0,3

 

 

Student: I don’t really know, but I’ve heard a few terms or phrases connected with Marxism, like class struggle, exploitation, alienation … things like that.

Karl: Yes, all those concepts are important, although it’s important that you first understand the intellectual foundation of Marxism, which is something called historical materialism.

Student: That sounds pretty abstract and not particularly relevant. What does it mean?

Karl: I can understand your reservations. But historical materialism is not a difficult concept to grasp, and it is certainly relevant to your life—even if you don’t realize it yet. Let me explain. The term materialism refers to the simple, yet profound, idea that economic or material forces play the key role in shaping the world as we know it, as well as our individual consciousness. In the feudal world (the era just prior to capitalism), for example, society was organized around a predominantly agricultural mode of production. This meant, among other things, that those who owned or controlled agricultural land—the landlords—necessarily occupied the top rungs of the socioeconomic hierarchy, while those who actually worked the land (i.e., peasants or serfs) were much, much lower down. Even more important is the nature of the relationship between lord and peasant, which was based on inequality and exploitation. In other words, the economic organization of feudal society virtually dictated an oppressive system of class relations centered on land ownership. As you know, however, the feudal system did not last. It eventually gave way to capitalism, which, needless to say, is based on an entirely different mode of production—one that privileges ownership, not of land per se, but of the means of production. With a new mode of production, as you might guess, comes a new set of class relations, which helps to explain why, to a significant extent, the monarchs, lords, and other rulers of the past have been reduced to fodder for the tabloids and paparazzi of today (especially in the most advanced capitalist countries). Consider the British royal family: the Queen has the “right to rule,” but her role is mostly symbolic. Even if she were to exercise her royal prerogatives (e.g., the right to refuse a government’s request to dissolve Parliament and call an election), real power lies not with the royal family, or even with the British Parliament. Instead, real power lies with Britain’s leading capitalists. Of course, the concept of materialism is a little more complicated than this, but I think you get my point, do you not?

Student: Yes, I believe I do. But why do Marxists use the phrase historical materialism?

Karl: Marx used the adjective historical before materialism to highlight his point that history is marked, or defined, by epoch-making shifts in the dominant mode of production. As I just mentioned, the shift from feudalism to capitalism represented one such shift. But we can also expect another: from capitalism to communism (where socialism is a sort of intermediate step). In this sense, historical materialism is a theory of history (Crane 1991, p. 9), since it purports to tell us not only how history unfolded in the past, but how it will unfold in the future. In a similar vein, Marx used the term historical materialism because he saw human societies as embedded in their own past, and thus he regarded history as the necessary method for any adequate understanding of the world (Abrams 1982, p. 35).

In this sense, historical materialism is also important because it establishes the basis for a scientific understanding of society. To Marx, in other words, history was obviously not some random or haphazard (and therefore unpredictable and unexplainable) series of events, but part of a single, nonrepetitive process that obeys discernable laws. While Marx recognized that historical laws are different from the laws of physics or chemistry, he believed that they could still be used as the basis for a scientific understanding of human society. Here, I might mention that Marx saw the process of historical change as comparable to the geological sciences, which are based on an understanding and analysis of a cumulative but continuous process of change (Berlin 1973, p. 57). Unlike the geological sciences, however, the historical development of humankind, in Marx’s view, is irreversible and necessarily progressive: every new epoch is characterized by greater freedom, equality, and prosperity. Capitalism, in this view, is an unequivocally significant improvement over feudalism. But it is not, contrary to what liberals may have you believe, the final stage of our historical development.

 

 

Figure 2.11. Historical Stages in Marxist Thought

 

 

Marx argued that human society evolves, moving from one historical stage to the next. Each new stage represents a progression from the previous stage, but each stage is necessary. In other words, it is not possible to skip a stage (as the Soviet Union, China, and other so-called communist countries tried to do). Why? Because each stage provides the material or economic foundation needed to move forward: moving to the communist stage requires the productive capacity of capitalism.

 

Image Source: Created by author. Image of stage is from http://www.pptbackgrounds.net/stage-background-content-slide-backgrounds.html (according to site, “This [background image] is free to download, ready to use.”

 

 

Student: So what you’re saying is capitalism is better than what it replaced, but that it still represents a lower stage of historical progress?

Karl: Precisely.

Student: I’m not sure I buy your argument. The world is not perfect—obviously— but that may only be because so many people are still misinformed about the virtues of a liberal economy. I really can’t realistically imagine a better world than the one we have right now. And, honestly, hasn’t the market mechanism proven itself to be superior to anything else? Hasn’t it created a more efficient, more prosperous, more democratic, and ultimately freer society for all of us? Wouldn’t we all be better off if we simply accepted the rationality and efficacy of market forces—even if they don’t produce perfect results for everyone?

Joanna: Ahh, the joy of a lesson well taught . . .

Karl: Don’t interrupt, Joanna. You had your turn, now let me have mine!

Joanna (grinning): I’m so sorry, Karl. Please go ahead.

Karl: Thank you. Now where was I? Oh, yes. The points you bring up, young man, are ones that Marx addressed long ago. You see, historical systems, like capitalism, originally arose to meet the needs of human beings. However, as time went by, these systems began to acquire a life of their own, so to speak. Thus, rather than simply serving the interests of the humans who created them, they eventually came to be seen as existing independently of—or even above— those needs and interests. When this happens, the relationship between people and the institutions they create is sometimes reversed in that the requirements and values of the system take precedence over the needs of human communities. This is precisely the case with the capitalist system, which many people portray as a natural creation whose authority and basic values we—as human beings—have no business challenging. Why, just look around you. Our lives are increasingly being governed by what the supposedly impersonal market “says” is good or bad.

                        The market, for example, tells us that greed (or unfettered self-interest) is good, because it creates an incentive for innovation and higher levels of productivity. If greed also results in extreme income inequality, this is good, too. This is not an exaggeration. For example, in an article in Forbes magazine, a bastion of free-market ideas, John Tamny wrote, “income inequality is beautiful” (2013). It is beautiful, Tamny argued, because it is “gaps in wealth that drive creativity among the citizenry. Seeing the immense wealth possessed by the most successful, those not in the rich club strive mightily to join the wealthy; their innovations redounding to individuals of all income classes.” If income inequality is beautiful, then equality must be ugly, or bad. This means, too, that exploitation is good. After all, it is what the market dictates.

In particular, we are implicitly, but unmistakably, told that the entire world should be governed by one overriding principle, namely, efficiency. We are told that capitalism should not be questioned because it creates unrivaled efficiency (which it does). But why should efficiency be the preeminent value of human society? What about other values—or measures of progress—such as distributive and social justice, human rights, political equality, and so on? Why should all these values be trumped by efficiency? [Note: The question of which values take precedence over others is a core element of the definition of international political economy we discussed in chapter 1.]

Clearly, though, historical systems do not achieve dominant status on their own. Instead, such status is achieved through the assiduous efforts of those social classes who hold power. During feudal times, for example, it was the landowners and the nobility who took on this role. I should stress, though, that the effort to legitimize an oppressive system is not merely the result of brute force; rather, it is a comprehensive effort that involves imposing an intricate set of cultural, religious, and intellectual values on society at large. Thus, even though these values are meant to benefit a single, privileged class, they are often accepted by everyone in society, including those on the very bottom rung of the social ladder. This is a point made brilliantly by Antonio Gramsci in his writings about cultural hegemony. [Note: We briefly discussed Gramsci’s ideas in chapter 1.]

Getting people to believe that God himself ordained that some men should be kings and others peasants is a perfect example of this. Today, of course, we know that kings are no closer to God than the rest of us. In our present system, however, a new king—or, dare I say, God—has arisen: the market. The fact that you, dear student, can’t imagine a better world than what exists now or that you are ready to passively accept the premise that whatever the market dictates must, by definition, be just and good, only means that you have been successfully co-opted—your consciousness is the consciousness of the market.

But don’t feel bad. Just as in feudal times, those in power have used their control over intellectual life, the political and legal systems, and cultural and religious institutions to get you to think the way you do. In short, they have created an illusory common interest, which can fool even the most steadfast blue-collar worker. In this regard, Joanna is simply another mouthpiece for the capitalist class, even if she doesn’t realize it. Her liberal ideas, which she holds up as “science” and “the truth,” are nothing more than a self-serving justification for an inherently exploitative and oppressive system.

Text Box: Figure 2.12. Marxist View of the State

 

In Marxism, the state is at least partly an instrument—or puppet—of the dominant class. Thus, when Bush advocated for a war against Iraq, he was not expressing a national interest; instead, he was expressing a class interest.

Source. Image of George W. Bush is in the public domain. Added elements are by the author. 
Student: I never thought of it that way. But when I think of how we use the market to justify all the inequality, injustice, oppression, and exploitation that continue to exist in the world, I find it hard to dispute what you say. Why, just the other day I read an older article that justified child labor by saying that, no matter how horrible and exploitative it may be, it is acceptable as a mutual “exchange that benefits both the buyer and the seller” (Khoury, 1998). And then, during the 2012 Republican campaign for president, Newt Gingrich talked about abolishing child-labor laws in the U.S.: he actually wanted to replace adult janitors at schools with children as young as nine years old, saying that such labor was good for the kids who otherwise would never develop an appropriate work ethic. I must admit that the argument is somewhat appealing, but it does seem perverse to hire children to take the place of adult workers.

Karl: I do believe, my boy, that you are beginning to truly understand a little about Marxism. Your education, however, is still incomplete. Capitalism, as I have said, is a unique historical system. As with all historical systems, it has its own distinguishing characteristics, the most salient of which is a class structure consisting of those who own the means of production (i.e., the bourgeoisie, or capitalists) and those who possess only the capacity to work (the proletariat). The division between these two social classes is a fundamental feature of the capitalist system and, as such, is the starting point for any analysis of society. This, I might note, is a critical difference between Marxism and the other perspectives you have heard about. Liberals, for example, start their analysis with a sort of generic individual, while mercantilists start with the state. You do understand why this is a significant, don’t you?

Student: Well, I can’t say I do. What does it matter whether you focus primarily on the individual, the state, or class?

Karl: My dear boy, it matters a great deal. A focus on the state, for example, presupposes that it is an independent or autonomous actor, which means that a state’s actions are not shaped or determined by a more basic force. To Marxists, however, there is a more basic force, namely, class relations. Look at it this way: any attempt to understand the actions of the state that starts—and ends—by examining the attributes and/or actions of the state itself would be like trying to understand the movement of planets by examining only the attributes of planets themselves. In other words, just as one cannot ignore gravitational forces in astronomy or physics, one cannot ignore class forces in political economy. Consider the following question: Why do states go to war? If we try to answer this question by focusing solely on the attributes of states, we might say that states go to war in order to protect the national interest or, perhaps, to correct an imbalance of power. We might even say that some states are naturally aggressive. But does this really tell us what we need to know? Certainly not. We do not know, for example, how or why the national interest is defined the way it is; nor do we know what causes an imbalance in power relations or why states become aggressive in the first place. Class-based analysis, by contrast, can answer all of these questions. While Marxists may not agree on the exact causes of specific wars, most would agree with the idea that wars are invariably fought to protect or promote the interests of the dominant class, which in the modern period is the capitalist class. Thus, when the typical soldier lays his life on the line, he is not protecting his interests or even the interests of his country at large, but the relatively narrow interests of the ruling class. Sadly, then, the people who actually fight the wars (i.e., the proletariat), usually work against their own interests—more often than not, in fact, they end up killing those with whom they should be making common cause. I think, my boy, you could not ask for Text Box: Figure 2.13. Picture of a Foxconn Factory Sign

 

Foxconn, a subsidiary of Hon Hai Precision Industry Co. (Taiwan), produces products for Apple, Amazon, Sony, Nintendo, Microsoft, and others, using low-cost labor in China.

Image is licensed under GNU Free Documentation License.
much more relevance than this.

A focus on individuals, I should add, is misguided for precisely the same reasons—i.e., an over-emphasis on the role of agency. As with states, individual actors cannot operate independently of the social structures in which they find themselves. What this means is that individual action does not and cannot take place in a social vacuum—social or class structures, in other words, force individuals to take up definite roles in relation to one another and to the means of production. We would expect, therefore, fundamental differences in interests and opportunities to exist between, say, Tim Cook (CEO of Apple) and one of the thousands of young Chinese who work for less than $17 a day (Economix Editors 2012)—a princely sum for many poor Chinese—in one of the many factories used by Apple in China, factories that are operated, I might note, by Foxconn, a subsidiary of Hon Hai Precision Industry Co., which is a Taiwan-based company.

Student: I think I’m getting a clearer picture of the importance of both social class and structure, which neither Friedrich nor Joanna even mentioned. But is Marx saying that agency is completely irrelevant?

Karl: Not at all. As Marx himself so aptly put it (in one of his most famous quotes): “Men make their own history, but they do not make it just as they please; they do not make it under circumstances chosen by themselves, but under circumstances directly encountered, given and transmitted from the past” (Marx 1853 [1963], p. 15). Marx, in other words, understood that people have the capacity of free will, but that it is always exercised within a particular context that cannot be ignored or dismissed.

Student: Okay. I see what you’re saying. But, how does all this talk about social class and structure relate to the real world?

Karl: The answer, my boy, is clear. As I said earlier, social or class structures force individuals to take up definite roles in relation to one another and to the means of production. This means, for example, that capitalists and workers are locked into a fundamentally antithetic relationship—in other words, it is almost impossible for the two classes to form a peaceful and mutually advantageous relationship. Why? Quite simply because the actions of both sides are shaped and constrained by their positions in the division of labor and by the broader dynamics of the capitalist system. This is easy enough to see in the real world. Consider, if you will, a good-hearted capitalist—a man who wants to provide a better life for his workers and his community. What would he do to achieve this? Well, he could raise wages, provide better health benefits, build a childcare center, fund a generous retirement program, etc. As a capitalist, however, doing such good deeds raises his costs; his firm, in short, becomes less competitive vis-à-vis other firms. If these other firms further undercut him by moving production offshore—say, to Indonesia, where the state uses its military and police powers to break unions and suppress wages—the good-hearted capitalist may not be able to compete at all. If this happens, he may be forced to lay off workers, reduce wages, eliminate health care benefits, ignore environmental and safety regulations, or cease production altogether; or he may move his factory overseas, too. So, the next time you hear of a factory closing its doors to move to another country—or even to another state where unionization is weaker—you might want to keep this in mind. The main point, to repeat, is this: the imperatives of competition, accumulation, and profit-maximization make it necessary for capitalists to exploit the workers they employ. This is the essential context of capitalism, which we simply cannot ignore.

Text Box: Figure 2.14. Base and Superstructure 

 

In the Marxist view, the base determines the types and functions of the institutions that make up the superstructure. Thus, the state enforces private property rights over other “rights” in a capitalist society; education serves to produce a small elite, and a much larger pool of “worker bees”; religion serves as the “opiate of the masses”; while the mass media encourage consumption and spread the “gospel” of the market.

Image designed by author.
                        Of course, we need not feel sorry for the capitalist, for he is the exploiter, not the exploited, which is where the problem really lies. For the more a capitalist exploits his workers, the better off he is. Obviously, to Marxists at least, this is a fundamentally untenable relationship, but one that survives—and even thrives—because the capitalist system is based on an equally fundamental truth: an imbalance of power between the classes. I have already talked about one of the more important implications of this imbalance, namely, the fact that those who have power in virtue of the division of labor also possess the power to define the legal and political superstructure of society. (The superstructure is what exists above the foundation, which in Marxism, is comprised of the ideologies and institutions of society.) That is, they can use the state, religion, law, custom, and academic institutions to naturalize their positions of economic advantage. But the structural imbalance of power also means that the tension between the exploiter and the exploited can never really disappear; in other words, the capitalist system—as with the other exploitative systems of the past, like slavery and feudalism—contains within it the seeds of its own destruction in the form of inherent contradictions, of which the primary one is usually seen as being between the forces of production on the one hand, and the relations of production on the other. 

                        Here it is important to understand, as I suggested earlier, that Marx saw capitalism as the most productive system ever known to humankind. In this regard, Marx also understood that capitalism was a necessary stage in human history, for only capitalism can provide the material and social basis needed for the next and final stage of historical development, namely, communism. Liberals, of course, focus only on the productive potential of capitalism. What they fail to see or account for, however, is that a system of production premised on ever-increasing levels of inequity between classes cannot survive forever. Ultimately, the increasing disparity and out-and-out impoverishment of the working class will lead to the transformation, or overthrow, of the capitalist system.

Student: Hold it right there. You had me going for a while. But, how can you talk about the impoverishment of the working class when clearly that isn’t happening? Certainly, in all advanced capitalist economies, the general population—capitalists and workers alike—have been earning more and doing better. Even in some former developing countries such as South Korea, Taiwan, Brazil, and a few other places, the industrial working class has made huge strides in the past few decades. If we just extrapolate from these experiences, it is easy enough to see that, while poverty may never be completely eliminated, most of the world’s population will reach a reasonable, even comfortable, standard of living in the foreseeable future.

Karl: First, my dear boy, you are committing a serious error in logic, something called the fallacy of composition. This is to assume that what one actor can do in a given set of circumstances must also be possible for an indefinite number of actors to do simultaneously (without, that is, leading to undesirable consequences). For example, one man may pollute a river and still catch edible fish. But if a million men were to do so simultaneously, the river would be destroyed and no one would be able to catch any fish, edible or not. So, yes, it is true that some workers have fared quite well, but always at the expense of other workers—to paraphrase an old saying by Will Rogers, in order to pay Paul, and thereby gain his support, the capitalist must rob Peter. In other words, the phenomenon of the so-called affluent working class is a politically expedient solution to a much deeper problem.

Second, you must also remember that Western capitalism was literally built on the blood, sweat, and tears of the peoples of the Third World. This is something that most liberals conveniently ignore, but it is an irrefutable historical fact: early capitalism was based on the enslavement and super-exploitation of African, Asian, and Indian peoples; the outright plunder of astronomical amounts of precious metals, land, and raw materials, particularly from Latin America; and the political, social, and economic domination of most of the world—which, obviously, continues to this day. None of this, I might mention, has anything to do with individual choice, freedom, or democracy, despite what liberals would have you believe. Finally, you are ignoring the contradictory tendencies that underlie the capitalist system. Consider this simple example: to continue to make a profit, capital must minimize costs, especially the cost of labor. This is done by outsourcing to underdeveloped countries, by automation, or some other laborsaving method. These strategies are effective: over time, they lead to a decline in wages. But as wages decline, so does the market for many goods—since workers make less and since there are fewer workers, there are fewer consumers with money to buy goods. This, of course, means lower profits, which puts more pressure on firms to cut costs. Do you see the dilemma?

Student: Umm, yes, I think I do . . .

 Karl: (Interrupting the student) I am, of course, simplifying a complex argument. Marx’s writings have been subject to a great deal of debate and scrutiny, among both those who are sympathetic to the Marxist view and those who are not. The main point, however, is that exploitation, poverty, and suffering have not diminished at all on a global scale. And there is no sign that this will or can change in the future. As any number of commentators (Marxists and non-Marxists alike) have pointed out, while the post–World War II period has seen a tremendous increase in global wealth, we have also seen a tremendous increase in the disparity between the richest and the poorest of the world. Among the three perspectives, only Marxism has been able to explain, in theoretically consistent terms, why this has happened.

Figure 2.15. The figure compares the population-weighted Gini index (red line) with the unweighted Gini index (black line). The population-weighted index declines almost consistently from 1962 onward. This is mainly due to the phenomenal economic growth in China and India relative to richer countries. Because China and India together account for over one-third of the world’s population, these two countries have a very strong impact on the population-weighted Gini results. But if China and India are removed from the calculation, the population-weighted Gini index trends upward after 1982 (as does the unweighted Gini index), meaning that overall income inequality is increasing in the rest of the world. This graph, we should note, does not unequivocally support the Marxist view, but it does show that global inequality, even accounting for China and India, was significantly higher in 2000 than it was in 1960.

 

Source (graph and text): http://www.conferenceboard.ca/hcp/hot-topics/worldinequality.aspx

 

Student: There are many parts of your argument that make sense to me. After all, any reasonable person would have to wonder how such an extraordinarily productive system as capitalism can, after hundreds of years, still leave millions, really billions, of people in utter destitution, both materially and spiritually. It also seems apparent to me that, if liberals were correct, those capitalist societies with the least amount of government intervention would not only be the most prosperous, but also largely free of social problems and societal conflict. But this certainly doesn’t seem to be case. Why? Perhaps it is because, as both you and Friedrich have pointed out, liberals almost entirely ignore questions of power. Still, I must admit that I am now more confused than ever. You all, in your own way, seem to be stating the truth. But that can’t be possible, given the fact that your disagreements are obviously more than skin deep. The only conclusion I can reach, then, is that there is no truth, at least in IPE.

(Looking dejected) I think I’ll just major in something else.

vvv

            The foregoing conversation suggests that IPE is not only a highly contentious field, but also a seemingly chaotic one. My intention, though, is not to confuse you, but to encourage you to open your mind to different ways of understanding and/or explaining the international or global political economy. As I hope you have seen, none of the three major perspectives (and all their derivatives) should be ignored or, worse yet, dismissed. Indeed, despite the somewhat playful tone of the conversation among Friedrich, Joanna, Karl, and the student, I believe that all three of these major perspectives must continue to be taken quite seriously. This is important whether or not you already hold strong views. For those of you who don’t already have a clear position, for example, taking the three perspectives seriously will, I hope, encourage wider and more disciplined thinking about global political economy. That is, by looking at any particular issue, event, or process from Marxist, mercantilist, and liberal perspectives you will not only consider a larger range of possibilities than would otherwise be the case, but you will do so in a more systematic manner. For those of you who may already have a clearly defined position, taking the other perspectives seriously will, I hope, compel you to critically examine—and defend—your own assumptions and ideas at a deeper, more substantive level than you may be used to. At the same time, the foregoing conversation is intended to encourage, on your part, a healthy degree of skepticism: any claim or analysis that purports to tell “the truth” about the operation and dynamics of the global economy, in other words, should be treated with caution and care.

None of what is covered above is meant to suggest that the only important conversation (or debate) in IPE is among Marxists, mercantilists, and liberals. There are many other conversations, some of which are discussed in more depth in chapter 3. The question for now, then, is: How should students of IPE deal with or approach these various conversations? Here are a few suggestions. First, use these conversations as a convenient way to identify the key points of contention in IPE. These key points of contention are not always visible at the surface; many are hidden in the differing assumptions, axioms (self-evident truths), and core ideas on which all theories are based. This chapter is meant to highlight some of the most important of these. Second, use the ongoing debates in IPE as a still vital foundation for understanding and explaining the global political economy. For, despite the limitations of each of the various perspectives, a great deal of valuable and useful work has been done on a variety of important issues, areas, and problems. As long as you explicitly recognize that research done from a Marxist, liberal, or mercantilist perspective reflects a particular set of assumptions and values (rather than a timeless truth), such work can be, and almost certainly is, quite useful. Third, use the theoretical conversations as an important starting point for reexamining old questions, or for asking entirely new questions about the international or global political economy. As discussed in chapter 1, the questions researchers ask are vital to the process of understanding and explanation. Keep all these suggestions in mind as you continue your exploration of the global political economy.

 


Chapter 3

 

 

 

 

Contemporary Theories of International Political Economy

 

 

 

Introduction

            In the previous chapter, we learned about three of the most important foundational schools of thought, or theoretical perspectives, in IPE. These foundational theories, as we have seen, are still relevant today, and still inform the thinking of scholars and nonscholars alike. In fact, those of you reading this book likely subscribe to one or the other of the foundational theories—at least in part—because the ideas from those theories have become so deeply embedded in our world. There are, however, a number of relatively new, or contemporary, theories about which any student of IPE needs to be aware (“relatively new,” in this case, means that they have been around for the past three or four decades—still a long time, but recall that Marxism, the youngest of the foundational theories, dates back to 1867, while liberalism dates back to the late 1700s). These include revisions and/or offshoots of traditional approaches, but also include a number of approaches that do not fit easily into the mercantilist, liberal, or Marxist camps. Indeed, contemporary theories include a diverse body of approaches that question the most basic assumptions upon which the traditional theories rest. Variously called postpositivist, constitutive, constructivist, or reflectivist, this category of approaches challenge the still widely held belief that the social sciences can produce an objective or value-free truth. Instead these approaches—I will use the term constructivism as a catchall—argue that the social world is unavoidably subjective. Admittedly, the arguments made by proponents of constructivism can be abstruse at times, but the basic lessons are well worth considering. This will be the topic of the last part of this chapter.

            To begin, however, we will take a good look at what is known as hegemonic stability theory (HST), which we briefly discussed in chapter 2. HST is a derivation of mercantilism, although as many scholars have pointed out, it is a hybrid theory, as it contains elements of mercantilism (or realism), liberalism, and even Marxism. Its closest association, however, is with mercantilism. The connection with mercantilism may not be immediately apparent, but it is not difficult to discern. As you read about hegemonic stability theory, then, consider these basic questions: Who is the main actor in this approach? Who or what has power, and how is power distributed? How is power exercised? Following the discussion of hegemonic stability theory, we will turn our attention to post-hegemonic theory. This is an admittedly nebulous and potentially confusing term, which is perhaps unavoidable, since it refers to a wide variety of approaches, all of which attempt to explain the dynamics of international or global political economy in an era in which hegemonic power—and more specifically, U. hegemony—has become more and more contestable over time. On this point, consider the title of one particularly influential book—After Hegemony, written by Robert Keohane (1984). Keohane himself uses the term post-hegemonic many times to emphasize how key dynamics in the international political economy, especially cooperation, are not strictly determined by hegemony. Keohane was primarily interested in the role of international regimes, but other post-hegemonic arguments seriously consider the role and power of nonstate, or transnational, actors in the global political economy. More specifically, in this extended section, we will consider the role of multilateral institutions—e.g., the IMF, the World Bank, the WTO—and corporate and nongovernmental organizations (NGOs). We will also discuss, in some depth, the basic problems with state-centric approaches. Finally, we will look at the two-level game approach. Although often treated as a theory, the two-level game approach works better as a method of analysis. My intention, I should make clear at the outset, is not to provide an in-depth exploration of contemporary theories and approaches. Instead, I will provide an admittedly broad and simplified outline of several of the more important—or more interesting—perspectives to emerge in the past few decades or so, as either useful extensions of, or viable challengers to the three traditional approaches.

 

Text Box:  Figure 3.1. North Korea and Currency Convertibility

 

Few countries today will accept North Korea’s official currency (known as the won) as payment for their goods. The reason is quite simple: North Korean currency is exceedingly difficult to exchange for U.S. dollars or other hard currencies. This means that North Korea must pay for its imports (1) with U.S. dollars or other hard currencies, or (2) through barter (that is, exchanging actual goods for goods from another country). Earning U.S. dollars is difficult for North Korea, since it produces few internationally competitive goods. This is a major reason why the country relies heavily on arms exports, and has been implicated in counterfeiting U.S. currency. 

For most countries today, currency convertibility is not a problem, but North Korea remains an exceptional case because of the inflexibility of its leadership, which remains committed to central planning and near-total control of North Korean society.

Pictured: North Korean 10 won note (image is licensed under the Creative Commons Attribution-Share Alike 3.0)
Hegemonic Stability Theory

The theory of hegemonic stability got its start in the 1970s with the work of Charles Kindleberger (1973), who focused on the reasons for the Great Depression. His basic argument was simple: the root cause of the economic troubles that bedeviled Europe and much of the world in the 1920s and 1930s was the absence of a benevolent hegemon—that is, a dominant state willing and able to take responsibility for the smooth operation of the international (economic) system as a whole. Taking responsibility, in large part, meant acting as an international lender of last resort, as well as a consumer of last resort (DeLong and Eichengreen 2012). More specifically, as a lender of last resort, the hegemon provides access to loans (especially long-term loans) when the normal flow of international lending has dried up; this is also referred to as counter-cyclical lending. Counter-cyclical lending, in turn, is critical to the maintenance of currency convertibility, which refers to the ease with which a country’s domestic currency can be converted into gold or a hard currency. Here is the basic problem: when a currency becomes relatively inconvertible, trade in goods tends to decrease, since many countries are unwilling to accept the inconvertible currency as payment. After all, if currency (from Country A) cannot be converted into gold or, say, U.S. dollars, then the only way it can be used is to buy goods from Country A. (See figure 3.1 for a contemporary example of this issue.) International trade is also negatively impacted if there is no consumer of last resort. In this case, as the consumer of last resort, the hegemon maintains an open market, and encourages other countries to follow suit. If the hegemon or potential hegemon closes or restricts access to its market, as the United States did in 1930 with the Smoot-Hawley Tariff Act, the effects on international trade are usually disastrous. The Text Box: Figure 3.2. Smoot and Hawley, April 1929
 
Willis C. Hawley (left) and Reed Smoot (right), standing together shortly after the signing of the Smoot-Hawley Tariff Act.
Source: Library of Congress. According to the library, there are no known restrictions on the use of this photograph. 

Smoot-Hawley Tariff Act had particularly damaging effects because of the asymmetrical (economic) power of the U.S.: as the dominant economy in the world, the U.S. decision to erect higher protectionist barriers essentially gave the green light to all other countries to do the same. However, once the U.S. fully accepted its role as the hegemon, as it did in the immediate aftermath of World War II with the construction of the Bretton Woods system (BWS), an opposing dynamic was set in place. At the same time, when U.S. commitment began to waver (a phenomenon referred to as benign neglect), international confidence in the dollar-based monetary order quickly began to wane.

The hegemon is willing to take on these responsibilities, it is important to emphasize, for self-interested reasons: as the world’s dominant economy, the hegemon has the most to gain, relatively speaking, from a stable and growing international economic system. At the same time, the existence of a hegemon does not prevent economic shocks and downturns from taking place. Instead, it plays a central role in ensuring that such events do not devolve into full-blown economic crises or depressions. As Kindleberger (1973) explained it, prior to the onset of the Great Depression:

 The shocks to the system from the overproduction of certain primary products such as wheat; from the 1927 reduction of interest rates in the United States … from the halt of lending to Germany in 1928; or from the stock-market crash of 1929 were not so great. Shocks of similar magnitude had been handled in the stock-market break in the spring of 1920 and the 1927 recession in the United States. The world economic system was unstable unless some country stabilized it, as Britain had done in the nineteenth century and up to 1913. In 1929, the British couldn't and the United States wouldn't. When every country turned to protect its national private interest, the world public interest went down the drain, and with it the private interests of all. (p. 291)

The period between the two world wars (also known as the interwar period, which ran from 1919 to 1938) was, as Kindleberger suggests, a transitional period. The old hegemon, Great Britain, had lost the capacity to stabilize the international system, while the new (latent) hegemon, the United States, did not yet understand the need to take on that role—or the benefits of doing so. The result, of course, was a worldwide depression that did severe damage to almost every major economy. Kindleberger’s analysis of the interwar period was convincing to many. It led to a flurry of additional scholarship and to the emergence of hegemonic stability theory as a widely accepted explanation, not only for the dynamics of the world economy, but also for the dynamics of international relations more generally. Indeed, hegemonic stability theory has been embraced by realists—who represent the long-dominant school of thought in international relations theory—as a general explanation for the existence of cooperation among states through most of the postwar period. In the realist view, the hegemon mitigates the effects of anarchy by acting as the rule enforcer for the international system (as a concept in international relations theory anarchy refers to a situation in which there is no overarching political authority that exists beyond individual states). Without a rule enforcer, states are usually unwilling to cooperate on a sustained and universal basis. Consider, on this point, a simple question: Would people, in general, willingly pay taxes to maintain, say, a police and judicial system if there were no threat of sanctions for  noncompliance? To be sure, some people, perhaps even most, would voluntarily comply. But a significant number of others would not. For the latter group, they obey because there is a clear-cut rule-enforcer—a higher authority in the form of the state—that is capable of compelling compliance when necessary by punishing rule-breakers. (See figure 3.4 for the real-world case of rampant tax evasion in Greece.) The issue is dramatically compounded if a public good (mentioned in the previous chapter) is at issue. A public good is something that is both nonexcludable and nonrivalrous. In simpler language, this means that once a good is created, it is available to everyone, but also that the use or “consumption” of that good by one individual does not reduce its availability to others. Domestic law and order is a public good: once a police and judicial system is in place, everyone benefits from it, even if they did not contribute to its creation; nor does use of the police or judicial system reduce their availability to others. The problem is easy to see: if an individual can benefit from domestic law and order (or any other public good) without paying for it, he will simply not pay. Instead, he will let others pay, and free-ride on their contributions. Free-riding may seem unfair, but it is both common and completely rational behavior. Internationally, free-riding is a major impediment to cooperation. After all, if every state is sovereign, then no other state, group of states, or international organization has the authority to collect taxes or otherwise compel compliance with international agreements and norms meant to produce global public goods.

Hegemonic stability theory, however, posits that a dominant actor can fulfill the role of a higher authority by using its overarching power and disproportionate control of resources. The hegemon, to put it bluntly, can either force or cajole “lesser states” to comply with the rules (of the international political economy). In this regard, it begins to look like “might makes right,” but hegemonic stability theory also presumes that the hegemon generally has more to gain from encouraging voluntary cooperation than from engaging in conflict and violence to compel cooperation. This is especially the case when considering economic issues, such as the construction of a liberal international economic order. This leads to a key question: Why does the hegemon actually play the role the theory describes? We already have a general answer—namely, because doing so brings a net benefit to the hegemon. But another way to answer this question is to examine specific and important cases. One such case is the international monetary system (IMS), which refers to the rules and established practices that facilitate international trade and investment. In the postwar period, the IMS has largely revolved around the Bretton Woods system, the core elements of which stayed in place until 1971. Since we will examine the IMS and Bretton Woods system in some depth later in this book, for now, we will take a broad look at them.

 

Bretton Woods, the Hegemon, and the International Monetary System

At first glance, the construction of the postwar IMS seems to be a very good fit for hegemonic stability theory. The Bretton Woods system, for example, was clearly a product of American power and influence. Delegates from around the world met in the United States to create the system, but there is no doubt that the United States was in charge of the entire process. In this respect, it is worth noting that Great Britain was given an important role to play, but British interests and desires were clearly secondary. U.S. dominance was manifested, in particular, by the adoption of the U.S. blueprint for the IMF, one that defined the IMF not as a world central bank, but instead as a promoter of economic growth through international trade and financial stability (Boughton 1998). The U.S. plan was chosen over a competing one prepared by John Maynard Keynes, the most influential and well-respected economist of the time (the U.S. plan was written by Harry Dexter White). Not surprisingly, the Bretton Woods system codified the U.S. dollar as the international currency. This gave the United States an advantage that few other countries enjoyed. Thomas Friedman (1994) provides an easy-to-understand explanation of this advantage:

Text Box: Figure 3.5. Harry D. White and J. M. Keynes at Bretton Woods

 
Assistant Secretary of the U.S. Treasury, Harry Dexter White (left) and John Maynard Keynes, honorary advisor to the UK Treasury at the inaugural meeting of the International Monetary Fund's Board of Governors in Savannah, Georgia, United States, March 8, 1946.

Image Source: The image is in the public domain in the United States because it is a work prepared by an officer or employee of the United States Government as part of that person’s official duties under the terms of Title 17, Chapter 1, Section 105 of the U.S. Code.
[The United States] prints green paper with George Washington’s and Ben Franklin’s and Thomas Jefferson’s pictures on it. These pieces of green paper are called “dollars.” Americans give this green paper to people around the world, and they give Americans in return automobiles, pasta, stereos, taxi rides, hotel rooms and all sorts of other goods and services. As long as these foreigners can be induced to hold those dollars, either in their mattresses, their banks or in their own circulation, Americans have exchanged green paper for hard goods. (Cited in Cohen 1998, p. 124.)

At the same time, the system of rules and institutions—including the establishment of a pegged rate or adjustable peg currency regime, the gold-exchange standard (GES), and two international financial institutions, the IMF and the World Bank—was designed to ensure general stability in the international monetary system, a benefit to all. Significantly, in the very early postwar period, international monetary relations remained shaky, in part because the IMF lacked the resources necessary to deal with the financial demands following the end of World War II. Fortunately, by the end of the 1940s, the United States had become willing to shoulder much more of the burden for maintaining global monetary stabilization. Cohen (2001) notes that American hegemony was exercised in three ways. First, the United States itself maintained a relatively open market, giving rebuilding economies a place to sell their goods. In fact, the United States not only maintained an open market, but also allowed some countries, especially Japan, one-way access to the U.S. market (that is, while Japan was given access to the U.S. market, the U.S. did not demand reciprocal access to the Japanese market). Second, the United States provided significant long-term loans; initially, this was through the Marshall Plan and related programs, and later funding went through the reopened New York capital market. Third, “a liberal lending policy was eventually established for provision of shorter term funds in times of crisis” (p. 97). U.S. policymakers, Cohen also points out, did not necessarily intend to take on a hegemonic role, but once that happened, “they soon came to welcome it for reasons that were a mixture of altruism and self-interest” (ibid.).

 

Criticisms of Hegemonic Stability Theory

Text Box: Table 3.1. Marshall Aid Recipients, Amounts
Country		Cumulative (U.S.$ millions
Austria	 	468
Belgium	 	777
Denmark	 	385
France	 	2296
Germany	 	1448
Greece	 	376
Iceland	 	43
Ireland	 	13
Italy	 	1204
Netherlands	 	1128
Norway	 	372
Portugal	 	70
Sweden	 	347
Switzerland	 	250
Turkey	 	137
United Kingdom	 	3297
Source: Schain (2001)
In a detailed analysis of the relationship between hegemony and the international monetary system, Eichengreen (1987) concluded that hegemonic stability theory is helpful for understanding the smooth operation of the early Bretton Woods system, and it is also useful in explaining some of the difficulties in the interwar period (p. 57). At the same time, he argues that hegemony alone only tells part of the story. One reason is clear: hegemony is essentially a transitory phenomenon (p. 58). On this point, Corden (1990) asserts that hegemony typically lasts only two or three decades. And while there is a great deal of debate over the starting and end points of hegemony,[3] if Corden is correct, this means that there must be other factors and forces at play to explain why, for example, the postwar international monetary system remained relatively stable even after American hegemony was lost in the early 1970s. Not surprisingly, Corden himself has an argument: the international monetary system has worked, he writes, “because on the monetary side it is more like a ‘nonsystem,’ every country doing what it chooses, though in some consultation with the other six [members of the G7] … with only intermittent acts of coordination, especially with regard to exchange-rate intervention” (p. 19). This statement clearly suggests that the stability of the international monetary system was perhaps never dependent on hegemony; rather, it was based, at most, on a loose collaboration among major economic powers. In this scenario, the United States might still be the dominant player, but it is not U.S. dominance per se that ensures stability.

            Others do not go quite as far as Corden. I already mentioned Robert Keohane (1984), who argued that hegemony—while on the wane—continued to exert influence in the international political economy into the 1980s (and beyond). More importantly, he and others have argued that international systems (such as the international monetary system) do exist, and are partly dependent on hegemony, but also partly (really, significantly) dependent on international regimes. One of the most oft-cited definitions of regimes is by Stephen Krasner, who defined a regime as a set of explicit or implicit “principles, norms, rules, and decision-making procedures around which actor expectations converge in a given issue-area (1983, p. 1).[4] Together, Keohane, Krasner and others helped develop regime theory, which provides an alternative to HST. Regime theory is premised, in part, on the idea that shared interests lead to voluntary cooperation among states, and that over time, cooperation on certain issues can become embedded through the creation of regimes. Regimes, it is important to note, serve three key purposes: (1) they provide information about the behavior of all participating states, usually through monitoring and self-policing; (2) they reduce the costs of future agreements (i.e., a regime makes it unnecessary for states to repeatedly negotiate over the same issues); and (3) they generate the expectation of cooperation among members. There is, of course, much more to regime theory, but the main point is that regimes make long-term cooperation possible without a hegemon.

            Another, less evident, criticism revolves around the tacit ethnocentric bias of hegemonic stability theory. This bias becomes particularly evident when the theory is applied to the United States. To a significant extent, advocates of hegemonic stability theory portray the United States as a “benevolent” hegemon, which suggests that much of what the United States does as hegemon is based less on self-interest, and more on nobility and largesse—doing good for the whole world, even if the rest of the world is ungrateful or too self-serving to realize this. Robert Gilpin, one of the best-known advocates of hegemonic stability theory, for instance, blithely wrote, “Societies freely enter into extensive market relations only when the perceived gains are much greater than the perceived costs or when the market relations are forced on them by a superior society” (emphasis added; 1981, p. 129). Grunberg (1990), moreover, argues that hegemonic stability theory has “a built-in, ethnocentric bias simply in the sense that it links the fate of the world with the United States “ (p. 247). This suggests that the world needs the United States to fulfill its destiny as the most powerful and dominant state, and to question, still less challenge, U.S. dominance only invites a self-defeating struggle.

            Closely tied to the criticism of ethnocentric bias is the contention over what constitutes a global public good. We are told by advocates of hegemonic stability theory that the international system of free trade, the international monetary system, and a neoliberal economic order more generally, are all clear-cut examples of public goods. Critics charge, though, that viewed “from below”—that is, from the perspective of poorer, less industrialized, and less privileged countries—these “goods” are primarily good for the richest and most powerful countries, and especially for the United States. They tell us that we must question the often-unquestioned assumption of what constitutes a public good at the international or global level (Snidal 1985, p. 613). Free trade, for example, does little to help the poorest countries that have extremely limited capacity to compete effectively in international markets. This is true even when comparative advantage is taken into account. After all, the poorest countries are competing against other poor countries on the basis of low labor costs, and the richest countries continue to produce a full range of products and services irrespective of their comparative advantages. To top it off, when richer countries selectively protect their most vulnerable markets from international competition—as the United States does with many of its agricultural markets (see figure 3.6, “U.S. Farm Subsidies”)—the notion of hegemonic benevolence becomes extremely problematic.

Text Box: Figure 3.6. U.S. Farm Subsidies (2005)

 

This graph shows the level of subsidies provided to U.S. agricultural industries in 2005. Subsidies are a form of protectionism, as they lower the cost of production for producers of a particular good, making it more difficult for foreign producers to compete.

Image source: The image is licensed under the Creative Commons Attribution-Share Alike 3.0 Unported license.
            There are other criticisms as well. For our purposes, however, it is enough to understand that, as appealing as the theory is on first glance, closer analysis reveals potentially significant flaws. This is not to say that hegemonic stability theory should be dismissed. Far from it. Instead, as with any theoretical approach (including the three traditional approaches we covered in the chapter 2), the key assumptions, principles, and insights should be evaluated with a critical eye. We should also remember that the international or global political economy is not an event, but an ongoing process. This means, in part, that dynamics within the system can, and likely do, vary over time. Thus, it could very well be the case that hegemonic stability theory explains important aspects of international economic stability in certain situations, but not in others. Or, equally likely, it could be that stability is a product of multiple independent factors. On this point, recall our discussion in chapter 1 on the distinction between international political economy and global political economy. In IPE, the focus is on what states do. In GPE, by contrast, analysts understand that states are only one of many actors, and that the basis of state power (which lies largely in the security structure) is neither the only, nor necessarily the most important, basis for power in the world economy. These last points provide a useful segue into the next section, a discussion of post-hegemonic theories.

Post-Hegemonic Theories

In the 1970s and 1980s, when hegemonic stability theory was at its peak, many (but certainly not all) IPE scholars took for granted that states were necessarily the key players in the world economy. It is no surprise, then, that a lot of analytical attention was paid to the most dominant state—the hegemon. We might say that the world has changed since then, although it is probably more accurate to say that there has been a change in the ways in which many scholars and other observers view the world. Whatever the case, one thing is fairly clear: states are not the only significant actors, and even the most powerful of states (i.e., the hegemon) has a limited capacity to influence and shape the global political economy. This is a starting point for post-hegemonic theories (PHTs)—a vague and potentially confusing term. To simplify our terminology, then, we can say that PHTs, most generally, are theories that pay serious attention to actors or entities other than the state. These other actors include international organizations (or multilateral institutions), international regimes, corporations, nongovernmental organizations, epistemic communities, social movements, and so on. There is also a much stronger emphasis on the nonsecurity or nonmilitary sources of power, especially power in the production and knowledge structures, but also including intangible or intersubjective sources of power such as ideology, culture, norms, and values.

Among the many types of actors, one area of particular interest has been multilateral institutions, some of the most prominent of which are the United Nations, the International Monetary Fund (IMF), the World Bank, the World Trade Organization, and the European Union. We might legitimately ask: Aren’t these institutions just proxies for state power, and don’t they simply reflect the interests of states? The short answer is: yes and no. On the one hand, multilateral institutions are unequivocally the product of state action and interests. Even more basically, their membership is comprised entirely of states. In this view, then, there is no reason to focus on the institutions themselves, since they simply do what the states, and especially dominant states, want them to do. Consider, on this last point, how power is exercised in the IMF: voting power is linked explicitly to financial contributions (called quotas) from member states; the higher a state’s contribution, the more votes it gets (see figure 3.8, “Distribution of IMF Quotas by Income Group”). Quotas, in turn, are based on a member’s relative economic position in the world.[5] Even in the United Nations, the most important decisions are left to just five countries—the United States, Russia, China, France, and Great Britain—who all retain permanent seats on the Security Council, and who can exercise veto power on any decision made by the Security Council. From a Marxist perspective, then, we might say that institutions are the tools of the dominant states. (It is important to recognize, however, that not all Marxists would agree with the foregoing statement, since social classes, not states, are the dominant actors.)

 

Figure 3.8. Distribution of IMF Quotas by Income Group, 1948–2004

This graph shows the percentage in quotas/votes for the U.S. individually, richer industrialized countries (except the U.S.) as a group, and developing countries as a group. Although the U.S. share has declined significantly over the years, it is still the largest single contributor by far; moreover, the richer (mainly Western countries) still have well over 50 percent of total votes. In fact, in 2011, just eleven countries (the U.S., Japan, Germany, the UK, France, Italy, Canada, Netherlands, Belgium, Switzerland, and Australia) accounted for more than 50 percent of total votes, out of a total membership of 187 countries.

 

Source: The graph is reproduced exactly from Blomberg and Broz (2006), p. 20.

 

On the other hand, once created, multilateral institutions often—although not necessarily—begin to take on a life, and interests, of their own. That is, they begin to operate, at least to some extent, as independent actors in their own right. Thus, while the United States and other rich industrialized countries may have preponderant voting power in the IMF, this does not automatically mean that everything the IMF does reflects U.S. power. Among scholars who focus on multilateral institutions, there is a great deal of debate regarding the autonomy—or lack thereof—on the part of multilateral institutions, but almost all would agree that institutions, at a minimum, serve a vital function as mechanisms for coordination, collaboration, and cooperation at the international level. Whatever the case, institutions are an increasingly important part of the world economy, and an increasingly important reason—many scholars argue—for international stability, in both the economic and political realms. With this in mind, let’s take a closer look at the role of institutions.

 

The Role of Multilateral Institutions

            Theoretically speaking, institutions (and regimes) fit comfortably within the liberal tradition, although institutionalists can be found in a range of theoretical camps within IPE. With this caveat in mind, in the liberal tradition, there has long been an emphasis on pluralism and the potential for cooperation (as opposed to conflict). The key question, from this perspective, has centered on how the interests of multiple actors, including both state and nonstate actors, can be reconciled in a manner that provides stability and mutual benefit to all. Or in simpler terms: How is cooperation possible? Before answering, I should emphasize that liberals and others who focus on institutions must deal with the issue of anarchy, a key concept in international relations theory. Anarchy, most simply, is the absence of overarching political authority within a particular political system, such as the international system today. The existence of anarchy makes cooperation at the international or transnational level very difficult to achieve, since all states are ostensibly sovereign, and therefore of equal standing in the international community. In hegemonic stability theory, this problem is resolved by the existence of a hegemonic state. In some versions of post-hegemonic stability theory, by contrast, the problem is resolved through multilateral institutions. So what exactly do these institutions and regimes do?

What Do Institutions Do?

Text Box: Figure 3.9. Criteria for Joining the European Union
 

According to the European Union’s website, any European country may apply for membership if it respects the democratic values of the EU and is committed to promoting them. More specifically, the EU lays out three key criteria for membership:

§	Stable institutions guaranteeing democracy, the rule of law, human rights and respect for and protection of minorities.
§	A functioning market economy and the capacity to cope with competition and market forces in the EU.
§	The ability to take on and implement effectively the obligations of membership, including adherence to the aims of political, economic, and monetary union.

In addition, the EU has specific rules in 35 different policy areas that are negotiated separately. Candidates for membership must agree on how and when to adopt and implement rules in these 35 areas, and must provide guarantees on the date and effectiveness of the measures they intend to complete in order to satisfy the negotiated requirements. (Source: http://ec.europa.eu/enlargement/policy/conditions-membership/index_en.htm)

The image of the EU flag is in the public domain.
Most generally, international institutions help to create a stronger, more durable basis for trust between and among states (trust can be defined as a belief in reciprocal cooperation—i.e., “If I do x, then you will do y”). Trust is a major component of cooperation. Or, to put the issue in negative terms, we can say that the greatest obstacle to international cooperation is distrust, or the prospect of cheating, by other states. International institutions help to alleviate the problem of distrust by providing a forum in which the intentions of various countries are revealed, and by providing a mechanism for monitoring, reporting, and policing the activities of all participants, and, in some cases, adjudicating disputes. In principle, these functions could be carried out without institutions, but in practice this is extremely difficult to achieve when more than a handful of states are involved. Thus, institutions provide, perhaps, the most (and often only) viable means for effective and efficient trust-building in world politics. In particular, institutions allow for regularized interactions. Regularized interactions raise the level of communication, increase the flow of information, and generally provide a stronger basis for sustained interaction. Institutions can also significantly reduce the likelihood of free-riders. This is true even where there is no effective enforcement mechanism (which is the case in the large majority of international institutions). Instead of enforcement, for example, institutions may require minimum standards for membership: that is, before a state is allowed to join, it must agree to, and make concrete progress toward, specific benchmarks (see figure 3.9, “Criteria for Joining the European Union”). To put it more simply, the motto of some international institutions might be: “No Free-Riders Allowed.” The effectiveness of entry requirements presupposes clear benefits to membership in an international institution.

            Rather than continue this discussion at an abstract level, we will consider one of the more effective and important international institutions—the World Trade Organization (WTO). The WTO is a descendant of the General Agreement on Tariffs and Trade (GATT). The most salient difference between the WTO and the GATT is that the former is a permanent or standing international institution, while the latter was a series of trade negotiations (or “trade rounds”) that began in 1947. The GATT, however, was not originally intended to be the main multilateral forum for the discussion of international trade. Another organization, the International Trade Organization (ITO) was supposed to play this role, but concerns, especially within the United States, on how the organization might infringe on domestic economic matters, led to the ITO Charter never entering into force. While the ITO floundered, the GATT succeeded in liberalizing international trade, primarily through tariff concessions: the first round of negotiations resulted in a package of trade rules and 45,000 tariff concessions affecting $10 billion of trade, which was about one-fifth of total world trade at the time (WTO, n.d. [a]). Subsequent rounds also primarily focused on tariff concessions, but in the Uruguay Round (1986–1994) a much broader array of issues was negotiated, including the establishment of the WTO in 1995. The Uruguay Round was a mixed bag: there was obvious success in the end, but achieving that success took nearly a decade, which included several periods of near collapse in talks (1988–89).

 

Table 3.2. GATT Trade Rounds

 

Year

Place/name

Subjects covered

Countries

1947

Geneva

Tariffs

23

1949

Annecy

Tariffs

13

1951

Torquay

Tariffs

38

1956

Geneva

Tariffs

26

1960–1961

Geneva
Dillon Round

Tariffs

26

1964–1967

Geneva
Kennedy Round

Tariffs and anti-dumping measures

62

1973–1979

Geneva
Tokyo Round

Tariffs, nontariff measures, “framework”
agreements

102

1986–1994

Geneva
Uruguay Round

Tariffs, nontariff measures, rules, services, intellectual property, dispute settlement, textiles, agriculture, creation of WTO, etc.

123

 

Source: WTO (http://www.wto.org/english/thewto_e/whatis_e/tif_e/fact4_e.htm)

 

            Since its creation, the WTO has not only continued the work of the GATT, but has also created a much stronger basis for trust building and cooperation. One of the main developments in this regard is the Dispute Settlement Panel (DSP), which has the power to resolve trade disputes between and among member countries. While not an enforcement mechanism per se, it provides the basis for ensuring reciprocity in a manner that few other international institutions can match. More specifically, if a state is found to have broken a multilateral trade rule, the WTO allows affected states to legally retaliate by imposing countervailing tariffs. Not surprisingly, more than a few states have threatened to withdrawal from the WTO when DSP decisions have gone against them. Significantly, though, none have (Balaam and Dillman 2011, p. 143). Other countries, including the United States, have refused to publicly acknowledge wrongdoing in the face of an adverse WTO ruling, but have nonetheless complied. In 2003, for example, the Bush administration dropped duties on imported steel (the United States referred to these as “safeguard measures”) following a decision by the DSP that the duties were illegal. Interestingly, the U.S. trade representative at the time, Robert Zoellick (who would later become president of the World Bank Group from 2007–2012) argued that the decision had been made independently of the WTO ruling; indeed, at no time did the United States ever admit it had breached WTO rules (BBC News 2003). If the U.S. had not dropped its duties, however, the affected countries were ready to retaliate legally—that is, with the blessing of the WTO.

Text Box: Figure 3.10. United Nations Climate Change Conference, Doha, Qatar, 2012
 
Source: UNclimatechange. Licensed under the Creative Commons Attribution 2.0 Generic license.
None of this is to say that the WTO has solved all issues in international trade. It has not. Recent WTO trade rounds have seen deep fissures develop, especially between wealthy and poorer countries on the issues of agriculture, services, intellectual property rights, and other areas.[6] These fissures have revealed the limits of cooperation and trust within even a strong institutional framework when the interests of participants are very far apart. At the same time, without an institutional framework, many broad-based cooperative arrangements could likely never be created. International trade, in this regard, provides a good case in point. There is, however, another issue that provides an even better example: global climate change.

 

Institutions and Global Climate Change

Global climate change is an immensely complicated and controversial issue, both scientifically and politically. Building consensus and cooperation among states, therefore, has been a monumental task. Nonetheless, once policymakers in most states—influenced, in no small part, by an epistemic community made up of climate scientists and others (an epistemic community, according to Peter Haas, is a network of professionals with expertise and competence in a particular domain and an authoritative claim to policy-relevant knowledge [1992, p. 3])— concluded that global warming represented a real and significant threat, they were able to use existing institutional arrangements, especially the United Nations, to address the issue on a collective, rather than individual, basis. More specifically, multilateral negotiations designed to address global climate change began in 1990 under the auspices of the UN General Assembly; by May 1992, a convention on climate change was adopted at the UN Conference on Environment and Development (popularly known as the Earth Summit, in Rio de Janeiro, Brazil). Negotiations continued through several institutional bodies, leading to the adoption of the Kyoto Protocol in December 1997. The protocol entered into force in February 2005 after the requisite minimum of 55 parties had deposited their instruments of ratification, acceptance, and approval of accession. The protocol required monitoring and recording of CO2 emissions, and for some states (referred to as Annex I Parties) a specified reduction in greenhouse gas emissions. There were many countries that delayed ratification; ultimately, however, the only major holdout was the United States, which accounted for 36 percent of all 1990 emissions. The noncooperation of the world’s largest emitter of greenhouse gases—and therefore the world’s biggest free-rider—should have doomed any compliance with the protocol, but it did not. Indeed, among the 36 Annex I Parties committed to an actual reduction in greenhouse gas emissions, the majority did achieve lower emissions.[7] Moreover, the Kyoto Protocol, which was set to expire at the end of 2012, was extended to 2020 during the Doha Climate Conference.[8]

The foregoing summary obscures the complexity and extraordinary difficulty of reaching agreement on climate change. As Haas (2000) points out, the first scientific warnings of global climate change appeared in the 19th century, but even after renewed scientific warnings occurred in the 20th century, it took 23 years to finally develop the first, admittedly weak, international measures (p. 567), which were embodied in the UN Framework Convention on Climate Change (UNFCCC) in 1992. In this regard, it is easy to dismiss the importance of international organizations, but once the UNFCCC was established, international organizations shifted from being passive arenas for negotiations to independent actors. In particular, the United Nations Environment Program (UNEP), according to Haas, was “able to translate new information emanating from the scientific community into effective policy-oriented programs,” while its executive heads became “vigorous exponents of environmental protection and research in public, in private with heads of state, and also in private negotiations.” Even more, Haas continued, “[t]hey were generally able to effectively cope with disagreements among member states and avoid institutional deadlock” (2000, p. 568). International organizations carried out a number of other functions as well: they encouraged the dissemination of innovations to other actors; they set the agenda for member states; distributed information; built national monitoring and research capacity; helped industry and societal groups identify new practices; trained and assisted governments to enforce international commitments; structured bargaining forums; and empowered new national and transnational political coalitions (Haas 2000, p. 571).

A Transnational World Full of Transnational Actors

Despite the focus on institutions, the foregoing discussion still seems to depict a state-centric world. Appearances, however, can be deceiving. The WTO, for instance, can be considered an independent actor insofar as its decisions do not merely reflect the interests of dominant states, and especially the hegemonic state. The ruling against the United States on steel imports is a good indication of this. Similarly, the Kyoto Protocol seemed to contradict U.S. interests (although, it should be noted that the protocol was endorsed by the executive branch but rejected by the legislative branch). While the United States failed to ratify the protocol, not only did the institutional framework that made the protocol possible survive, but so too did the protocol itself. It is also important to emphasize, on this last point, why the U.S. legislature refused to ratify the Kyoto Protocol. The main reason can be boiled down to domestic politics. This means, most generally, that it wasn’t the United States as a monolithic state actor making the decisions, but rather individual politicians responding to pressures from industry-based interest groups, oil companies, and other corporations that rely heavily on traditional sources of energy. Even more, the issue became part of a partisan political struggle, with support for climate change legislation becoming identified with the Democratic Party. This meant that Republicans, even those who might have been concerned with climate change in the past, chose sides based on partisanship. There were additional actors playing a role as well, especially environmental NGOs, who were strong advocates of ratification.

To say that domestic politics helps explain the U.S. rejection of the Kyoto Protocol suggests that the actors involved were purely domestic actors involved in domestic-only politics. But this was hardly the case. (We will look more closely at domestic politics below.) The key nonstate actors clearly had interests and concrete connections or relationships that transcended national borders, and even without such interests or connections, the consequences of their actions have had ramifications beyond the borders of the United States. As such, these actors were all involved in “international relations” in almost the same way that states are. Yet, because these actors are not states, it is more accurate to say that they were involved in “transnational relations.” Thus, they are transnational actors, a term that is preferable to nonstate actor, as the latter suggests that states are naturally dominant (Willets 2001, p. 357). In recognizing that transnational actors can influence and even dictate the policy of states, we automatically enter a new, far more pluralistic world. In the state-centered world, there are around 200 actors (states), the large majority of which seem to have little impact on the world economy or on world politics. If we disregard the poorest and smallest states, this would leave us with a slate of just a few dozen main actors in the interstate world. In the transnational world, by contrast, there are hundreds, if not thousands, of potential main actors. There are, for example, more than 82,000 transnational corporations with 810,000 subsidiaries spread across the world; thousands of nongovernmental organizations whose activities and interests routinely cross national borders; and dozens of diasporic, epistemic, and other types communities. Of course, just as with states, not every one of these transnational actors has an impact on the world economy or world affairs more generally. But some clearly do. It is hard to discount, for example, the power of global corporations such as Exxon Mobil: this company is one

 

 

 

 

 

 

 

 

 

 

 

Table 3.3. Number of TNCs and Foreign Affiliates, Selected Years and Countries

Country

Year

Parent Corporations

Foreign
Affiliates

United States

1990

2000

2005

3,000

3,235

2,360

14,900

15,712

13,667

United Kingdom

1991

1996

2003

1,500

1,059

2,607

2,900

2609

13,176

Germany

1990

2002

2006

6,984

6,609

5,935

11,821

9,268

9,631

Japan

1992

2001

2006

3,529

3,371

4,663

3,150

3,870

4,500

South Korea

1991

2002

2007

1,049

7,460

7,460

3671

12,909

14,869

China

1989

2002

2005

379

359

3,429

15,966

424,196

280,000

 

Source: Cited in Selden (2010), table 1. Original source data derived from table 1, Saskia Sassen, “The Global City Perspective. Theoretical Implications for Shanghai,” in Shanghai Rising: State Power and Local Transformations in a Global Megacity, pp. 9–10, based on UNCTAD, World Investment Report (1998, 4; 2004, 273, 274; 2008, 211). The figure for China’s foreign affiliates bears further checking.

 

 

of the largest in the world by revenue, with 37 oil refineries in 21 countries that, together, have a combined daily refining capacity of 6.3 million barrels. Another major oil company, Royal Dutch Shell, controls oil production in the Niger Delta region of Nigeria. Since the early 1990s, Shell has been engaged in a struggle with an indigenous population, the Ogoni. While conventional IR theory tells us that the Nigerian state calls the shots, it is fairly clear that Shell has been able to exert tremendous influence over the Nigerian state in its dealings with the Ogoni people. In one particularly infamous case, Shell was implicated in the unlawful execution of nine Ogoni activists; in 2009, Shell agreed to pay $15.5 million to settle a legal action in which the company was formally accused of having collaborated in the execution of the “Ogoni Nine” (Pilkington 2008).

            It is relatively easy to argue that major transnational corporations are important actors in the world economy, but can we make a similar argument about other transnational actors? Let’s consider a rather unorthodox example of an NGO: Al-Qaeda. Mosisés Naím asked, in Foreign Policy (2002, p. 100) magazine, “What does al-Qaeda have in common with Amnesty International and Greenpeace?” The answer, according to the author is that “all three are loose networks of individuals united by a shared passion for a single cause, and thanks to cheaper communication and transportation, each can project globally.” I mention al-Qaeda as an example of an NGO because its influence on global affairs is unquestioned. Yet if this single, poorly funded (relatively speaking) nongovernmental organization can have such a major impact, it stands to reason that other NGOs, even those that use far less violent tactics, can also have an impact. Consider, on this point, a far more innocuous NGO, the International Air Transport Association (IATA). The IATA partners with governments and international organizations to provide an effective regime for navigation, safety standards, and the general regulation of commercial aviation (cited in Willets 2001, p. 372). Clearly, this is an important function in the world economy.

In short, to argue that the power and presence of these actors do not matter—as many traditional IPE and IR scholars do—is becoming less and less defensible. We already covered why this is the case in chapter 1, so I will not go over the same ground here. Instead, let us consider the issue from a slightly different perspective by addressing the question, why has a state-centric approach become increasingly problematic?

 

Problems with the State-Centric Approach

Text Box: Figure 3.11. A Borg from Star Trek

 
The image depicts an individual Borg from the TV series Star Trek. The Borgs are individuals, but they think and act as a single collective unit.

Source: Science Fiction Museum and Hall of Fame. The image is licensed under Creative Commons Attribution-Share Alike 3.0.
I have already suggested one reason for questioning the focus on states—namely, the notion that states are somehow unitary or monolithic actors. Put another way, state-centric analyses tend to take for granted that states are holistic entities, almost literally thinking with one mind and speaking with one voice. This notion depicts the state, at least to some extent, as the Borg, a fictional alien race in the television series Star Trek. The Borg is a collection of individual species that have been thoroughly assimilated into a collective, or “hive.” There are no individual interests or desires—only a single “hive mind” that connects and controls every individual Borg. Although this comparison is admittedly oversimplified and overdrawn, it is not entirely unfair. Analysts who treat the state as a holistic entity presume that the interests of the many specific groups and organizations within any country, no matter how powerful they may be, are largely immaterial to explaining state behavior. In their view, there is only one relevant interest in explaining state behavior: the national interest. Critics argue that this is an unrealistic view of states’ decision-making process. We need to consider what goes on within the state, how different groups influence, shape, and even determine state behavior. Once we admit that there are competing groups shaping state policy, each with varying degrees and types of power, and each with its own interests, the picture becomes more complex, but also more realistic. And, again, when those interests are no longer contained within a single set of national boundaries, we have the beginning of transnational (as opposed to international) theory. One particularly influential argument in this vein is given to us by Anne-Marie Slaughter, who asserts that states, while not disappearing, are “disaggregating into … separate, functionally distinct parts.” These parts, which include the courts, regulatory agencies, executives, and even legislatures, are not only disaggregating, they are also “networking with their counterparts abroad, creating a dense web of relations that constitutes a new, transgovernmental order” (1997, p. 184).

            The second problem with the state-centric approach is the tendency for realists and others to treat all states as essentially equal because all states are sovereign. Specifically, the state-centric approach implicitly treats the most powerful states as the benchmark for all states. Thus, because the United States or China or Germany are still capable of influencing the world economy in major ways, so too must be Costa Rica, Albania, and Zanzibar. Of course, this is a gross exaggeration: no one would argue that Zanzibar has the capacity to impact the world economy to the same extent as the United States. And yet, the state-centric approach still tells us, in principle, that the weakest of states is more important than the strongest transnational actor (see table 3.4 for a comparison of the largest and smallest states). Consider, on this point, the country of Nauru: it has a population of 10,000 people and a GDP of $60 million. Is it reasonable to imply that tiny Nauru is more significant than, say, Exxon Mobil, a company with revenues of $354 billion, profits of $30.4 billion, and assets of $302.5 billion (in 2009)? Exxon Mobil even has eight times more employees—82,100—than Nauru has people. Indeed, even many NGOs exceed some countries in one or more significant dimensions. The Art of Living, an educational and humanitarian NGO, for example, operates in 153 countries and has 2.5 million members. NGOs such as Oxfam, CARE, World Vision, and Save the Children have significant financial strength and large contingents of workers—World Vision, for example, has over 23,000 employees. Politically, too, many of the largest, most well known NGOs have significant clout, at both the national and transnational level. We should not forget, too, that NGOs include criminal, guerilla, and terrorist organizations, and it is almost impossible to deny their (increasing) influence in world affairs. In fact, one tiny, highly dispersed organization, al-Qaeda, arguably has had a larger impact on world affairs than all but the most powerful states over the past decade or so.

 

Table 3.4. Differences between Largest and Smallest States, Selected Indicators

 

Country

Largest

Measure

Smallest

Country

China

1,339,190,000

ß Population à

10,000

Nauru

China’s population is almost 134 thousand times the population of Nauru.

United States

$14,991,300,000,000

ß GDP à

$37,000,000

Tuvalu

U.S. GDP is 405 thousand times larger than Tuvalu’s GDP.

United States

$711,000,000,000

ßMilitary Spendingà

$4,600,000

Gambia

U.S. annual military budget is more than 154 thousand times Gambia’s military budget.

Russia

17,098,242

ßArea (km2)à

1.95

Monaco*

Russia is 8.77 million times larger in landmass than Monaco.

*Vatican City, a sovereign entity, is smaller than Monaco.

 

Transnational Theory

Once we acknowledge that states are not monolithic and that there are dramatic differences among states, on the one hand, and between states and nonstate or transnational actors on the other hand, it becomes clear that we need a framework that can accommodate a full range of actors, relationships, and issues. We need, in short, a transnational theory. A transnational theory requires a multidimensional, structural understanding of power, one that does not privilege military power (or power in the security structure). It also requires an understanding of how power is manifested or exercised in different places, and with respect to different issues: security, finance, the environment, production, trade, human rights, democracy, migration, poverty, and so on. In different places and on different issues, state and transnational actors will play different roles, have varying degrees of influence and interest, and have varied tools at their disposal. In some places and on some issues, states may be dominant, while in other places and on other issues, transnational actors will play the central roles.

Equally important, the distinction between “high politics” and “low politics” must be eliminated. In the past, state-centric approaches classified all nonmilitary, nonsecurity issues as low politics, clearly implying that these were less important issues—and also implying that security was a separate realm, largely disconnected from economic, social, and cultural processes. But this distinction has always been extremely problematic. To see this, all one has to do is consider the fallout of the Great Depression: the consequences were not just widespread economic misery, but the most destructive international war—i.e., World War II—the world has ever witnessed. The causes of World War II are undoubtedly complex, but as David Kennedy (n.d.) writes, the genesis of the war can be found in the “lingering distortions of trade, capital flows, and exchange rates occasioned by the punitive Treaty of Versailles.” To this list, Kennedy adds that a “rigidly doctrinaire faith in laissez-faire, balanced budgets, and the gold standard … added up to a witches’ brew of economic illness, ideological paralysis, and consequent political incapacity as the Depression relentlessly enveloped the globe” (n.p.). Whether or not one agrees entirely with Kennedy’s analysis, one thing is clear: economic policies and (ideological) principles played a key role in the process leading to World War II. Economics, in short, simply cannot be subordinated or dismissed as “low politics.” And what was true in the first half of the 20th century is even truer today, in an era of deeper interdependence and globalization.

Text Box: Figure 3.12. French Support of Mohamed Bouazizi

 

A demonstration in France in support of Mohamed Bouazizi, the “Hero of Tunisia.”
Source. Antonine Walter. The image is licensed under the Creative Commons Attribution-Share Alike 2.0 Generic license.

The example above also highlights a third point: the importance of keeping firmly in mind the inescapable linkage among political, economic, and socio-cultural processes. To a large extent, this is what international/global political economy is all about. These connections have always been present, but they are more acute and pervasive today than ever before. This leads to the final key point in transnational theory—that the increasing interconnectedness of the world must be given full consideration. The old saying, “When the United States sneezes, the world catches a cold” (meaning that what the U.S. does or does not do has an impact everywhere) is still pertinent. But it also works increasingly from the other end; that is, given the high degree of interconnectedness today, seemingly inconsequential events in “faraway” places can have global repercussions. One of these “faraway” places is Tunisia, a place the large majority of Americans could not find on a map. On December 17, 2010, however, the action of a single person led to an extraordinarily significant series of events that has affected the entire world. On that date, a street vendor named Mohamed Bouazizi set himself on fire as a protest against police harassment. His act of self-immolation led quickly to widespread anti-government protests in Tunisia, and ultimately to the ousting of President Zine El Abidine Ben Ali only 28 days later. Ben Ali had ruled Tunisia for 23 years. The protests in Tunisia inspired protests and anti-government movements—dubbed the Arab Spring—throughout the Middle East and North Africa (MENA): in Algeria, Yemen, Egypt, Bahrain, Syria, Lebanon, Libya, and elsewhere. Interconnectedness, in this case, created a snowball effect that, to a significant extent, reached almost every part of the globe. The European Union and the United States, in particular, were drawn into several conflicts, the most salient of which was Libya. In Libya, the EU committed military forces, via a no-fly zone, on behalf of anti-government forces. The result was the capture and death of Muammar Gaddafi in 2011—a man who had ruled Libya for more than 40 years. Of course, Bouazizi’s action was not the cause of the Arab Spring; there were many other factors. But his action is a clear demonstration of how interconnected and transnational the world has become.

Summing Up

For our purposes, a full-blown discussion of transnational theory is not necessary. For now, it is enough to keep in mind that, as reasonable as it may seem, an exclusive focus on states is shortsighted and even wrongheaded. In many cases and on many issues, states are only one of several important actors, and not necessarily the most important. Keep in mind, too, that the very concept of the state as a unified and rational actor that speaks for an entire country is an abstraction. When we talk about the state, in other words, we are not talking about a singular entity making decisions in the same way an individual person does. Instead, a state is a complex and oftentimes disjointed amalgamation of institutions and agencies, of people, of norms and (legal) principles, of ideologies and values, of territory and resources. Thus, we must be careful not to anthropomorphize the state. This is not to say that the state-as-an-actor approach is a bad thing. It is not. Conceptually, it is useful to think of the state as a coherent entity, and many important insights have been derived from this abstraction. At the same time, the abstraction is frequently taken too far, leading analysts to neglect other important actors within a country, and, equally important, what goes on inside of states. This leads us back to an issue we encountered briefly above, namely, the relationship between domestic and international politics.

 

Two-Level Games and IPE

            Earlier I mentioned the importance of domestic politics. My emphasis was on the implicit relationship between domestic and international/global politics. This relationship has been an important area of inquiry among IPE scholars and others, especially those who focus on the formulation of foreign policy. They have even developed a special term for this relationship: two-level games. The general concept of two-level games has been around for a long time, but Robert Putnam, a well-known political scientist, is generally credited with formalizing the term. To Putnam, the opening question is simple: Do domestic politics influence international politics, or do international politics influence domestic politics? His answer is equally simple: “Both, sometimes” (Putnam 1988, p. 427). Putnam’s basic argument echoes much of what we covered in the previous section, but he provides us with a cleaner framework of analysis, one based on a game-playing metaphor.

            In this metaphor, domestic or international negotiations can be depicted as a two-level game. The first level (Level 1) involves the people who actually negotiate an issue, endeavoring to reach an agreement that is satisfactory to all parties. The negotiators might be high-level cabinet officials, heads of state, diplomats, party leaders, union heads, CEOs, and so on. In the simplified model, we presume that each side (there may be many more than two sides) is represented by a single leader or chief negotiator. In Putnam’s model, we also presume that the chief negotiator has no independent policy preference, but is primarily concerned with achieving an agreement that he can sell to his constituents. The second level (Level II) involves separate discussions with various constituent groups, whose support is needed to “ratify” a negotiated agreement (Putnam uses the term ratify generically to refer to any process that is required to endorse or implement a Level I agreement). These two levels are not meant to be “descriptively accurate” (p. 436); Putnam understands that actual negotiations are far more complicated, interactional, and fluid. Still, as with the state, the two levels are useful analytic constructs that allow us to more easily grasp the core elements of a public policy decision. On this point, the requirement that any Level I agreement must by ratified at Level II provides a crucial theoretical link between the two levels.

            Another key element of Putnam’s framework is what he calls “win-sets.” As the term implies, a win-set is the set of all possible Level I agreements that would “win”—that is, gain the necessary majority among the key constituent groups. In this regard, even before (international) negotiators come to the table, they are influenced by domestic factors. The reason is clear: if the chief negotiator goes beyond the range of the win-set, he or she knows that an agreement made at Level I will not be ratified at Level II. This is both a constraint and an advantage. As a constraint, the chief negotiator has little wiggle room for bargaining; however, as a bargaining advantage, the negotiator can use a small domestic win-set as an excuse for a hard-line position: “I’d like to accept your proposal, but I could never get it accepted at home” (p. 440). The relative size of domestic win-sets, of course, will vary considerably depending on the issue being negotiated, but it is safe to say that virtually all issues negotiated at the international level are influenced by domestic win-sets. On this point, too, Putnam puts forth clear guidelines on the circumstances that affect win-set size. He argues that there are three sets of factors:

§  The distribution of power, preferences, and possible coalitions among Level II constituents.

§  Level II political institutions (including constitutional rules and procedures as well as established institutional norms): For example, a two-thirds vote for treaty ratification versus a majority vote; informal consensus building among all major constituent groups (as in Japan); autonomy of the state from social forces (in general, authoritarian states have larger win-sets because Level I decisions do not always require Level II ratification, as they do in democracies).

§  Level I negotiators’ strategies: Individual negotiators can employ strategies to expand or constrict the size of the domestic win-set. For example, in the Carter White House, many inducements were offered to wavering senators to ratify the Panama Canal Treaty.

 

In addition to the three sets of factors listed above, Putnam also discusses three other factors that can affect the relationship between the two levels. These are: (1) uncertainty about the size of the win-set (both the opponent’s and one’s own) on the part of Level I negotiators; (2) international pressures, which reverberate within domestic politics, tipping the domestic balance and thus influencing international negotiations; and (3) the role of the chief negotiator and his or her personal preferences (this could lead to a Level I decision that cannot be ratified and implemented at Level II).

 

Text Box: Figure 3.13. A Highly Simplified Representation of the Two-Level Game

 

This image shows the (partial) interaction between the domestic and international levels. Negotiators in Level I do not act in a vacuum: their positions almost always reflect the interests, concerns, and power of domestic constituents (Level II); this is manifested in the win-set they develop and propose. Once an agreement is reached at Level I, it must then be “ratified” at Level II. This does not always happen—e.g., the Kyoto Protocol was rejected by domestic actors in the U.S.
Image Source: Created by author, but based on similar image in Borsuk et al. (2010), available at http://engineering.dartmouth.edu/sedg/climate_change.html 
There is much more detail and substance to Putnam’s argument than is evident in this summary, but the basic point should be clear: understanding and explaining how states make decisions requires taking into account the “entanglements,” as Putnam puts it, of international and domestic politics. The two-level games approach is one way to do this because it recognizes that “central decision-makers strive to reconcile domestic and international imperatives simultaneously” (p. 460). We should also note, as Putnam does, that his two-level approach can accommodate, or “in principle be married to such diverse perspectives as Marxism, interest group pluralism, bureaucratic politics, and neo-corporatism” (p. 442). This makes Putnam’s two-level approach more a method than a theory, but this is not a problem. Indeed, I recommend that, as we move through our discussion of specific issues in international/global political economy in subsequent chapters, you see how this two-level method can be usefully “married” to the traditional and contemporary theories of IPE.

 

Constructivist Approaches to IPE

Our theoretical discussion thus far has centered on approaches that, for the most part, assume that the world’s economic and political systems have a primarily concrete, or objective, existence. This should not be a surprise. It is, after all, just common sense, right? In fact, almost everyone engages in this sort of “theorizing” all the time. General and taken-for-granted statements such as, “Be realistic,” “That’s just the way it is,” or “There’s nothing you can do to change it,” reflect the same sort of assumptions about the world. These statements, in other words, are based on the presumption that there is a fixed, entirely objective reality. This means, in turn, that there are certain things in the world—certain aspects of our reality—that just cannot be changed, no matter how much we would like them to or no matter how hard we try. For much of the 20th century, moreover, social scientists generally subscribed to the same view of the world.

More specifically, most rarely questioned the idea that the world of economics and politics was different, in a fundamental sense, from the world that physicists, chemists, geologists, and other natural scientists focus on in their studies—i.e., the natural or physical world. But consider this difference: while natural scientists study things (material objects), social scientists study, well, human beings. (To be sure, some natural scientists—e.g., biologists—study human beings, too. But they are primarily concerned with the physical processes, structures, and functions of humans as organisms.) More generally, social scientists study the social world. This means that social scientists study human society—the motives, decisions, and actions of individuals, groups, organizations, institutions, and whole countries, as well as (or especially) their interactions. The ultimate objects of study in social science are human beings—beings not only capable of thinking, feeling, understanding, and learning, but also of creating, sustaining, and changing the core elements of the world in which they live. This makes studying the social and physical worlds fundamentally different. Even more, social scientists, as human beings themselves, are an integral part of the world they study, which makes separating themselves from that world exceedingly difficult, if not impossible. This is the starting point for constructivist theories, which, at the simplest level, challenge the assumption of objectivity.

Objectivity vs. Subjectivity

Before continuing with this point, let us take a step back and briefly reconsider the notion of objectivity in the social sciences, particularly in regard to the theories with which we are familiar. Mercantilism, liberal economic theory, and Marxism all operate on the assumption that there are basic, largely invariable laws that govern human society. One of the main goals of each theory, then, is to identify and understand these laws so that we can explain and predict how the world works. Thus, liberal economics tells us that human behavior can be reduced to self-interest, a characteristic that is unchanging across time and space. Mercantilism, in turn, assumes a world of unremitting struggle for power and security among self-interested, sovereign states, while Marxism posits that material forces largely determine the type of world we live in. Built into each of these assumptions is the firmly held conviction that our theories of the social world are completely separate from that world. Put another way, in an objective view of the world, we assume, for instance, that what we think about how states interact with each other at the international level has nothing to do with how they actually interact. Again, we are told, this is only common sense. This view, however, has been challenged by an increasing number of scholars, including natural scientists. Indeed, it is within the “hardest” of natural sciences—physics—that many cogent arguments can be found. As Niels Bohr, a Nobel Laureate in physics, said, “It is wrong to think the task of physics is to find out how nature is. Physics concerns what we can say about nature …” (emphasis added, cited in McCloskey, 1994, p. 41). Even more to the point, Werner Heisenberg wrote, “Natural science does not simply describe and explain nature; it is part of the interplay between nature and ourselves: it describes nature as exposed to our method of questioning” (emphasis added; cited in ibid.). Of course, Bohr and Heisenberg are speaking of a field in which the primary objects of study are not human beings. In the social sciences, then, we can expect that the interplay between our reality and ourselves will likely be even more important.

What does this mean in more concrete terms? To find out, let’s revisit an example we considered earlier—namely, the primary manner in which states interact with one another. There are many ways to think about this, but mercantilism (and realism) tells us, as we have seen several times already, that the nature of interstate relations is defined by a constant struggle for power and security. In this view, our understanding of the world (of international relations) tells us that no state can let down its guard, even for a minute. The world is “Darwinian”: only the strongest survive and prosper. Thus, if political leaders in Country A believe this view of the world, they will naturally put it into practice. This means that they will build a strong military, create and produce destructive weapons, be suspicious of neighboring countries and peoples, and so on. When other countries see what Country A is doing, they do the same. We then come full circle: in Country A, the original views of the political leaders are seemingly confirmed as neighboring countries build up strong (and threatening) armies. This means Country A must redouble its efforts, which leads to another round of military build-up, and then another, and so on. This is a simplistic example, but one that nicely illustrates how our theories of the nature of international relations can lead to a self-fulfilling prophecy (Wendt 1992). The end result is, in fact, an objectively dangerous reality. But it is a reality that is in large measure socially constructed; that is, it is a reality that is fundamentally premised on—and continuously reproduced via—our ideas (beliefs, perceptions, and theories) as much as, if not more than, on objective conditions.

On first glance, all of this may sound terribly—and dangerously—naïve, as critics of constructivist theory have repeatedly pointed out. On this point, though, just think about two real-world cases. The first case is U.S.-Soviet relations during the Cold War. Both sides operated on the assumption that relations between capitalist and socialist states were unavoidably conflictual; thus, it was necessary to do everything possible to protect against the threat the other side presented. In the 1980s, however, something strange happened: within the Soviet Union, the once firmly held belief that the West represented an existential threat began to break down (Wendt 1992, p. 420). Ultimately, this lead to a radical shift in Soviet policy—and a voluntary dissolution of the Soviet Union—based on a very different understanding of what the United States represented. Instead of viewing the United States as a threat, the Soviet leadership began to see the possibility of a peaceful relationship. U.S. reassurance further reinforced these views, and a new type of relationship was born, a relationship symbolized by the ending of the Cold War. The second case is the formation and development of the European Union, a process that began rather modestly with the formation of the European Coal and Steel Community (ECSC) in 1950. The motivation for the ECSC, however, was more than just economic: it was widely viewed as the first concrete building block for pan-European cooperation, integration, and peace. In this regard, it represented a belief that Europe could construct a new postwar reality. To a very significant extent, this new reality was created, as war and violent conflict among the major Western European countries, thus far, has been completely eliminated since the end of World War II.

The cases of U.S.-Soviet relations and the European Union suggest that socially constructed changes only work in one direction. This is far from the case. One counter-example should suffice to make this point. In 2002, George W. Bush introduced the phrase “Axis of Evil” in his State of the Union address. This was part of a long process of constructing certain countries as existential threats, not only to the United States, but also to the entire world. The three members of this evil axis were Iran, Iraq, and North Korea. Significantly, none of these countries had or has world-beating military capacity, and although all three were accused of attempting to build nuclear weapons—which North Korea succeeded in doing—there are other countries with significant nuclear arsenals, such as Pakistan and China, that were and are deemed less threatening. This tells us that objective military power does not determine what is dangerous and what is not. Identifying a country as an existential threat, however, must be reinforced in practice. The United States did this with Iraq until the 2003 invasion, and continues to do so with Iran. The construction of Iraq as a mortal threat worked exceedingly well: most Americans and much of the world believed that Saddam Hussein’s regime was a real danger, and this allowed the Bush administration to carry out an unprecedented preemptive war against what was, in essence, a state with little actual or potential military capacity. In the case of Iran, distrust of the regime is palpable and has been so for a very long time. Yet, this did not have to be the case: shortly after 9/11, for example, Iran extended an olive branch to the Bush administration, but this diplomatic gesture was summarily rejected. Instead of pursuing the possibility of a better relationship, the Bush administration further aggravated the hostility between the two countries.

Constructivist Theories and IPE

Text Box: Figure 3.15. Income Inequality in OECD Countries
 
This chart shows the Gini Index numbers for all OECD countries. The key point to note is the significant differences among the countries, which indicates that, while some inequality cannot be avoided, there are many things individual countries can do to effect the overall level of income inequality.
Source: OECD (http://www.oecd.org/social/inequality.htm)
How does this relate to an understanding of the international or global political economy? Most saliently, it strongly suggests that so-called objective theories do not necessarily describe a reality that has to be, but instead describe a fundamentally malleable reality. Marxism, for instance, argues that class struggle in an unavoidable aspect of capitalist reality, and that capitalism necessarily produces severe exploitation, alienation, inequality, and so on. In the Marxist reality, these social problems can be overcome, but only with the complete collapse of the capitalist system; reform, in other words, is not possible. Constructivist theories recognize that capitalism has very real, very serious consequences; they also recognize that capitalism is a deeply embedded, extremely powerful structure. Yet they leave open the possibility of significant change to this structure through purposeful collective action. This is precisely what many of the anti-globalization protestors are trying to do in their actions against global neoliberalism: they are challenging the idea that the values of the (neoliberal) market—e.g., efficiency, deregulation, unfettered competition—should take precedence over other values such as equity, social and environmental justice, and so on. It is easy to dismiss these efforts as meaningless noise, but consider this reality: in the world today, there are already several types or varieties of capitalist systems that produce very different results—in terms of equity and social and environmental justice—for their societies. Among the wealthiest capitalist economies, for example, there are significant differences in terms of income inequality (see figure 3.15, “Income Inequality in OECD Countries”). Scandinavian and some Eastern European countries have extremely low levels of income inequality, while others—most notably the United States, Mexico, and Turkey—have very high levels of income inequality. This is not an accident. There are also significant differences, to cite one more example, in terms of the level of state intervention in the national economy. In some countries, the state plays a direct and ubiquitous role in the economy (e.g., China), while in other countries, the state’s role is much more limited (although never absent). There are many other important and complicated issues related to how capitalist systems vary; for now, though, just keep in mind that the reality of capitalism is socially constructed.

Constructivist approaches, to repeat, tell us that there are always different possibilities. Put another way, they tell us that current realities, because they are not God-given or somehow predetermined by human nature, are historically contingent social structures. This means, in part, that they emerged through an interactive process involving a complex mix of material and nonmaterial—or ideational—factors, which includes collective human action. Capitalism, in particular, did not emerge fully formed; instead, it developed and expanded gradually (on a global basis) over a very long stretch of time. Significantly, too, the early development of capitalism took place hand-in-hand with the emergence and development of the (modern) national state in Western Europe. The combination of capitalism and a national state proved to be highly effective in making war[9]—an extremely important attribute during a period of almost constant warfare in Europe. Both capitalism and the national state, it is crucial to add, required an ideational base before they could thrive. On this point, it is worth remembering that Adam Smith was a philosopher more than an economist, and his idea that the pursuit of self-interest was good for society as a whole—and not just the self-interested individual—was a crucial part of his overall argument about the virtues of capitalism. The ideational basis of the national state is even clearer. Consider, on this point, the following questions: Why do people give their allegiance to a state? Even more, why are people willing to give their lives for a state? The answer to both questions, most simply, is nationalism. Yet nationalism is an ideology, a belief that one belongs to a particular political community. This sense of belongingness, in turn, acts as a deep source of collective mobilization, collective responsibility, and—when necessary—collective violence. This is a very complex discussion, so suffice it to say that nationalism is a powerful ideological force in the world.

The dual development of capitalism and the national state enabled Western Europe to dominate the rest of the world, and eventually to impose that system on a global basis. Again, this was not a predetermined outcome, but was contingent on a host of material and ideational factors. Once this Eurocentric structure took hold, however, it became the picture of reality for much of the world, a picture that defines, but does not wholly determine, the parameters of our present existence. It is important to note that this Eurocentric system has itself changed significantly over time; indeed, it is no longer Eurocentric, but is instead U.S.-centric. The shift from a Euro- to U.S.-centric world economy, it is important to emphasize, brought with it more than just a change in leadership. In the U.S.-dominated system, we saw the ascendance of a “hyper-liberal” form of capitalism (Cox 1995, p. 37) that had an almost messianic quality. Indeed, as Gill (1995) describes it, the “relentless thrust of capital on a global scale … has been accompanied by a neoliberal, laissez-faire discourse which accords the pursuit of profit something akin to the status of the quest for the holy grail” (p. 66). And, like a religious quest, any deviation from the orthodoxy “is viewed as a sign of either madness or heresy, a view which acts to disarm criticism and to subvert the development of alternatives” (ibid.). This is the key point: ideas (values, culture, beliefs, etc.) become seen as natural or essential to our very way of life. The notion that ideas (including ideologies, values, culture, and intersubjective meanings) reflect/support the interests of the dominant class in global or national society is, of course, not unique to constructivist theories. Unlike most traditional views, however, constructivist theory sees the relationship among ideas and material forces, in a given historical structure, as invariably open-ended, no matter how entrenched they may appear to be.

 

Chapter 3: Conclusion

I hope this brief survey of contemporary theories in IPE/GPE has inspired you to think differently about the world and about the factors and processes that influence and shape the world political economy. And even if you remain unconvinced about the validity of any of the perspectives we have discussed, I trust that you will not dismiss any of them out of hand. For, just as is the case with the three “grand narratives” (mercantilism, Marxism, and liberalism), I believe the contemporary narratives discussed in this chapter all have something of value to say. After all, each of them reflects the work of many very smart people who offer the benefit of their specialization and willingness to “trade” or share their knowledge with you. Still, we are left with the questions: Who is right? Which theory provides us the clearest route to the truth? As should be clear, I do not think that proponents of any one of the perspectives can make the claim that they speak the truth. Certainly, you might find one perspective more compelling than another. But be aware that your choice is not necessarily governed by your superior grasp of reality. Rather, your choice is governed by a host of personal, institutional, historical, and, yes, even structural biases. I am not saying this to confuse you, but to urge you to keep an open but critical mind. For if you really want to learn more about the world, you simply cannot afford to ignore the myriad strands of knowledge, thinking, and understanding available to anyone willing to listen, see, and absorb.


Chapter 4

 

 

 

 

Politics, Economics, and Cross-Border Trade

 

 

 

 

 

 

The Long History of Cross-Border Trade

            Most people today, it is probably fair to say, take international trade for granted. That is, most of us consider international or cross-border trade to be a natural or an inevitable part of the world in which we live (for the purposes of this chapter, the terms international trade and cross-border trade will be used interchangeably). History seems to bear this out. We can, for example, find ample evidence of significant trade between and among many ancient and medieval powers. As Winham (2014) put it, “Trade lay at the centre of state revenue and state power in ancient Athens, Ptolemaic Egypt, the Italian city states of Venice, Florence, and Genoa, and the German Hanseatic League” (p. 110). In Asia, a vast and complex trading route called the Silk Road, which can be traced back to 2000 BCE, connected much of the old world through economic (and cultural) exchange. (The Silk Road was actually a network of routes, not a single road as the term implies.) From China, the Silk Road extended a total of 4,000 miles by land and sea through much of the rest of Asia (including India), to the Middle East, to North Africa, and to the Mediterranean and European regions.

Despite the long history of trade, it is important to recognize that the scope and scale of cross-border trade is, today, immensely greater than at any other time in human history. Consider, on this point, some basic statistics: before the global recession beginning in 2008, cross-border merchandise trade grew (in real terms) at an average annual rate of 6.2 percent a year between 1950 and 2007, compared to a much lower 3.8 percent a year between 1850 and 1913 (WTO 2008b). The growth in trade was an even more impressive 8.2 percent between 1950 and 1973. In dollar terms, this meant an increase in the value of trade from a relatively paltry $84 billion in 1953 to $13.6 trillion in 2007 (WTO 2008a). Equally significant, the growth of trade over the same period far outpaced the world GDP growth rate, which coincidentally also averaged 3.8 percent between 1950 and 2007. As a consequence of this disparity, the share of world GDP accounted for by merchandise trade (imports and exports combined) grew from only 18 percent in 1960 to more than 50 percent by 2008. This was a remarkable development, and one that has continued to the present time (2014), albeit with a significant, but short-lived decline in the first couple years of the global recession.

            At the same time, for as long as there has been cross-border trade, there have been disputes and frequently serious tensions over trade. Especially over the past century or so, these disputes and tensions have ebbed and flowed, but never disappeared. Why this should be, at least on the surface, is perplexing. After all, the growth of trade has, in an important respect, brought unparalleled prosperity to the world. Of course, this is no surprise to liberal economists (and many other social scientists): they are almost universally united in their conviction that cross-border trade is beneficial, both for individual national economies and for the world as a whole. Even within the general public (especially in wealthy capitalist economies), most people acknowledge, although perhaps only tacitly, that the antithesis of trade—namely, autarky (i.e., a policy premised on complete economic independence or self-sufficiency)—is essentially impossible and self-defeating in the industrial and postindustrial eras. In this chapter, then, one of the main goals will be to try to make sense of the continuing debate over cross-border trade. As we will see, the debate revolves around both practical political issues—e.g., who benefits and who is harmed by trade—and also around deeper theoretical disagreements.

The debate over the desirability and utility of international trade, however, will be neither the only nor necessarily the main focus of this chapter. Instead, this chapter will also examine the intersection and inextricable linkage between politics and economics in the construction (and de-construction, as the case may be) of the political and institutional framework necessary for cross-border trade to thrive. There will be a strong emphasis on the period from the post–World War II era until the present, but a discussion of the early part of the 20th century is also important. The reason for emphasizing the last 70 years or so is simple: as already noted, the post–World War II era not only marks a period of unprecedented growth and expansion of cross-border trade, but is also the first era in history in which liberal trade rules became firmly and widely embedded in the international system. Understanding how, why, and when this happened, therefore, will go a long way toward developing our understanding of the study of international or global political economy. Before moving on to a substantive discussion of cross-border trade and international political economy (including the debate over trade), however, it will be useful to begin with an overview of basic—but essential—concepts.

 

Basic Concepts and Data on Cross-Border Trade

            Most readers already have a basic and basically sound understanding of trade. If I give you my laptop computer in exchange for your brand new iPad, we have engaged in trade. The exchange of a good or service for another illustrates a particular type of trade, referred to as barter trade. In the contemporary period, of course, the great preponderance of trade involves the exchange of money for goods and services. This type of trade can take place entirely within a domestic economy, and is, in principle, no different than cross-border or international trade. But there are a number of critical distinctions between domestic and cross-border trade. In particular, in cross-border trade the exchange of goods and services is, in the most minimal terms, mediated by at least two different national governments, each of which has it own set of interests and concerns, and each of which exercises (sovereign) authority and control over its national borders. In practice, this means that even “free” trade is never entirely free. Here, for example, is a standard definition of free trade: “The unrestricted purchase and sale of goods and services between countries without the imposition of constraints such as tariffs, duties and quotas” (from investopedia.com). Yet, even today—in an era of unparalleled neoliberal globalization—all but a handful of countries or territories (i.e., Macao, Hong Kong, Singapore, and Switzerland[10]) continue to apply tariffs to a range of manufactured goods. In 2010, the (weighted) mean tariff rate for all countries and all manufactured products was 2.7 percent (see figure 4.2). While very low by historical standards, any tariff is still a government-imposed restriction on trade. Moreover, while tariffs on manufactured goods have declined dramatically over the last few decades, tariffs on agricultural products continue to be high: according to a report by the U.S. Department of Agriculture, at the turn of the 21st century, the global average tariff on agricultural products was 62 percent (Gibson et al., 2001, p. 34). Since then, the rate has come down, but it remains significant.

It would be useful to consider a concrete example: the North American Free Trade Agreement, or NAFTA, a trade agreement among the U.S., Canada, and Mexico (NAFTA is also a type of regional trade agreement, or RTA, which is discussed later in this chapter). Under the original terms of NAFTA, which came into force in 1994, Mexico was obligated to eliminate an import licensing system (a type of nontariff barrier, discussed below) for its agricultural sector; at the same time, NAFTA allowed Mexico to replace that system with tariff-rate quotas and ordinary tariffs. In other words, “free trade” under NAFTA initially meant a lesser degree of governmental constraints in cross-border trade, but not an elimination of government action. The tariffs were eliminated by mutual agreement in 2008; at the same time, both Mexico and the U.S. also agreed that “import-sensitive sectors”

 

 

 

 

 

 

 

Figure 4.2. Tariff Rates (Weighted Mean), All Products

 

Description: Macintosh HD:private:var:folders:3r:pz8t0txj4lvcmd_nlwkbdfzc0000gq:T:J1rBLr:Google Chrome.png

 

Source: World Bank (graph created by author using tools provided by the World Bank online database, available at http://data.worldbank.org/)

 


could be protected with emergency safeguard measures in the event that “imports cause, or threatened to cause, serious injury to domestic producers” (USDA 2008). This same exception is embedded in trade regimes more generally, including in Article XIX of the General Agreement on Tariffs and Trade. In addition, under NAFTA, each country explicitly retains the right to determine, for itself, the standards and protections deemed necessary to protect local consumers from unsafe products, or to protect domestic crops and livestock from the introduction of dangerous pests and diseases. While many people would see this as a reasonable qualification, it is, again, a restriction on trade. The key point is simple: free trade remains as much an abstraction as an actual practice. This would be the case even if tariffs were completely eliminated tomorrow, since there would still be significant constraints on the flow of goods and services across borders.

It is important to add that the terms of “free trade” have always been, and, for the foreseeable future, will continue to be, set by the states that manage the extent to which markets are open (this point will be taken up again later in the chapter). From a broader historical perspective, in fact, the default position between sovereign states has been a mercantilist or protectionist position, whereby different kinds of barriers to trade have been intentionally erected to either minimize or control imports and sometimes even exports. Under mercantilism, maintaining a positive trade balance (i.e., a situation in which the value of exports exceeds the value of imports) has long been a primary goal of states. To achieve this, national governments have typically engaged in some form of protectionism.

Text Box: Figure 4.3. The VCR (Videocassette Recorder)

 

The VCR was a ubiquitous, high-tech product from the 1980s to the 1990s. Ampex, an American company, originally developed the technology, but Japanese and European companies were the first to successfully commercialize the product for consumer markets. Competition was intense and began in earnest in the mid-1970s. By the early 1980s, however, Japanese companies—Sony, Panasonic, Toshiba, and JVC—were the industry leaders. 

Image: A Toshiba VCR. GNU Free Documentation License, Version 1.2
As an economic policy, protectionism “protects” domestic producers from foreign goods and producers, typically by placing a tax on specific imported goods (tariff), prohibiting their importation (import ban), or imposing a quantitative restriction (import quota). The latter two policies are examples of nontariff barriers, or NTBs. Other types of NTBs include domestic health, safety, and environmental regulations; technical standards (i.e., a set of specifications for the production or operation of a good); inspection requirements; and the like. NTBs, it is important to recognize, have, since the 1980s, become a more important source of domestic market protection than tariffs. One reason for this, which is discussed in more depth later in the chapter, is clear: by the 1980s, multilateral negotiation (through GATT) had significantly reduced tariffs, but many countries were still intent on protecting their markets. To do this, they turned to NTBs, which are less obvious and more subject to interpretation (that is, it is not immediately clear that an NTB is, in fact, meant to be a barrier to trade). On this point, it is useful to keep in mind that domestic regulations, practices, and standards are not always directly concerned with cross-border trade; other times, however, they are only thinly disguised efforts to inhibit foreign imports. One of the most famous examples of the latter situation is the Poitiers case, which involved France and Japan. In the early 1980s, French officials wanted to protect their market from imports of Japanese VCRs (see figure 4.3). The French government came up with an interesting strategy: it required that all customs inspections of Japanese VCRs be routed through a facility in a small town called Poitiers, in the center of France, that was staffed by only eight inspectors. The importation of Japanese VCRs into the French market slowed to a crawl. This policy, however, attracted such strong and immediate negative reaction that it was abandoned in less than a month (Jovanovic 2002, p. 248).

There is no need to feel sorry for the Japanese in this case, for Japan is well known for its extensive use of NTBs. In the automobile market, for example, Japan has not only required foreign automakers to meet very high safety standards, but has also required that the cars be submitted for a lengthy governmental inspection. According to Don Whitehouse, a retired Ford executive with long experience in Japan, these inspections were “brutal.” The inspectors, Whitehouse noted, “would check off every defect, even if it were well within generally accepted tolerance. They gun-sighted everything with magnifying glasses and flashlights to see if it had to be repaired” (cited in Hoffman 2009, n.p.). The result has been very low foreign car sales in Japan: in 2010, there were 4.2 million new vehicle registrations in Japan (for passenger cars), but of this total, imports by non-Japanese manufacturers were only 180,255—a scant 4.29 percent of total sales (figures Japan Automobile Manufacturers Association 2012). Inspections on cars, however, seem reasonable and can be relatively easily justified. Other nontariff barriers, by contrast, bordered on absurdity. In one case, Japanese official barred the import of foreign skis, claiming that Japanese snow was different, and in another, restrictions on foreign beef were based on the contention that the intestines of Japanese people were not the same as intestines of Westerners (Reed 1993, p. 37).

Protectionism sometimes stands alone, but it is also part of a broader policy toward international trade (and domestic economic development) that involves the use of subsidies, dumping, and industrial policy. Subsidies are designed to give domestic exporters an edge in international market competition by, most commonly, lowering the effective price of domestically produced goods. Specific types of subsidies include an assortment of practices from tax credits (and tax holidays), to access to below-market rate loans, to in-kind subsidies (for example, government funded road, sewage, and electrical service for a single factory), to the purposeful devaluation of local currency (a devalued currency makes exports cheaper and imports more expensive). Dumping is the practice of selling an exported good at a price that is lower in the foreign market than the price charged in the domestic market; frequently, dumping involves selling a product in foreign markets for less than its cost of production. A primary motivation for dumping is to capture market share—this could lead to foreign competitors being driven out of business in their own markets. For this reason, dumping is also referred to as predatory pricing. Interestingly, dumping is legal under WTO rules, unless and until it causes or threatens to cause “material injury” to a domestic industry in the importing country. Dumping can be part of a larger industrial policy, the latter of which can be loosely defined as a coherent set of polices designed to create comparative advantage in trade, to increase the competitiveness of domestic industries (vis-à-vis foreign competitors), or to nurture and develop strategic industries (i.e., industries considered vital to future economic growth and development). Neo-mercantilists consider industrial policy key to the postwar economic success of Japan, South Korea, Taiwan, and China.

There are many other basic terms related to cross-border trade, but they will be introduced in later sections so that they can be discussed in context. With this in mind, let’s turn to the first substantive section of this chapter, the debate over international trade.

 

Cross-Border Trade: A Still Contentious Debate

            From chapter 2, you are already generally familiar with the theoretical debate over cross-border trade. You know, in particular, that the liberal argument centers on the principle of comparative advantage, while mercantilists and Marxists expound upon power differentials between national economies, or on class inequality and exploitation. In this section, elements of that debate will be extended and deepened, but with a focus on “making sense” of the debate. In other words, rather than rehashing the different sides of the debate, the focus will be on explaining why disagreements over trade have not gone away. As the following section will show, there are some obvious and not-so-obvious reasons for continuing concerns about trade, despite a basic consensus among scholars, policymakers, business people, and consumers that cross-border trade is preferable to alternatives, including autarky and a purely mercantilist trade policy. As will be shown, too, the most important disagreements are not about the costs and benefits of cross-border trade per se, but are instead about the costs and benefits of free trade specifically. Let’s begin with a quick review of the more obvious (or, more accurately, widely understood) reasons.

Domestic Politics and the Debate Over Cross-Border Trade

            Perhaps most obvious reasons for the continued debate over cross-border trade revolve around the inevitable fallout from international economic competition. By definition, competition produces “winners” and “losers.” Losers, of course, want and demand “protection” from market forces (i.e., competition), but even some winners have a vested interest in protecting their positions of advantage. That is, winners often want to turn their (temporary) positions of market dominance into permanent positions of advantage. (This point connects to the issue of infant-industry protection, discussed in the next section.) Both losers and winners—organizing themselves through coalitions or interest groups—may turn to the political system for protection. Put in different terms, cross-border trade almost always has significant domestic consequences that impact specific groups within a society in different ways. In particular, cross-border trade, as Alt and Gilligan (2010) explain it, “affects the distribution of wealth within domestic economies, which raises the question of who gets relatively more or less, and what they can do about it politically.” Thus, while the rising level of cross-border trade, it is fair to say, increases overall wealth within an economy, not everyone benefits equally, and some may not benefit at all (especially in the short and medium run). There is nothing at all surprising about this observation, and this is the main point: because cross-border trade invariably leads to unequal domestic outcomes, there will always be a debate about, or tension over, how free or unhindered trade should be. Some economists have provided very useful models for analyzing the political-economic dynamics of increasing cross-border trade at the domestic level, the best known of which is the Stolper-Samuelson model. (You can find additional discussion of the Stolper-Samuelson model or theorem in figure 4.4.)

 

 

Figure 4.4. The Stolper-Samuelson Model of International Trade

 

Wolfgang Stolper and Paul Samuelson “solved conclusively the old riddle of gains and losses from protection (or, for that matter, from free trade)” (Rogowski 2010, p. 365). Most simply, the Stolper-Samuelson theorem tells us that protectionism in the form of tariffs invariably produces both winners and losers. Significantly, the theorem tells us that free trade—i.e., the lack of protectionism—also invariably produces winners and losers. To understand why can get a little tricky, and for our purposes the details are not necessary. Suffice it to say that free trade acts to lower the real wage of the scarce factor of production, while protection from trade raises it. Consider two countries: China and the United States. China has an abundance of labor relative to the United States. In other words, labor is, relatively speaking, a scarce factor of production in the U.S. Thus, without any protectionist policies in place, cross-border trade between the two countries will tend to reduce the wages of American workers in import-competing industries. These American workers are the “losers.” At the same time, other American workers and American consumers will benefit from trade. Of course, if the losing workers can successfully convince the U.S. government to protect their industry through the imposition of tariffs, their wages will go up. But this will have a detrimental effect on other Americans. Again, no matter how you slice it, there will be winners and losers due to any change in trade policy.

 

The Stolper-Samuelson theorem, however, is based on a number of simplifying assumptions, which have opened it up to criticism from a number of sources. Interestingly, some of the most voracious critics are other economists. Donald Davies and Prachi Mishra (2004), for example, argue that, while the Stolper-Samuelson theorem “has the hallmarks of great economic theory ... an enormous problem arises when we try to apply the Stolper-Samuelson theorem, unthinkingly, specifically to the question of the consequences of trade liberalization for the poorest or least skilled in poor countries.” The main problem is clear: the simplifying assumptions have very little relevance to real-world economies, which means that when it is actually applied to the real world it cannot provide reliable answers; but more than simply being wrong, “it is dangerous,” because those answers often end up as the basis for policies (p. 4).

 

Image: Paul Samuelson, American economist and Nobel Prize winner (1970). The image is licensed under the Creative Commons Attribution 1.0 Generic license.

 

 

            It is worth re-emphasizing that the tension between winners and losers, and the effort by domestic actors more generally to influence trade policy, underscores the importance of two-level games, or the interconnectedness of the domestic and international levels. While this point has been made several times already, a short review is in order. In his classic study, “The Second Image Reversed: The International Sources of Domestic Politics” (1978), Gourevitch argued convincingly that international processes and structures almost certainly have a profound impact on domestic-level politics. At the same time, Gourevitch made it clear that internal political processes cannot be ignored. These processes shape (policy) outcomes in contingent and sometimes extremely significant ways. As he put it, “[t]his idea is captured in the old concept of logrolling: the need to make bargains changes the outcome. The importance of organizations, political parties, elections, ideologies, vision, propaganda, coercion and the like as well as the more obvious aspects of economic interest arise from this need. What must be illuminated is how specific interests use various weapons by fighting through certain institutions in order to achieve their goals. Each step in this chain can affect the final result” (p. 905).

Consider the dynamics of the Cold War. The overarching rivalry between the United States and the Soviet Union, it is easy to argue, had a profound impact not only on international relations, but also on individual countries around the world. Within individual countries, however, there was great leeway in dealing with the pressures of this international rivalry. In Japan, for example, conservative political and economic leaders used the rivalry to tremendous advantage: they purged leftists and labor leaders from positions of influence in the national economy (Gourevitch makes the same point about domestic politics in the United States [p. 905]), and they used the rivalry as justification to resurrect (albeit in a different form) the prewar, highly nationalistic ideology of achieving rapid industrialization at almost any cost. This, in turn, gave rise to a powerful economic bureaucracy with strong ties to large, market-dominating business groups (which, in the prewar period, were known as the zaibatsu) with tremendous political influence within Japan. The result, to oversimplify a complex story, was the very rapid, and mostly unexpected, economic ascendance of Japan to the position of third largest economy in the world by the 1960s—a development that has played an important role in shaping the global political economy ever since. To understand and explain the debates over cross-border trade, then, it is essential that careful attention be paid to domestic-level political processes, and especially to the dynamics of inter- and intragroup bargaining.

Infant-Industry Protection

            Two other widely understood reasons for the continuing debate over cross-border trade focus on national-level interests. The first concerns so-called infant industries, and the second concerns national security. The infant-industry argument, which was alluded to in chapter 2, reflects core principles in the mercantilist or neo-mercantilist position. It is based on the idea that late-industrializing countries need to develop their own domestically based industrial capacity, lest they be permanently disadvantaged in the struggle for economic and political power internationally. The logic of the argument is clear-cut: in the real world, there are advanced and very powerful countries—with well-developed economies, competitive industries, and formidable military forces—and there are weak and industrially “backward” countries (as well as many countries that fall in-between these two poles). The countries with the most advanced market economies have a decided advantage over everyone else. This is because their industries are more developed and economically competitive. As a result, they can easily dominate ostensibly free or open markets and international trade. Text Box: Table 4.1. U.S. Agricultural Subsidies, 2012 (Selected Crops)
Crop	Number of Recipients	Total Subsidy
Corn	9	$2,702,462,268
Soybeans	7	$1,469,484,005
Wheat	12	$1,109,821,903
Cotton	53	$560,924,418
Dairy	42,229	$447,081,952
Tobacco	58,506	$188,776,927
Barley	4	$51,815,935
Sorghum	N/A	$141,933,892
Peanut	N/A	$51,011,029

Source: EWG, “The United States Summary Information” for 2012. Reproduced from a table available online at http://farm.ewg.org/region.php?fips=00000&progcode=total&yr=2012. 


Moreover, while the most powerful countries may preach free trade, they are not shy about engaging in protectionism when it suits their purposes. The United States is a case in point. In agriculture, the United States government heavily subsidizes American farmers: between 1995 and 2012, according to the Environmental Working Group (EWG 2013), the U.S. government provided $292.5 billion is subsidies. Subsidies to some individual growers are immense. For example, nine corn farms received a total of $2.7 billion in subsidies in 2012, an average of $300.2 million per recipient. Ironically, the U.S. also subsidizes Brazilian agribusiness with a $147.3 million yearly handout (Grunwald 2010). Why? Because this was the price the United States agreed to pay so that it could continue to subsidize U.S. cotton growers. (The backstory: Brazil brought a case against the United States in the WTO, which Brazil won; the U.S. had a choice to either end cotton subsidies altogether, allow Brazil to impose countervailing duties per the WTO decision, or make a deal with Brazil. The U.S. made a deal.[11])

            Most countries, of course, do not have the economic power of the United States. This was especially true in the early postwar period, when all of Western Europe and Japan were recovering from the devastation of the war. Their industries, in general, had a lot of catching up to do. More specifically, for those countries that hoped to break into key manufacturing sectors—e.g., steel, shipbuilding, aviation, and automobiles—their disadvantages were immense and seemingly insurmountable. Consider the case of automobiles, one of the most important early postwar industries. In the first few decades after the war, the major U.S. auto companies—Ford, General Motors, and Chrysler—held dominant and nearly unchallengeable positions, not just in the U.S. but also throughout the world. Japanese producers, in particular, had essentially no chance of competing head-to-head against U.S. car companies in the 1950s and 1960s: technologically, and in practically every measure of productive capacity, the Japanese were decades behind the Americans. Thus, if the Japanese had opened their domestic market fully to American automobiles, it is all but certain that the U.S. companies would have destroyed the still nascent domestic industry, which was just beginning to emerge behind the efforts of Toyota, Nissan, Isuzu, and Mitsubishi. Significantly, this is largely what happened in the 1920s, when Ford and GM first established operations in Japan (as will be discussed in more detail in chapter 6, U.S. car companies began engaging in overseas production—in both Europe and Japan—in the early part of the 20th century). It was crystal clear to everyone in Japan that the only way a domestic auto industry could develop and thrive was through infant-industry protection. And this is precisely what happened. In the early 1950s, the Japanese government essentially closed the domestic market to foreign producers through high tariffs and a strict quota. For a time, in fact, imports were limited to no more than 1 percent of the Japanese market (Cusamano 1988), although this ceiling was not strictly enforced. At the same time, Japanese companies—both final assemblers and auto parts producers—were provided numerous government-based incentives, including subsidized loans, which encouraged expansion and strong domestic competition.

            Not surprisingly, the growth and development of Japan’s “infant” car industry took a fairly long time to mature. In the first few years following the end of the war, Japan produced no passenger cars at all: between 1945 and 1947, production was exactly zero. By 1950, the Japanese auto industry had taken a few baby steps, producing about 1,600

 

 

Table 4.2. Japan’s Passenger Car Statistics (Total Production, Exports, Imports), Selected Years

Year

Total Production
(Passenger Cars)

Exports
(Passenger Cars)

Exports/Total
Production**

Imports
(Passenger Cars)

Imports/
Domestic Sales†

1950

1,594

7

0.44%

0

0.00%

1955

20,268

2

0.01%

0

0.00%

1960

165,094

7,013

4.25%

3,540

2.24%

1965

696,176

100,716

14.47%

12,881

2.16%

1970

3,178,708

725,586

22.83%

19,080

0.78%

1975

4,567,854

1,827,286

40.00%

45,480

1.66%

1980

7,038,108

3,947,160

56.08%

46,285

1.50%

1985

7,646,816

4,426,762

57.89%

52,225

1.62%

1990

9,947,972

4,482,130

45.06%

251,169

4.60%

1995

7,610,533

2,896,217

38.06%

401,836

8.52%

2000

8,363,485

3,795,854

45.39%

283,582

6.21%

2004

8,720,385

4,214,027

48.32%

281,526*

6.25%


* For 2003  ** Author’s estimates  † Author’s rough estimate (domestic sales determined by subtracting exports from total production)

Source: Japan Automobile Manufacturers Association (JAMA), http://njkk.com/statistics/index.htm

 


cars. By 1960, however, the industry was clearly growing up: production stood at just over 165,000 that year. Still fully protected from foreign competition, the “teenage” years saw even more dramatic growth. In 1965, Japan was producing almost 700,000 cars, and by 1970, well over 3 million cars. In addition, moving into the 1970s, Japanese auto producers had achieved an important degree of international competitiveness, at least in the compact car segment—in part because U.S. producers were not particularly interested in manufacturing small, relatively low-profit automobiles. This allowed Japan to become a successful exporter; in 1970, almost 23 percent of total production was being sold outside of Japan (all figures cited in Japan Automobile Manufacturers Association 2012). Japan, of course, is not the only country to follow this strategy: South Korea, for example, implemented Japan’s infant-industry model almost exactly—and with very similar results (although it took South Korean automobile producers, the best-known of which is Hyundai, longer to break into major Western markets). China is also using classic infant-industry protection policies to nurture its domestic auto industry (and is doing the same in a number of other industries today, including commercial aviation and steel).

The neo-mercantilist argument is crystal clear: Japan and similarly situated countries could not have succeeded without infant-industry protection. For poorer, late-industrializing economies today, this means that state support is not an option, but an absolute requirement (for the most part, this has always been true). To repeat, neo-mercantilists argue that active, interventionist states are necessary to provide space for fledgling domestic firms to emerge, survive, and develop. These states can provide crucial assistance in a number of ways. One prominent advocate of the neo-mercantilist approach, Alice Amsden (1989), explains it this way:

Countries with low productivity require low interest rates to stimulate investment, and high interest rates to induce people to save. They need undervalued exchange rates to boost exports, and overvalued exchange rates to minimize the cost of foreign debt repayment and imports…. They must protect their new industries from foreign competition, but they require free trade to meet their import needs. They crave stability to grow, to keep their capital at home, and to direct their investment toward long-term ventures (p. 13).

 

Text Box:  
Figure 4.5. Paul Krugman, American economist and Nobel Prize winner (2008)

Image source: Prolinserver. GNU Free Documentation License, Version 1.2.
Surprisingly, perhaps, not all mainstream economists wholly dismiss the neo-mercantilist argument. Thus, while liberal economic theory generally rejects any argument that subordinates the invisible hand of the market to the visible hand of the state, the verdict on the infant-industry argument is more open-ended. Indeed, as Baldwin (2001) notes, the “classical infant-industry argument for protection has long been regarded by economists as the major ‘theoretically valid’ exception to the case for worldwide free trade” (p. 295)—although we should note that Baldwin himself does not agree with many of his colleagues. Today, many liberal arguments about infant-industry protection draw from what is known as new trade theory (NTT), which is most closely associated with the work of Paul Krugman, a Nobel Prize–winning economist turned pundit. In a 1979 article, Krugman argued that the principle of comparative advantage could not adequately explain the level of trade that takes place between similarly situated economies. More specifically, the logic of comparative advantage suggests that the bulk of cross-border trade should occur between, say, a country with high agricultural productivity and a country with high industrial productivity (e.g., trading rice for cars instead of trading cars for computers). Yet, this was (and is) not the case, and Krugman showed why. To put it in the simplest terms, he told us that consumers have a preference for variety or diversity of goods, while production favors economies of scale. These dueling preferences create situations in which countries specialize not only in the production of certain types of goods (as comparative advantage predicts), but in certain brands or styles (e.g., Mercedes, BMW, Lexus, and Volvo automobiles).

Text Box: Figure 4.6. The Heckscher-Ohlin Model
 
The Heckscher-Ohlin model, or theory, explains why countries trade goods and services with one another. The model is based on the assumption that countries differ with respect to the availability of the factors of production (i.e., capital, labor, land). If one country, for instance, has abundant capital (in the form of machines), while another country has abundant labor (workers), both countries will benefit from trading if they specialize production based on the factor that they have in abundance. Of course, the theory is far more sophisticated than this, but sophisticated economic theories are often built from basic insights. 

Image. Eli Heckscher (left) and Bertil Ohlin (right). Both images are in the public domain.

This insight did not, at first, connect directly to a rationale for infant-industry protection, except in a very general manner. To wit, in demonstrating that there was a clear theoretical counterpoint to comparative advantage (and more specifically to the mainstream Heckscher-Ohlin trade theory [see figure 4.6 for additional explanation]), Krugman and others opened the door to the idea that infant-industry protection could be justified from a liberal economic standpoint. Over time, this is exactly what has happened. “In fact,” as Maneschi (2000) puts it, “much of the new trade theory can be regarded as providing sophisticated arguments for some forms of protection to provide favorable initial conditions, such as subsidies for research and development activities” (p. 10). A main thrust of these arguments is that the initial costs associated with protection can be recovered once a country successfully develops its own specialization within a particular industry. The example of the automobile industry mentioned above is a good example: in Japan and South Korea, tariffs, quotas, and a host of NTBs prevented foreign competition for many years, which meant consumers in those countries paid a short- to medium-term premium for inferior products from domestic manufacturers. Over the long run, however, those initial costs became insignificant as the Japanese and South Korean auto industries developed into world-class competitors.

            The upshot is simple: despite a good degree of continuing disagreement, the infant-industry argument does provide a fairly strong justification for protectionism, at least on a temporary basis. The same might be said for a third area of debate: national security.

The National-Security Argument

            The national-security argument, for many observers, makes the most intuitive sense. The logic is straightforward: war and conflict are facts of (international) life; thus, countries have to ensure that they do not unduly aid potential (still less actual) enemies by selling or transferring to them goods or technology that could be used in, or as, weapons. Achieving this objective clearly requires some restrictions on trade. In practice, however, it is often difficult to determine what, if any, restrictions are effective or reasonable. Thus, while most people would agree that a prohibition on selling materials to produce gas centrifuges that could be used to enrich uranium, which could then be used in nuclear weapons, to an actual or potential military-strategic rival is a threat to national security, other cases are less clear-cut. One particularly interesting case in this regard involved the French food giant (and yogurt maker), Danone. In 2005, it was rumored that PepsiCo of the United States was preparing a bid to take over Danone. The rumors were enough to cause an immediate uproar throughout France, and even prompted the French government to list Danone as a “strategic industry” to preempt any takeover bid. The case of Danone, it is important to note, was likely much more about French national pride than national security, [12] but for some countries food production is considered a genuine national-security issue. This is especially true for Japan, which argues that its restrictions on rice imports are part of an overall effort to achieve and maintain food, and national, security (Williams, Grant, and Fisher 1990). 

Japan’s argument about food security is straightforward—namely, in case of an international crisis, having a secure source of domestic food production means that the country cannot be threatened with an embargo that could literally starve its population. This same argument is used for a variety of other industries or economic sectors—e.g., aerospace, petroleum, heavy equipment, steel, and armaments, among others. In addition, the shift from weapon-centric warfare to what has been labeled network-centric warfare has put a premium on countries having their own domestically based expertise in information (including cyberspace) and computer technology. More generally, many countries see the high-tech sector as an essential area with military-strategic, economic, and politico-social implications all rolled into one broad imperative. This is particularly apparent with regard to China (although China is certainly not alone), which in 2006 officially launched a set of policies dubbed “indigenous innovation.” At the core of China’s indigenous-innovation initiative are 16 so-called megaprojects designed to make China a world leader in a full range of high-tech sectors, including core electronic components, high-end general use chips and software, large-scale integrated circuit manufacturing, broadband wireless and mobile communications networks, large advanced nuclear reactors, large aircraft, manned space-flight, and so on. Most of these sectors are designed to create dual-use technologies—that is, technologies that have both military and civilian applications. Significantly, three of China’s 16 megaprojects are currently deemed classified (for further discussion, see McGregor, 2010), suggesting that they are considered national-security priorities. Tai Ming Cheung, a leading scholar on China’s defense industries, believes that one of these classified projects is the Beidou Satellite Navigation System, a second-generation project that will provide global coverage in two modes: free or open services available to commercial customers with 10-meter location-tracking accuracy, and restricted or authorized services providing positioning, velocity, and timing communications estimated at 10-centimeter accuracy. Use of this technology would be limited to the Chinese government and military. The Beidou 2 satellites are also designed to withstand electromagnetic interference and attack. Presently, China still relies on the United States’ GPS and Russia’s GLONASS satellite navigation systems, which are subject to deactivation in times of conflict (cited in Raska 2013). Thus, from a national-security perspective, the development of Beidou 2 is crucial, as it will free the Chinese military from reliance on two potential adversaries, while providing essential space-age technological capability.

            Another integrally related element of the national-security argument—but one that is worth discussing separately—revolves around the issue of defense procurement. Generally speaking, defense procurement refers to the process by which military-related equipment is purchased. Liberals would argue that, from a strictly economic (i.e., cost-effective) perspective, it would make sense for countries to buy their military equipment from companies that provide the best quality product at the lowest price, regardless of where those companies are located. In other words, buying from the open (global) market would allow countries to get the most bang for their buck (excuse the pun). For some defense-related procurements, this is the policy of the U.S. Department of Defense (DoD)—which, by some measures, is the largest single purchaser of contract goods and services in the world (GovWin Network 2010). In fact, the DoD purchases all its microchips for certain military equipment from overseas vendors. Significantly, though, some congressional representatives and senators deemed this sort of reliance on foreign technology unacceptable (McLean 2005). In response, the DoD issued a report that showed that building and maintaining a domestic base for producing the needed chips would have been extraordinarily costly: a facility would have cost $2 billion to build, and a few hundred million dollars a year to maintain and upgrade. Ultimately, it was determined that the benefit of having more secure domestic production was not worth the cost of making the chips in the United States. Interestingly, McLean (a lieutenant colonel in the U.S. Air Force) asserts that the DoD generally prefers to source its military needs through the open market (p. 10), while the U.S. Congress prefers a “Buy American” policy.

The tension between Congress and DoD has resulted in a policy that is a little of both; yet, because Congress can write laws, it has been able to exercise greater influence over defense procurement practices. As far back as 1933, in fact, Congress passed the Buy American Act (BAA), which is still largely in force today. The BAA requires all government purchasers to validate that the products and services they buy are at least 50 percent American-made; it also gives a six percent advantage to domestic businesses when competing against a foreign company (McLean 2005, p. 14). Given the longevity of the BAA, it is fairly clear that most Americans (including policymakers) accept the notion that the military needs of the United States should, to a significant extent, be met by American companies and American workers producing goods on American soil. Of course, the U.S. is far from alone. Most countries, including European countries, have similar and sometimes far stricter defense procurement policies. It is important to emphasize, however, that the issue is not simply one of national security. Defense procurement policies are also extremely politicized and generally reflect a high degree of domestic-level politics that may have little to do with national security and defense.

 

 

Unequal Exchange and the Fallacy of Comparative Advantage

            So far, this chapter has dealt with the most widely understood or obvious strands of the debate over cross-border trade. As noted at the outset, however, there are also a number of less obvious arguments. Many of these come from within the Marxist tradition, and more specifically from theories concerning unequal exchange. Briefly put, unequal exchange refers to the phenomenon in which certain commodities or assets, including labor and primary goods, are systematically undervalued or overvalued. While there are several variants of unequal exchange, Arghiri Emmanuel first coined the term in 1962—although it was not popularized until 1972, when the English-language edition of his book, Unequal Exchange, came out—and developed a theory around it in which he asserted that (most) developing countries fail to benefit from cross-border trade because wages in poorer countries are kept artificially low. His argument was explicitly designed, at least in part, to challenge Ricardo’s theory of comparative advantage, one key assumption of which is that both capital and labor are relatively immobile (Custers 2007). Emmanuel argued, however, that comparative advantage would not work as predicted if one or the other factor of production was mobile while the other remained immobile. Indeed, any examination of the real world tells us immediately that, especially since the mid-20th century, capital has become far more mobile than labor. That is, while capital can and does easily cross borders, for example, to take advantage of low-cost labor and resources, workers are stuck in place (not entirely, of course, but relative to capital, labor is highly immobile). The result is that international trade is at least partly driven by capital’s unremitting efforts to find the lowest-cost wage-labor in the world. To be clear, according to the theory of unequal exchange, it is the immobility of labor that creates and sustains large wage differentials between and among countries. This process exacerbates inequality and perpetuates poverty (rather than making everyone better off), because the poorest workers in poor countries are not able to earn the real (or actual) value of their labor power.

            Labor immobility is even a problem in regions of the world where, at least on paper, citizens have the right to move across borders and take up residence with relatively few restrictions. This is the case for legal residents of the European Union. Any person who holds the nationality of an EU country is automatically also an EU citizen, which confers the right to move freely around the European Union and settle anywhere within its territory (European Commission 2010). Despite having freedom of movement within the EU, labor mobility is still highly restricted. There are a host of reasons for this, one of which is simply cultural—especially language differences. However, most of the labor immobility in the EU has nothing to do with cultural factors, but instead is a product of government policy. Specifically, migrating from one country to another often means losing welfare and unemployment rights, pension assets, tax and housing benefits, and so on (Nonneman 2007, p. 5). These policies, in turn, reflect a profound division within the eurozone between so-called core and periphery states. David Marsh (2011) describes the eurozone as a “zone of semi-permanent economic divergence, corrosive political polarization and built-in financial imbalances, beset by a ‘perpetual penumbra of hope and pain’” (p. 3). The result is the reproduction of the same pattern of unequal exchange that afflicts the world in general.

 Text Box: Table 4.3. Summary Table on Main Economic Outcomes in Brazil, China, and India

 

Notes: (a) For GDP growth, (-) indicates growth below OECD average; (+) indicates growth between 2 and 5%; (++) indicates growth between 5 and 8%; and (+++) indicates growth above 8%. (b) GDP per capita variation is measured with respect to the OECD average and the latest year value is in 2005 constant USD. (c) FDI corresponds to the inward stock: (+) indicates that FDI inward stock has increased on average above 5%, (++) above 10%, and (+++) above 20%. The latest year available value is in thousand million current USD. (d) Trade to GDP ratio measures the average trade openness during each period: (+) indicates the ratio is below 20%, (++) between 20 and 40% and (+++) above 40%. (e) 2008 data or the latest year available is given for reference. (f) Poverty incidence refers to the variation in the share of the population living on less than USD 2 per day. (g) Income inequality refers to the variation of the Gini coefficient of household income or consumption.

Table source: Copied from OECD Secretariat (n.d., p. 4)
           Emmanuel’s argument was taken up and revised by others, including Amir Samin, who argued that it is the larger structure and dynamics of the world economy (not just capital’s mobility and labor’s relative immobility) that creates and perpetuates unequal exchange. The global structure is based on a specific division of labor between countries, wherein poor (or peripheral) countries are relegated to the role of providing low-cost inputs (i.e., land, labor, and resources), while rich countries (the center or core) occupy a privileged position. The privileged position of the core is based on historical processes, especially colonialism, but also on the relatively recent emergence and development of monopoly capitalism, which is the term used by Marxian analysts to denote the (present) stage of capitalism characterized by (1) an extreme concentration of economic/productive power, and (2) the centralization of economic decision-making among a small number of major corporations (for further discussion see Sweezy 2004). It is the power of monopoly capitalism (or of monopoly capitalists) that maintains the unequal division of labor between rich and poor countries, and which ensures that cross-border trade disproportionately and systematically favors the core over the periphery. This does not necessarily imply some sort of conspiracy, but instead, simply reflects the overarching dynamics of an inherently exploitative and unequal system. To be sure, short-term gains are possible, but because poor countries tend to depend on only a few exportable products (especially agricultural products and cash crops—usually a vestige of colonial occupation), or because foreign capital controls industrial production and technology, long-term gains and dynamic economic growth are rare. (The few exceptions, we should note, are those cases in which the “rules” of free trade are purposely violated by poor countries with strong and interventionist governments. On this point, then, there is some common ground between Marxist and neo-mercantilist analysis.)        There is, it is important to emphasize, a huge amount of quantitative research that has been done to demonstrate that increased trade via liberalization either results in large net gains for all countries (which contradicts the Marxian argument on unequal exchange), or fails to help developing countries grow their economies. Other chapters in this book cover some of the relevant data. For example, statistics on the significant decline in absolute poverty (chapter 1) indicate that trade liberalization is helping a range of poor countries overcome the worst conditions of poverty. On the other hand, chapter 5 shows that liberalization may be radically increasing the debt burden of many developing countries. Certainly there are other data that can be used, but for each set of statistics one side uses, the other side can usually find contradictory evidence. One point, however, is clear: there have been some significant success stories among developing countries. The most salient is China, which, partly through cross-border (but not free) trade, has been able to increase the standard of living among its people dramatically over the past two to three decades. Other major cases include India and Brazil. In these three success stories (at least by liberal standards), a significant increase in cross-border trade during the 1990s and 2000s has, according to the OECD, led to a decline in severe poverty (defined as living on less than $2.00 a day in PPP terms): in China, over 600 million people have been lifted out of poverty since the late 1980s, while in Brazil, the number is 11 million. In India, by contrast, the decline is only in the percentage of the population living in poverty, while the actual number of poor increased by 40 million. Significantly, perhaps, while severe poverty declined in all three countries, inequality did not. Brazil did witness a small decline in inequality of 9 percent (measured in terms of the Gini coefficient), but it remains one of the most unequal countries in the world. China and India, by contrast, saw increases in inequality of 24 and 4.5 percent respectively. We are left with a somewhat mixed message (especially if we include all developing countries), which is another reason why the debate over the costs and benefits of cross-border trade remains unresolved.

 

Cross-Border Trade, Negative Externalities, and the Global Environment

            Even if the trading/economic success of China, India, and Brazil were unequivocal, the world would still be left with perhaps the most important, albeit long-term, cost of increasing global trade and the intense economic competition (as well as increase in world economic growth) it creates: environmental degradation. This, too, is a less-than-obvious concern about cross-border trade and free (or unregulated) trade in particular. The basic problem can be expressed in economic terms: pollution and other economic practices that damage the environment are a type of negative externality. A negative externality, in the most general terms, is a cost that is suffered by a third party to an economic transaction. Externalities are most common in situations in which ownership of a particular asset or resource cannot be determined, or is uncertain. The most common examples come from the natural environment: the air or atmosphere, the oceans, rivers, the ozone layer, etc. The basic problem is clear: if the natural environment is not owned by anyone or anything, then economic actors can pollute or damage the environment without fear of incurring additional costs. Indeed, pollution and environmental destruction become fully rational actions under these circumstances. After all, why would any business owner incur additional expenses—expenses that might make his or her company less competitive—to prevent or mitigate pollution and environmental destruction if engaging in such practices is “free” to that company?

At the domestic level, negative externalities can be addressed through legislation or state action: economic actors can be compelled or forced to engage in environmentally sustainable activities. Of course, this is sometimes easier said than done, but it is clearly possible, especially when dealing with environmental issues that are tied directly to a local environment, such as pollution of a river or lake. However, when the environmental issues are tied to the global environment (e.g., global warming), the ability to develop effective legislation becomes significantly more difficult, since there is no world government that has binding, legal authority over all states (although, as we will see in the following section, the world does have legislative and juridical analogues to domestic governments).

Text Box: Figure 4.7. CO2 Emissions by Region/Country, 1990–2012

 

Source: Reproduced from Olivier, Janssens-Maenhout, and Peters (2012), “Trends in Global CO2 Emissions; 2012 Report.” Available at http://www.pbl.nl/en/publications/2012/trends-in-global-co2-emissions-2012-report 
An added problem is that cross-border trade makes it easier for the industrialized economies of the core to transfer their domestic environmental concerns to peripheral countries, the latter of which typically have governments either unwilling or unable to address environmental issues. That is, by outsourcing production of many (especially consumer) goods, most of which are sold back to and consumed by individuals in wealthy economies, the core effectively exports its environmental problems to the periphery. This has especially been the case in China, the “poster boy” for the transformative powers of international trade. China has indeed been transformed—into a world leader in environmental pollution and destruction. In 2007, for example, China overtook the U.S. in aggregate CO2 emissions (although, on a per capita basis, its emissions are still relatively low); in 2009, according to the International Energy Agency, China also overtook the United States in total energy use (2.252 billion tons of oil equivalent, compared to 2.17 billion tons in the U.S.); 16 of 20 of the most polluted cities in the world are in China; and, not surprisingly, China is the world’s top emitter of sulfur dioxide (a noxious gas with a pungent, irritating smell). Even more, the average daily discharge of polluted water in China is comparable to that of the U.S., Japan, and India combined, which has resulted in the serious contamination of over 70 percent of China’s rivers and lakes. This has exacerbated a water shortage in China to a critical threshold. Desertification is also accelerating, which is largely responsible for annual dust storms that spread toxic clouds of fine soil (called “yellow sand”) throughout East Asia every spring. The toxicity is from industrial pollutants contained in the dust (all figures cited in Morton 2009, p. 3).

The problem, to repeat, is that China’s environmental problems are all symptomatic of the larger economic system in which China operates, a system of global capitalism pushed forward by an acceleration of cross-border trade. To put the issue more generally, the commoditization of nature and the rejection of environmental preservation and protection as a hindrance to profit maximization are part and parcel of the capitalist process. China is simply the latest in a long line of capitalist economies that have already followed (and are still following) this path. China will start exporting its environmental problems to other, poorer countries, too, and as problems worsen in China, and as public outcry within China grows (a phenomenon that is already happening—see figure 4.8), it is likely that this process will accelerate. Of course, simply shifting environmental problems from one location to another is not a solution. More importantly, China’s sheer size makes this an increasingly untenable solution, primarily because China’s continued economic “development” will increasingly have more and more adverse global environmental implications: CO2 emissions, to cite the most salient example, have no respect for national borders.

 

Figure 4.8. Social Networking and Environmental Problems in China

 

Despite the fact that China is still an authoritarian political system, the government has had to respond to increasingly strong public opinion on environmental problems. Much of this opinion is expressed through social media, especially Weibo, China’s micro-blogging version of Twitter. For example, public outcry on Weibo compelled the Chinese state to publicize the particulate count of air quality in major cities (e.g., http://aqicn.org/city/beijing/). In addition, China has seen a rapid increase of environmental NGOs: between 1994 and 2000, the number of registered environmental NGOs went from essentially zero to as many as 2,000, although there are widely varying estimates, due in part to the ambiguous manner in which NGOs are defined in China (Schwartz 2004). Most analysts concur that China’s environmental NGOs presently exert very limited influence, but their rapid growth indicates growing domestic concern, both on the part of citizens and the Chinese state.

 

Image description: Haze over China. The thickest of the grey-brown haze conforms to the low-lying contours of the Yellow River valley, and the western half of the North China Plain.

 

Source. NASA. The image is in the public domain because it was solely created by NASA, an agency of the U.S. government.

 

Of course, not everyone agrees with this assessment. There have been a number of studies done that show, at best, an ambiguous relationship between expanding international trade and global environmental destruction. As with almost any political-economic issue, in other words, there is intense and ongoing debate fueled by competing claims and seemingly contradictory empirical analyses. It is well beyond the scope of this chapter to evaluate the competing claims. Nordström and Vaughan (1999), however, did a very good review and evaluation (albeit somewhat dated) of the literature on the relationship between cross-border trade and environment. Readers interested in exploring this issue in more depth are encouraged to read the Nordström and Vaughan study on their own.

Text Box: Figure 4.9. What Is Fair Trade?

Here is how the World Fair Trade Organization (http://www.wfto.com/) defines fair trade: “Fair Trade is a trading partnership, based on dialogue, transparency and respect, that seeks greater equity in international trade. It contributes to sustainable development by offering better trading conditions to, and securing the rights of, marginalized producers and workers—especially in the South.” In this view, fair trade is based on “just compensation” for everyone involved in the production, distribution, or sales of a product. This means that the small farmers and the unskilled workers, those who typically receive the smallest share of income, should be paid a living wage—that is, a wage that covers the basics of food, shelter, clothing, education, and medical care. 
 
 
Image is in the public domain.
The Social Construction of Free Trade

            In the foregoing scenario, it is important to recognize that the freer the trade (that is, the less governmental regulation and management), the worse the situation is likely to be. Free trade, by definition, necessitates a lack of government (or nonmarket) oversight over economic processes and transactions. Genuinely free trade means, therefore, that some negative externalities might never by solved or addressed. The solution, some would argue, is not the elimination of trade, but instead, a continued regulation and perhaps re-regulation of cross-border trade. This might mean fair trade (see figure 4.9) as opposed to free trade, or it could mean something else. This suggests, in turn, that it would be necessary to begin constructing a very different type of trading and economic system, one based on principles and ideals that diverge, perhaps significantly, from what liberal economic theory or neoliberalism suggest. 

The possibility of constructing a meaningfully different type of trading and economic system reflects the basic principles behind the constructivist perspective, which was discussed in the previous chapter. Constructivist approaches, to repeat (once again), tell us that there are always different possibilities, but that these possibilities must be constructed through an interactive process involving a complex mix of material and nonmaterial factors and forces. Different possibilities, however, are not always easy to see or even imagine, especially when there are powerful structures already in place—e.g., the current neoliberal economic framework governing cross-border trade and international economic relations. Indeed, “free” trade as both an idea and an ideal has become deeply ingrained in the global system, even if free trade as an actual practice has never fully existed. This paradoxical situation is not difficult to understand: all social structures require an ideological (or ideational) basis that shapes not only what people think (and how they define themselves and their interests), but also what they do. Ideas and actions do not always match up. The key point, though, is this: the creation of a free-trade system, to the extent that it exists today, is a product of concerted and sustained collective action. The free-trade system is, therefore, a social construction. It is a real structure that exercises significant power in the world today, but it is also a normative framework for understanding and defining how cross-border trade should be conducted.

            Asserting that the current system of cross-border trade is socially constructed may strike some readers as wrongheaded. Yet, if we examine the history of the international or global trading system we will see that there is nothing particularly controversial in that assertion. This is particularly evident in looking at the historical dynamics of cross-border trade throughout the 20th century, and especially since the end of World War II. For it was during the postwar period that a new international system of trade was quite consciously constructed through a variety of means. Hitherto nonexistent international regimes, rules, and institutions were created largely from scratch. Liberal principles were given renewed and even greater prominence; in those places resistant to liberalization (of which there were many), both sticks and carrots were used to ensure compliance. Very little of this occurred spontaneously, and very little of it could have occurred spontaneously. Indeed, underlying all these changes was a great deal of structural and material power, much of which was exercised specifically by the hegemonic power of the time, the United States. The construction of the postwar trading system, therefore, was also very much a profoundly political process. It is to this issue that we will turn next.

 

The Rise of “Free” Trade in the 20th Century, Part I

            Neither the idea nor the practice of free trade, as we have just seen, has always been readily or even mostly accepted, including in the major Western economies. In the United States, in particular, trade policy was generally protectionist until the mid-1930s—although, as we will see shortly, the U.S. began to pursue, in fits and starts, a policy of more open trade beginning in the last part of the 1890s. Still, the overall tone of America’s position on cross-border trade was solidly protectionist; this was clearly revealed in the general discourse of the time. Consider just one example: in 1895, shortly before he became president, Theodore Roosevelt wrote, “Thank God I am not a free-trader. In this country pernicious indulgence in the doctrine of free trade seems inevitably to produce fatty degeneration of the moral fiber” (cited in Irwin 2001, p. 61). To be sure, during that period, free trade was the subject of intense partisan bickering and posturing (at the time, Republicans championed protectionism, while Democrats were supporters of free trade), but it is nonetheless true that protectionists often held the upper hand. This was well reflected in the passage of the Smoot-Hawley Tariff Act (formally, the United States Tariff Act of 1930), which increased import duties on top of already high tariff rates in the United States. Indeed, the Smoot-Hawley Act—along with other important trade legislation of that era, including the Fordney-McCumber Tariff Act—are generally held up as exemplars of unabashed American protectionism in the early part of the 20th century. Still, support for a more liberal trading policy, as already noted, was not absent. Indeed, by the mid-1930s, a new, more open trade policy began to emerge with the implementation of reciprocal trade agreements (discussed in more detail below). This tells us that we need to view U.S. protectionist policies in the first part of the 20th century in a wider perspective.

In this wider perspective, the United States could be understood (as many scholars have argued) as a rising hegemon: economically, politically, and militarily, the country was moving from a position of relative equality with the European powers, to one of preeminence. At the same time, Britain (which had occupied a hegemonic position beginning in the early to mid-19th century) was clearly on the decline. In such a period of hegemonic transition, advocates of hegemonic stability theory argue, instability and especially protectionism are very likely. David Lake (1983) explained the situation as follows:

When no hegemonic leader exists and only a single supporter is present, there are no constraints on protectionism within the supporter. Although it will continue to value export markets and may attempt to lead the international economy, a single supporter will lack the resources to stabilize the international economy successfully, or to create and maintain a liberal international economic regime. If the supporter believes that it cannot preserve its export markets, the protectionist fires at home will be fueled. The growing flames may precipitate the abdication of whatever leadership role had been held by the supporter (p. 523).

Lake argues that, in the early part of the 1900s, the U.S. largely fit the description above as a “supporter.” This can be seen, for example, in the U.S.-led effort to maintain the Open Door policy in China at the turn of the 20th century. The Open Door policy—which was based on a series of diplomatic notes written by Secretary of State John Hay—was directed at maintaining equal and nondiscriminatory privileges among the major countries trading with China at the time: the United States, Russia, Germany, France, Japan, and Britain. It was, in other words, a “free” trade policy, albeit for a geographically limited area. The principles of the Open Door, it is important to point out, long predated the U.S. initiative. In fact, they were first articulated and enforced by the former hegemon, Great Britain, under the terms of the Anglo-Chinese treaties of Nanjing (1842) and Wangxia (1844). For half a century, the British maintained the open-door principles, but this started to break down in the 1890s as the major industrial powers began a scramble for spheres of influence in various parts of coastal China: within their respective spheres, they all claimed exclusive privileges. Importantly, Britain was not an exception: not only did the country abrogate its former (hegemonic) role, but the British also took part in staking out their own sphere of influence. The U.S., therefore, was essentially on its own in trying to maintain the open-door system. The U.S. did so, in large part, because it occupied a disadvantageous position in China (relative to the European countries and Japan), and had little to gain if China became a thoroughly closed market. The U.S., however, had a very limited capacity to change the behavior of the other states; indeed, one can argue that the U.S. effort was an abject failure. Nonetheless, as Lake asserts, the Open Door policy marked a significant first step by the U.S. toward assuming the hegemonic mantle. (The pursuit of reciprocal trade agreements in the 1930s was another important, and far more successful step; again, this point will be discussed below.)

As a fledgling hegemon, the U.S. commitment to maintaining an open system was not entirely consistent. This helps explain why the U.S. continued to follow an ostensibly protectionist line in its trade policy more generally. As already noted, in 1922 and 1930, two major tariff billed were passed: the Fordney-McCumber Act, and the Smoot-Hawley Tariff Act. Interestingly, earlier in the century, two other tariff bills were passed that lowered tariffs. These were the Payne-Aldrich Act of 1909 (which was nonetheless a protectionist bill) and the Underwood Tariff Act of 1913 (a partly liberal bill, in that it significantly lowered tariffs). Except for the infamous Smoot-Hawley Tariff Act, it is important to note, the various tariff bills were at least partly used as “bargaining tariffs” designed to extend the Open Door policy abroad (Lake 1983, p. 534). More specifically, they contained a provision for “flexible” tariffs that would allow the U.S. to impose higher tariffs on countries that discriminated against American goods. For a variety of reasons—not the least of which is that using protectionism to reduce protectionism was viewed as hypocritical by America’s trading partners—the strategy did not work. But the U.S. had few other tools at its disposal. Partisan and interest-group politics within the U.S. also complicated the issue (Eichengreen 1986), and many Americans remained thoroughly unconvinced of the virtues of free trade. The situation, however, would soon change.

Table 4.4. Duty Level by Tariff Act, 1897–1930

Tariff Act (Name), Date

Level of Duty on All Imports

Level of Duty on Dutiable Imports

Percentage of All Imports on Free List

Dingley, 1897

26.2

47.6

45.1

Payne-Aldrich, 1909

20.0

41.0

51.3

Underwood, 1913

8.8

26.8

67.5

Fordney-McCumber, 1922

13.9

38.2

63.5

Smoot-Hawley, 1930

19.0

55.3

65.5

 

Source: Lake (1983), p. 534. Original data are from the Statistical Abstract of the United States, selected years.

 

 

The Great Depression, the RTAA, and the Emergence of U.S. Hegemony

            In chapter 3 there was a brief discussion of the economic and political effects of the Great Depression. It was noted, in particular, that the decision by the U.S. to erect higher protectionist barriers through the passage of the Smoot-Hawley Tariff Act helped spur a significant worldwide rise in tariff rates and other discriminatory measures; not surprisingly, U.S. goods were a key target (Irwin 1998; also see table 4.5). While it is not at all clear that the Smoot-Hawley Act caused or even significantly contributed to the severity of the Great Depression,[13] it is nonetheless clear that it led to a serious rethinking within the United States about the efficacy of the tariff as the main instrument of trade policy. In this regard, the Great Depression played an important role in pushing the domestic political balance in favor of the democrats, who were then able to use their newfound voting advantage to pursue reciprocal trade agreements (agreements that would reduce tariffs on a bilateral, as opposed to multilateral, basis). It is important to note here that the shift in thinking and action also represented a more profound change. That is, it reflected the process by which a new reality for international trade was being constructed. The extant worldview—one premised on neo-mercantilism and beggar-thy-neighbor policies—had defined the basic nature of the international trade regime for decades. The failure of those policies pushed countries around the world toward a different normative framework, one that would later lead to institutional innovations such as GATT, the WTO, and the idealization of “free” trade.

Table 4.5. Average Tariff of Major U.S. Trade Partners, Selected Years

Country

Share of U.S. Exports

1913

1928

1932

Canada

18.2

26.1

23.3

27.4

United Kingdom

16.5

4.3

9.9

23.1

Germany

9.1

6.3

7.9

23.8

France

5.6

9.3

5.5

5.4

Japan

4.7

9.2

8.7

17.5

Argentina

3.5

17.6

18.8

28.8

Italy

3.2

7.4

6.7

23.5

Netherlands

2.8

0.4

2.1

4.7

Australia

2.7

17.9

22.4

41.2

 

Note: Figures are not comparable for all countries. For Canada, the listed tariff rate was for U.S. imports only; for other countries, the average tariff rate is based on the tariff revenue divided by total imports.
 

Source: Table reproduced from Irwin (1998), p. 339.

 

 

The basis for bilateral trade agreements came relatively quickly with the passage of the Reciprocal Trade Agreements Act (RTAA) of 1934. The successful passage and implementation of the RTAA has been intensively studied and debated by both economists and political scientists,[14] and for our purposes a detailed discussion is not necessary. Suffice it to say that the RTAA set in motion a largely virtuous cycle (as liberals would emphasize) that proved to be beneficial to the industrialized world, and especially to the United States. Because of this, the RTAA helped to cement a liberal trend in U.S. trade policy that had not previously existed. The path toward a liberal international trading system was, of course, interrupted by the outbreak of World War II. Significantly, though, World War II interrupted the process, but did not stop it. Indeed, in important ways, it may have ultimately accelerated the process of trade liberalization. As Hiscox (1999) argues, World War II had the effect of radically reducing—although only temporarily—import competition for U.S. manufacturers while simultaneously fueling a tremendous expansion in export demand for U.S.-made products (p. 685). The reason is easy to see: unlike most of the industrialized world, the United States homeland was essentially isolated from the devastation of the war. While other countries had to rebuild, the U.S. was able to build up from a still-intact, and very strong industrial foundation. Support of freer trade in the early postwar period, therefore, became almost a no-brainer, even for formerly diehard protectionists in the Republican Party. This was a major reason why, in 1948, the Republican platform dropped its strong prewar opposition to the RTAA (p. 686). By the time the postwar export boom for the United States began to peter out in the early 1960s, opposition to “free” trade had become too fragmented for an easy return to protectionism.

Another strongly related effect of the war was to leave the United States with no viable rival in the capitalist world. The physical decimation of the European economies and Japan meant that U.S. economic, military, and political supremacy was well-nigh indisputable in the early postwar period. This was crystal clear to practically everyone, both inside and outside of the United States. It was, therefore, relatively easy—almost natural—for the U.S. to assume an unchallenged leadership role in world affairs, which is exactly what happened. In short, the de facto transition from British to U.S. hegemony had basically been accomplished as a result of World War II. Armed with a new liberal outlook, the United States was now in a position to pursue its open-door policy in a more vigorous and far more effective manner than before, and it wasted no time in doing so, as we will see in the following sections.

 

A Theoretical Caveat

            In the next section, we will examine important elements of the U.S.-led process of trade liberalization in the postwar period; before doing so, however, an important caveat is in order. The foregoing analysis offers an interpretation of events in the first half of the 20th century primarily from the standpoint of hegemonic stability theory (HST), although strong elements of the two-level game approach can be seen as well. And, in a general and loose way, the following discussion will continue along these lines. Yet, as should be quite clear by now, theoretical interpretations differ, sometimes in dramatic ways. That said, it is useful to employ the HST framework, not only because it is one of the more widely accepted interpretations of international trade during the 20th century, but also because it can fit with a variety of approaches. As noted in chapter 3, HST is a hybrid theory that can contain elements of mercantilism (realism), liberalism, and Marxism.

Since previous discussion of HST focused on the mercantilist view, it will be useful to say something about how HST fits with Marxist approaches. Classical Marxist analysis did not have much to say about hegemony. Contemporary versions of Marxism, however, recognize hegemony as an important element of global capitalism in general, and of free trade more specifically. Perhaps the most important of these contemporary views is world systems theory (WST), which is primarily credited to the work of Immanuel Wallerstein. In Wallerstein’s view, hegemony “refers to those situations in which one state combines economic, political, and financial superiority over other strong states, and therefore has both military and cultural leadership as well. Hegemonic powers define the rules of the game” (Wallerstein 2004, n.p.). Given the dominant status of the hegemon, the state that occupies this position will generally use its power to support and maintain the system, and to ensure that any challenges to the system are eliminated or minimized. During the Cold War, for example, the efforts by the Soviet Union to create an alternative world system—one based on withdrawal from or nonintegration into the capitalist world system—compelled the United States to use its considerable resources to prevent the Soviet Union from expanding its sphere of influence, and to prevent other countries from “going communist.” U.S. policymakers at the time implicitly understood the need to expand the boundaries of capitalism on a global basis, and therefore saw Soviet efforts as an economic—as opposed to military-strategic—threat. To achieve their goals, U.S. policymakers had to not only “contain” the Soviet Union, but also deepen and expand capitalism anywhere it was possible to do so. This explains, for example, why the United States decided to support Japan’s emergence as a center of capitalism in Asia in the late-1940s—despite having just completed a vicious and hate-filled war with that country. This meant providing Japan with one-way access to U.S. markets in the 1950s and 1960s (that is, the U.S. allowed Japan to sell as much as it could to the U.S., while not requiring that Japan open its markets to U.S. goods), but it also meant fighting a war on behalf of the Japanese, namely, the war in Vietnam (on this last point see figure 4.10, “Japan, Vietnam, and the Falling Domino Principle”).

On the surface, all of this sounds quite similar to the mercantilist view, and the similarities are admittedly strong. However, there is one key difference: in WST, hegemony reflects class dynamics and class power. In other words, it is not unitary states making decisions and acting in the national interest, but dominant class actors who are directly or indirectly calling the shots. Even more, in the world-systems view, hegemony reflects the inherently exploitative nature of the capitalist world system: the world is divided into unequal zones (i.e., the core, semiperiphery, and periphery), and the hegemon plays a key role in ensuring the integrity of this structure. In this structure, wealth is systematically extracted from the poorer and weaker zones (the periphery and semiperiphery) and brought to the core, and one of the most effective ways to do this is by imposing a “liberal” world order, one ostensibly premised on free markets and free trade. Of course, countries with weak, industrially backward economies cannot effectively compete in such a world, and in those areas where they might be able to compete, such as agriculture, the hegemon and other core economies conveniently ignore the principles of the free market and free trade. The main point is that the concept of hegemony is represented in a variety of theoretical approaches. In addition, it can be asserted that this is no accident; in other words, hegemony has been embraced by a variety of perspectives precisely because it provides a useful basis for understanding and explaining the emergence and initial dynamics of the U.S.-led, liberal international trade system in the postwar period. (Keep in mind that one of the primary criticisms of arguments focusing on hegemony is not that the concept is flawed, but that hegemony is a relatively short-term phenomenon.)

 

 

 

Figure 4.10. Japan, Vietnam, and the Falling Domino Principle

 

Asserting that the U.S. fought the war in Vietnam to benefit Japan may seem a huge stretch, but to see why it is not, consider the famous falling domino principle. The domino principle is based on the idea that, while Vietnam in its own right may have been unimportant to U.S. national security, if the U.S. had allowed Vietnam to fall to communist forces, this would have led to other countries in the regime also falling to communism. Thus, the fall of Vietnam would have meant the collapse of several more pro-American allies in Asia, including Malaysia, Indonesia, Thailand, Burma, and the Philippines. The implication was that this would have seriously compromised the U.S. military-strategic position in the region and globally. Yet few people are aware of the original logic behind the falling domino principle, which was first enunciated by Dwight D. Eisenhower. In 1954, in response to a reporter’s question about the strategic importance of Indochina (the old name for Vietnam, Laos, and Cambodia), Eisenhower said this:

Text Box:
You have, of course, both the specific and the general when you talk about such things. First of all, you have the specific value of a locality in its production of materials that the world needs. Then you have the possibility that many human beings pass under a dictatorship that is inimical to the free world. Finally, you have broader considerations that might follow what you would call the “falling domino” principle. You have a row of dominoes set up, you knock over the first one, and what will happen to the last one is the certainty that it will go over very quickly. So you could have a beginning of a disintegration that would have the most profound influences . . .

[W]hen we come to the possible sequence of events, the loss of Indochina, of Burma, of Thailand, of the Peninsula, and Indonesia following, now you begin to talk about areas that not only multiply the disadvantages that you would suffer through loss of materials, sources of materials, but now you are talking really about millions and millions and millions of people.

 

Finally, the geographical position achieved thereby does many things. It turns the so-called island defensive chain of Japan, Formosa, of the Philippines and to the southward; it moves in to threaten Australia and New Zealand. It takes away, in its economic aspects, that region that Japan must have as a trading area or Japan, in turn, will have only one place in the world to go—that is, toward the Communist areas in order to live. (2005 [1954], p. 383; emphasis added)

 

Think about the italicized part: Eisenhower justified intervening in Vietnam in order to protect Japan’s “trading area”! It’s hard to imagine why the United States would commit its own resources and people to defending another country’s trading area, except if we take into account the world-systems view on hegemony.


Image. Dwight D. Eisenhower as general of the army (circa Dec. 31, 1943). The image is a work of a U.S. army soldier or employee taken as part of that person’s official duties. As a work of the U.S. federal government, the image is in the public domain.

 

The Rise of “Free” Trade in the 20th Century, Part II

            It is fairly clear that there is almost nothing spontaneous or automatic about the emergence and development of a freer or liberal system of free trade internationally. In the prewar period (we can argue), it was a lack of a centralized authority with sufficient material, structural, and political power that prevented a stable, widespread system of free trade from developing. Without a solid political foundation and structure, in other words, international or global markets could not fulfill their “potential.” Even more, without economic stability, the tensions inherent in the interstate system became increasingly difficult to contain, which made the outbreak of World War II—or, if not that war, then another conflagration between major states—almost inevitable. It is no surprise then that one of the first priorities of the United States in the postwar period was to create the framework for a more stable international economic system. In previous chapters, a key part of this framework was mentioned: the Bretton Woods system (BWS). As you learned, the Bretton Woods system is most closely associated with the creation of the international monetary or financial system. But another key (and strongly related) aspect of the BWS was the effort to liberalize international or cross-border trade on a sustained, multilateral basis. This effort was largely successful, as over several decades a new international trade regime was constructed. This almost assuredly could not have been achieved without the exercise of tremendous political will and power, and more specifically without the coordinating and stabilizing efforts of the United States.

            Creating an international trade regime, then, was not a foregone conclusion. As noted in chapter 3, one of the very first attempts in the postwar period to do this failed. Specifically, after several years of both bilateral and multilateral negotiations, a draft agreement known as the Havana Charter (formally, the Final Act of the United Nations Conference on Trade and Employment) was announced on March 24, 1948. The charter’s main objective was the creation of the International Trade Organization (ITO). Although signed by 53 of the 56 countries participating in the final UN conference on this issue, the ITO failed to materialize. One big reason for this was the unwillingness of the U.S. Congress to ratify the charter, and without U.S. commitment, it was no surprise that other countries likewise refused to move forward. In retrospect, it is not difficult to see why the U.S. Congress refused to approve the charter. Specifically, the ITO was, at the time, an extraordinarily ambitious idea. As Narlikar (2005) explains it, “The ITO envisaged by the Havana Charter had a far-reaching mandate, and an elaborate organization to implement it.... [B]esides covering the obvious areas of commercial policy, the 106 articles of the ITO extended to areas of employment, economic development, restrictive business practices, and commodity agreements. It gave recognition to the importance of ensuring fair labour standards, and also incorporated provisions that allowed governments to address their development and humanitarian concerns” (p. 12). More simply, the ideas behind the ITO were too “radical” and too much to handle for many U.S. congressional free-trade-wary representatives. This points to a larger lesson as well: the failure of the ITO could very well have spelled the failure of the entire push toward trade liberalization in the early postwar period.

            Fortunately (but not necessarily fortuitously), the ITO was only one prong of a multi-pronged strategy. Another prong was the General Agreement on Tariffs and Trade (GATT). Originally, the GATT was to be an interim arrangement until the ITO came into force, but it was also pursued by the Truman administration precisely because of fears that the U.S. Congress would oppose the more ambitious provisions of the ITO (Kaplan 1996, p. 53). Unlike the ITO, moreover, the GATT did not require ratification by the Congress (ironically, this was a result of the congressionally approved RTAA). As an interim agreement, the GATT was much less ambitious than the ITO; it focused on trade (or commercial) relations only and revolved around three basic principles: (1) nondiscrimination—i.e., the concept of most-favored-nation (MFN) status—in trade relations among participating countries; (2) a commitment by all participants to observe the negotiated tariff concessions; and (3) a prohibition of quantitative restrictions (quotas) on exports and imports. Not only was the substantive coverage much narrower, but so too was its legal and institutional basis. The GATT was, in essence, little more than a negotiating forum; it was not an international organization, nor did it even have a membership per se—instead it had “contracting parties” (Narlikar 2005, p. 16). Despite these shortcomings, the GATT not only survived (for 47 years, after which it was formally replaced by the WTO), but it proved to be an effective—albeit far from perfect—means for establishing the necessary groundwork for a significant expansion of cross-border trade.

            The effectiveness of the GATT was clearly premised on the support and leadership of the U.S. government; for just as the ITO failed without U.S. support, so too, it is fair to conclude, would have the GATT. This did not mean that support within the United States was undivided and consistent—it most certainly was not. Still, because sufficient trade-policy authority had been transferred to the executive branch, the GATT did not meet the same fate as the ITO. Thus, as each round of the GATT was negotiated, the agreements reached on tariffs were, without any “drama,” enacted. Until the Kennedy Round (1964–67), however, tariff negotiations had to take place on an item-by-item basis. Despite this cumbersome condition, the very first round of negotiations (the Geneva Round, in 1947) produced 45,000 tariff concessions. Subsequent rounds were less impressive, but significant tariff reductions were achieved overall—for example, the Torquay Round (1950) led to a cutting of the 1948 tariff levels by 25 percent. In the Kennedy Round, a new across-the-board method was used (to achieve this, the U.S. Congress had to give the president the power to abolish item-by-item negotiations, which was originally codified in the Reciprocal Trade Agreement Act of 1934). This resulted in average tariff reductions of 36 to 39 percent, worth about $40 billion. Put in different terms, by the end of the Kennedy Round, the average tariff on manufactured goods was about 10 percent, compared to a 40 percent average in 1947. From the discussion at the beginning of the chapter, it is clear that these reductions in tariffs (along with nondiscrimination and the elimination of quantitative restrictions) helped lead to a significant—really, unprecedented—expansion of cross-border trade through the 1950s, 1960s, and beyond.

The details of each round, while important, are not the main concern. The main concern is how the GATT negotiations laid the groundwork for the establishment of a liberal international trading regime (or a new socially constructed order for international trade). In this regard the move from item-by-item negotiations to across-the-board negotiations—which was partly related to the growing economic power and influence of the European Economic Community (later the European Community, and finally the European Union)—was an important step. Specifically, it made it more difficult for parochial domestic interests (i.e., special interests) to influence negotiations (Kaplan 1996, p. 68). Also, the successive rounds of multilateral negotiations helped to establish a practical and normative framework for talks and disputes over trade issues in general, not just tariffs. Tariffs, as you know, are not the only barriers to trade. NTBs (nontariff barriers) are equally, and potentially more, corrosive to trade, since they are nontransparent (meaning that it is not always obvious what constitutes an NTB). As tariffs declined, many countries began turning to NTBs. But before this trend could completely undermine the progress made through tariff reductions, talks on nontariff barriers were added to the GATT agenda during the Tokyo Round (1973–79). To be sure, early negotiations could not and did not immediately resolve or effectively address the issue, but they allowed for an ongoing dialogue among states. Subsequent rounds continued these discussions, and added other vexing issues, including intellectual property, agricultural subsidies, textiles, and dispute settlement. It is important to emphasize that as negotiations moved beyond tariffs—for manufactured as opposed to agricultural goods—agreements became harder and harder to come by (this was probably also the product of an expanding membership—from 23 participating countries in 1947, to 62 during the Kennedy Round, to 123 during the Uruguay Round). Yet, because a basic and sustained framework for trade negotiations had been created and institutionalized, the movement toward a more liberal trading order, while sometimes stalled, did not reverse itself or completely fall apart, as could have easily happened.

 

The Birth, Significance, and Troubles of the WTO

Text Box: Figure 4.11. Official Logo of the WTO
 
Image: The image is in the public domain.
            Recognition of the increasingly unwieldy and ineffectual arrangements of the GATT reinvigorated interest in the creation of a full-fledged international trade organization—an organization that would fulfill, at least in very general terms, the original intentions behind the ITO. Significantly, the United States was not, at first, receptive to the idea of creating a trade organization. In particular, U.S. policymakers had a number of concerns about the scope and degree of authority that a new international trade organization might have. The United States had, for example, no interest in having labor standards, commodity agreements, or monopolistic business practices included in the organization’s charter (Narlikar 2005, p. 26). In addition, the U.S. demanded additional trade concessions from the European Union before it would drop its opposition; the U.S. even demanded a name change—from the proposed “Multilateral Trade Organization” to the “World Trade Organization” (Narlikar 2005, p. 25). Only when the United States got what it wanted was it possible for serious discussion on the establishment of the WTO to move forward; this discussion, not coincidentally, took place under the auspices of the final GATT round, the Uruguay Round, which lasted from 1986 to 1993.

A major reason why the GATT had become unwieldy and ineffectual was its ad hoc nature. The many overlapping negotiations of the GATT had produced a range of agreements and measures, not all of which were consistent with one another. Thus, some sort of coordinating mechanism was required for creating order out of an increasingly chaotic situation. Creating a coordinating mechanism, however, was (in practical terms) not possible within the existing GATT framework. As a forum rather than an organization, moreover, the GATT could not directly coordinate its activities with international organizations such as the IMF and the World Bank—a function that was becoming increasingly necessary as the global economy had become more complex and interconnected. Even more basically, it is important to understand that reducing tariffs on manufactured goods was a relatively easy part of trade liberalization—the rest was far more difficult, as the issue with NTBs demonstrated. Indeed, the increased use of NTBs gave rise to the phrase new protectionism in the 1980s to indicate that a significant and potentially destabilizing change was taking place in the world trading system. In sum, then, the GATT had served its purpose, but without a solid institutional basis it might not have been able to sustain the progress that had been made. The significance of the WTO, therefore, would be found in its ability to reinforce the existing international trading system, but also to extend liberalization into the most politically sensitive and divisive areas.

On the first point, the WTO did, in fact, play a key role in reinforcing the existing system. One of the most important new features was the adoption of the single-undertaking principle. This principle required member countries to accept and implement all WTO agreements as a package rather than through a pick-and-choose method, which had been the practice under GATT. For the most part, this also meant that there would be no grandfathering of rights; that is, countries that had previously been exempted from certain agreements and articles because of existing (domestic) legislation were no longer allowed exempt status. This was a major change, in that it compromised the principle of state sovereignty by requiring member countries to change their domestic legislation if there was a conflict with WTO rules. Within the framework of the WTO, it is important to add, this was possible due to a Dispute Settlement Panel (DSP) that had been much strengthened via the Dispute Settlement Understanding (DSU). The new DSU made WTO rules easier to adjudicate, and easier to enforce through the principle of cross-issue retaliation (cross-issue retaliation allows a member country or countries to apply WTO-approved sanctions against a violator of an important area of trade in order to maximize the impact of the punishment).

More broadly, the DSU gave the WTO a power that few other international organizations had—a compulsory, legally binding process for resolving conflicts between member states. The compulsory element of the DSU was and is key: unlike the GATT (which had its own dispute-settlement procedures), or other international organizations that have a judicial process, the DSU did not require the consent of both parties to bring or hear a complaint. As you might guess, requiring consent from both parties to move a case forward can be a monumental obstacle. Under GATT rules, moreover, the (positive) consent of all parties—even the losing party—was required to make any decision legally binding. Under the new WTO rules, the “positive consent” principle was replaced with the “negative consent” principle. According to the latter, a ruling (or report) can only be rejected if all members decide by consensus not to adopt the report. Another important element of the DSU is the obligation of member countries to refrain from using unilateral measures to settle trade-related disputes; instead, they must bring their disputes to the WTO. All these and other rules raise an important question: Why would states voluntarily give up their (sovereign) rights and be bound by the rules and procedures of an international organization? The simple answer is this: the benefits of belonging to that organization outweigh the costs. On this point, recall the discussion of international institutions in chapter 3: international institutions allow states—operating in an environment of anarchy—to achieve cooperative goals that would be extremely difficult, if not impossible, to achieve on their own. Most generally, then, the rule-based framework of the WTO is widely seen as bringing economic benefits through cross-border trade that compensate for the diminution of autonomy and state sovereignty.

 

 

 

 

 

 

 

 

 

 

Figure 4.12. Map of the WTO Members and Observers (2013)


 
  Members

   Members, dually represented by the European Union

   Observers

   Nonmembers

 

Image: The copyright holder has released the image into the public domain.

 

 

            The WTO has been far less successful in extending liberalization into the most politically sensitive and divisive areas. This is clearly reflected in the current round of trade negotiations (under the WTO, high-level trade talks take place in ministerial conferences held every two years), known as the Doha Development Agenda, or the Doha Round for short. The Doha Round began in 2001, and was designed to take up a number of difficult issues, including agriculture, services, intellectual property regulation, environmental agreements, electronic commerce, regional trade agreements, transparency in government procurement, and trade facilitation, among others. The results have not been pretty: after the initial Doha meeting, the next ministerial conference in Cancún (2003) collapsed after just four days, and subsequent meetings in Hong Kong (2005) and Geneva (2009 and 2011) also failed to reach agreement. Things were so bad at one point that the biennial ministerial meeting in 2007 was cancelled (in addition to the ministerial conferences, there were also a number of lower-level, that is, nonministerial, meetings). The failure to reach an agreement during the Doha Round has raised the obvious question: Is Doha dead? As this book was being written, however, the answer appeared: no. In December 2013, there finally was a breakthrough in negotiations; although not completely resolved, preliminary agreement was reached on some particularly sticky issues, although it still remains to be seen how far things will move (see chapter 7 for additional discussion).

As suggested above, however, the obstacles are not only related to difficult-to-negotiate issues, but also to (1) a dramatic expansion of membership (by the Doha Round, the WTO had 155 members); (2) a more pronounced division among “developed” and “developing” countries; (3) consensus decision-making; and (4) an all-or-nothing principle (based on the idea of a single undertaking wherein, in effect, “nothing is agreed until everything is agreed”). Critics, of which there are many, point to the recent difficulties faced by the WTO and argue that the organization—after less than two decades—has already become irrelevant. This argument could only be valid if the pre-existing and still-robust system of trade relations is ignored, which would be an absurd thing to do. Even if the focus is shifted entirely to current issues (and the current impasse), however, the argument for irrelevance does not carry much weight. After all, without an established institutional framework for multilateral negotiations, there likely would be no multilateral negotiations at all. Certainly, we could not expect agreements covering hard issues to automatically appear. This takes us back to the broader point: creating a framework for cross-border trade or international trade is a profoundly political process. More specifically, in a world of ostensibly sovereign states, free or liberalized trade requires a political-institutional framework. The creation of such a framework does not guarantee constant “progress,” but it makes progress—in this case, liberalization—possible and sustainable. This, again, tells us that studying the economic without the political (and vice versa) is a hollow practice.

 

Politics within the WTO: Bargaining Coalitions

            If we look inside the WTO, it becomes almost immediately apparent why the latest round of trade negotiations has been so painfully protracted: strong disagreements over particular trade policies and issues are exacerbated by profound divisions between “developed” and “developing” countries (also referred to as the North-South division). Keep in mind, on this point, that until the Kennedy Round in 1964, there were relatively few “contracting parties”: the total number of participating countries ranged from a low of 13 (Annecy Round 1949) to a high of 36 (Torquay Round 1950). More importantly, while there were developing countries represented in these early negotiations, they largely stood on the sidelines or were effectively co-opted by the developed countries. This has changed dramatically over time as developing countries have become more numerous (they now account for 75 percent of WTO membership), more assertive, and less beholden to the wealthier, developed countries. But standing alone, individual countries in the developing world had little capacity to exercise power, even with the decision-making-by-consensus rule in the WTO. To effectively exercise power, then, it was necessary for developing countries to adopt a unified and collective stand. This is more difficult than it sounds. After all, the so-called developing world is tremendously diverse, and does not always have a consistent set of interests and concerns.

In WTO negotiations, the primary means to overcome this diversity has been through the formation of bargaining coalitions—many of which predated the WTO and some of which originated outside the GATT/WTO framework—some targeted to specific issues and others more broadly based. Among the many coalitions are the G77, the G90, the Like-Minded Group (LMG), the NAMA 11, Café au Lait (also known as the G20), the African Group, the Africa-Caribbean-Pacific (ACP) group, Caricom, the group of Small and Vulnerable Economies, the Cairns Group, the Cotton Initiative, and so on. There is not space to provide a detailed discussion of bargaining coalitions here, but suffice it to say that they have become a central feature of the negotiating process in the WTO, especially in the ongoing Doha Round. The impact of bargaining coalitions, however, is complex. As Narlikar (2003) and Rolland (2007) have noted, there are several types of coalitions: blocs, issue-based, and regionally based coalitions. Blocs are the largest and most diverse coalitions (e.g., the G77 and G90); they tend to play a negative role by stalling or blocking certain initiatives. Issue-based coalitions are smaller, and more focused, as the label suggests, on specific issues. The G20, or Group of 20, for example, was established to push for the liberalization of Western agricultural markets. In general, these have been the most effective coalitions. Regionally based coalitions typically come out of pre-existing regional trade agreements, such as ASEAN (Association of Southeast Asian Nations). These coalitions have had limited success, in part because their economic interests tend to conflict with one another, so group consensus is hard to maintain. (The one major exception, however, is the European Union—a point that is discussed below.) Overall, developing-country coalitions have changed the dynamics of WTO negotiations: they have proven to be a viable and influential vehicle for articulating the interests and demands of developing countries, and they have made the developing world a force to be reckoned with. Yet, it is precisely because of their effectiveness that negotiations in the WTO have stalled and remain in danger of complete collapse.

Developing countries, it should be stressed, are not necessarily the problem. After all, it takes two to tango. Consider the issue of agriculture: for their own, primarily domestic political reasons, many developed countries have steadfastly refused to liberalize their agricultural markets. The G20, which emerged primarily as a reaction to EU and U.S. intransigence on agricultural liberalization, played the lead role—in the Cancún meeting (2003)—in demanding that the European countries and the United States dramatically cut their domestic and export subsidies and provide greater market access. Although there were some signs that substantive negotiations over agriculture might take place, the overall trade talks collapsed over another set of issues, the so-called Singapore issues, which dealt with transparency in government procurement, trade facilitation, investment, and competition. Not surprisingly, the Singapore issues were a main concern of developed countries, especially the EU, Japan, and South Korea (over the years, South Korea had moved from developing-country status to developed-country status). In green room discussions, no consensus between the developed and developing countries could be reached on the Singapore issues (which happened to be the first agenda item), so the Mexican chairperson brought all negotiations to a close before the talks on agriculture and market access could even begin.

 

The WTO and Transnational Actors

            The discussion thus far has treated the WTO as if only countries or governments make decisions. Yet, as is clear from previous discussions of both the liberal and the Marxist perspectives, governments (or states) are almost assuredly not the only important actors: domestic political groups, social classes, and NGOs can be equally, perhaps even more, important. In the case of the WTO, it is not difficult to see how these groups impact politics within the organization. For example, on the issue of trade-related aspects of intellectual property rights (referred to as TRIPS), which was successfully negotiated at the end of the Uruguay Round and is now enforced under WTO rules, pharmaceutical companies played a central role in pushing for an agreement that protected their rights—to the detriment, some critics argue, of developing (and especially the least developed) countries. Specifically, one important element of the TRIPS agreement gives pharmaceutical companies exclusive patent rights on drug innovations for a period of 20 years. While patent rights have long been protected, the controversy surrounding TRIPS is that it denies poorer countries access to drugs vital to maintaining public health. This particular issue was addressed in the TRIPS agreement through two exceptions: (1) compulsory licensing, and (2) parallel importing. The first exception gave member states the authority to grant a license to a third party to produce a patented invention without the consent of the patent holder under certain conditions, like, for example, a public health crisis. The second allows a developing country to take advantage of differential pricing between countries. For example, if a drug costs $200 in Canada, but sells for $300 in South Africa, a South African company or the government can import the drug from Canada and sell it at a lower price without the consent of the South African patent holder. Despite these exceptions, pharmaceutical companies in the U.S. and elsewhere pressured their governments to sanction developing countries that attempted to take advantage of these provisions (Subhan 2006).

One of the most famous cases involved a 1998 suit brought against the South African government by the South African Pharmaceutical Manufacturers Association and 40 pharmaceutical manufacturers, most of them multinationals. The companies alleged that the South African government had violated TRIPS by authorizing the parallel importation of HIV/AIDS medication. At the start of the litigation, corporate interests lobbied their home governments to put pressure on South Africa. Corporate pressure at first worked quite well, as both the U.S. government and European Commission took up the cause of “Big Pharma,” and threatened to withhold trade benefits and impose trade sanctions against South Africa. Very soon, however, public outcry—led by NGOs and AIDS activists—changed the dynamics: then-presidential candidate Al Gore was accused of killing babies in Africa, and “[b]y the time the case finally reached the courtroom in May 2000, the drug companies could no longer count on the support of their home governments” (’t Hoen 2003, p. 44). Continuing public pressure eventually pushed the companies to drop their case. While much of this action took place outside the framework of the WTO, the controversy over TRIPS eventually found its way back into the Doha Round. Developed-country governments reverted back to pushing for changes on behalf of corporate interests, while developing countries—acting in part through coalitions, including the Africa Group—pushed for a larger public interest (for a more detailed discussion, see ’T Hoen 2003). The key point here is that the WTO does not just involve state actors. Indeed, any close examination of the WTO is bound to find a diversity of both state and nonstate actors influencing and shaping the organization in myriad ways.

 

Regional Trade Agreements

            As was noted early in the chapter, free trade is more an ideal than an actual practice. What the GATT and WTO have produced is a liberalized, but managed, system (or regime) of international trade. The GATT and WTO, however, are not the only institutional or political frameworks for international trade. Another major source are regional trade agreements, or RTAs (also referred to as free-trade agreements), which are defined by the WTO as “reciprocal trade agreements between two or more countries”; they include free-trade arrangements and customs unions.[15] According to the WTO, there were 546 notifications of RTAs (counting goods, services, and accessions separately), and 354 in force as of January 2013.[16] Despite their name, not all RTAs are primarily focused on trade. Some are concerned with investment or capital liberalization; some provide the underpinnings to strategic alliances (and are therefore part of a security arrangement); some are meant for domestic economic restructuring; and some are centered on political, as opposed to economic, integration (Whalley 1998). Indeed, it might be fair to say that, for many RTAs, reciprocal trade arrangements are a secondary issue. Whatever the primary goal, however, it is clear that RTAs are an important, and increasingly pervasive, part of the international trading system. Thus, while RTAs have been around for a long time, their growth has accelerated over the past decade. In 1995, for example, there were 124 RTA notifications; by 2005 that number had jumped to 330, a more than two-and-a-half-fold increase in a decade. The current number represents an almost four-and-a-half-fold increase since 1995.

 

 

Figure 4.13. The Growth of RTAs, 1948–2012     

Source: WTO Secretariat. Copied from WTO website and available at the following address: http://www.wto.org/english/tratop_e/region_e/regfac_e.htm.

 

This rapid growth raises an obvious question: Why has there been a proliferation of RTAs following the establishment of the WTO? From an economic standpoint, one logical answer is that the rapid growth of RTAs reflects frustration with the WTO process: the stalemate in the Doha Round has weakened confidence in the WTO and in broad multilateral negotiations. The solution, therefore, is to pursue a smaller-scale, more manageable approach to trade, which could then serve as the basis, or building block, for larger-scale, multilateral liberalization. In this view, the ideal model is to broaden or expand WTO trade rules within the RTA—this is referred to as a “WTO-plus” regional trade agreement. On the surface, the building-block explanation holds a lot of appeal, but few RTAs actually fit this model. Instead, as Sally (2006) bluntly puts it, a large majority involve “bogus” liberalization (p. 308). RTAs, in other words, tend to undermine multilateral liberalization—that is, they act as stumbling blocks—by creating more complicated trading arrangements between and among “regions” through the application of different rules for different products coming from different points of origin. Few are WTO-plus. (The issue of RTAs as building blocks or stumbling blocks is explored in much more depth by Robert Lawrence in his 1996 book, Regionalism, Multilateralism, and Deeper Integration). If the economic (and specifically liberalization) rationale is not the reason for the proliferation of RTAs, then what is? Some possible answers have already been suggested, but perhaps the simplest explanation is that the overarching motivation is political rather than economic. Of course, this does not tell us much. A slightly more specific explanation is that national governments have found RTAs to be useful and relatively effective economic tools for achieving political ends (Ravenhill 2005).

Consider the European Union, which began as the European Coal and Steel Community (ECSC), one of the first postwar RTAs. Certainly, there was an economic motivation for the ECSC, but there was arguably an even stronger political motivation—namely, to build a basis for lasting peace in Europe by consciously integrating the German economy, first into a regional economic arrangement, and ultimately into a regional political federation. To put it in more academic terms, underlying the ECSC was a security linkage. There were other political objectives, too, including creating a regional economic bloc that could negotiate on more even terms with the United States; in this regard, the RTA was used to increase multilateral bargaining power (this parallels the motivation behind bargaining coalitions within the WTO). The use of RTAs as a bargaining tool is, in fact, likely one of the more common reasons for their proliferation since 1996. In addition, for smaller-market countries, RTAs are an important method of guaranteeing access to larger markets. There is, of course, always a trade-off: to gain this access, the smaller-market economies must make numerous concessions to the larger-market countries. In a U.S.-Canada agreement, for example, Canada was able to secure an exemption from the use of anti-dumping and countervailing duties by U.S. producers, while the U.S. demanded that Canada maintain energy and investment policies favorable to the U.S.; Canada also made changes in pharmaceutical protection laws to parallel U.S. laws (Whalley 1998, p. 73).

There are other political motivations as well, but one of the key points is this: RTAs are not inherently vehicles for greater trade liberalization on a global scale. Instead, they are discriminatory (or preferential) trade arrangements that can easily contradict the nondiscrimination, or MFN, principle embedded in the WTO. This has created a great deal of tension, despite the fact that WTO rules explicitly allow for RTAs and can, at times, be applied in a manner that overrules certain RTA practices. Fiorention, Verdeja, and Toqueboeuf (2006) put the problem this way:

The tension in the RTA-WTO relationship has extensive ramifications and may pose a threat to a balanced development of world trade through increased trade and investment diversion, particularly if liberalization on a preferential basis is not accompanied by concurrent MFN liberalization; it also poses a threat to the business community and to the global production system on which it operates by raising costs through regulatory complexity and shifting production from comparative advantage to “competitive preferences.” (p. 26)

            The RTA-WTO relationship raises important and interesting questions about the future of the system of international trade. Will liberalization at an international or global level continue to unfold—that is, will the world continue to move closer to a situation of international free trade? Or does the popularity of RTAs mean, instead, a balkanization of global trade? There is no clear evidence to show what the answer is.[17] However, a couple of things are fairly clear: continued liberalization in the international system of international trade is far from assured, and whatever the outcome, the process will be profoundly political.

 

Chapter 4: A Quick Conclusion

            The liberalization of international trade is a divisive issue. As we saw in the first part of the chapter, there is still a contentious debate on how “liberal” cross-border trade should be. This debate holds strong despite a general consensus that cross-border trade is itself good, or at least necessary. Thus, states (or economic actors within states) continue to trade, and cross-border trade has grown tremendously, especially in the postwar period. Nonetheless, we know from historical and contemporary experience that cross-border trade will continue to be defined, to a significant extent, by political boundaries, and that the existence of these boundaries means managed, rather than free, trade. Managed trade, however, is not necessarily a negative term: it reflects the outcome of complex processes and relations of power, all of which play out within domestic, international, and global structures. In IPE, the study of these processes and relations of power is critical to understanding the shape of the global economy. This does not, it is important to re-emphasize, necessarily imply state dominance; indeed, it means something quite different. “Relations of power” tell us that we need to be attuned to, for example, state-state interactions, state-firm interactions, and firm-firm interactions. Nor can we ignore the power and influence of a plethora of other actors, both inside and outside the state: international organizations, organized labor, NGOs, citizen movements, and the like. The interactions among all these actors produce results that are not generally predictable. The type of cross-border trade regime that exists today, therefore, was not inevitable, just as so-called free trade is not inevitable or natural. 


Chapter 5

 

 

 

 

The Global Financial System

 

 

 

 

 

Introduction

            In the last chapter and through much of this book, the United States has been a major focus of attention. There is good reason for this. Beginning in the early part of the 20th century, the United States—not necessarily as a monolithic entity, but as a complex and diverse set of actors and institutions—gradually emerged as the predominant economic and political player in world affairs, especially in the capitalist world. The power of the United States, it is important to re-emphasize, was more structural than coercive. That is, U.S. power was based on its increasingly strong positions within the security, production, knowledge, and financial structures, positions that were considerably enhanced by the destructiveness of World War II. In the aftermath of World War II, U.S. dominance in the production structure in particular enabled the United States to dictate or shape—to a significant extent—the rules of the game for international or cross-border trade for a good part of the 20th century. Control of this process was, of course, never total, even in the early postwar years, but it is fairly clear that the major institutions of the international trade regime reflected and continue to reflect the interests and dominant position of the United States (and to a much lesser extent, Great Britain). In this view, it is not at all difficult to understand why a predominantly liberal set of trade rules developed around manufactured goods, while agriculture, to this day, is governed by a different set of largely protectionist principles. This double standard, to put it simply (and admittedly simplistically), reflected the interests of a hegemonic power, which pursued liberalization in those areas in which the benefits were clear, while it eschewed liberalization in those areas in which the benefits were less clear-cut. Such are the prerogatives of the hegemon.

            In the global financial system (which is composed of two tightly connected elements, discussed shortly; for now, though, just think of currency issues as one element, and credit issues as the other element), the same basic dynamic was visible in the early postwar years. It is obvious that the United States—again, with Great Britain playing an important, but secondary role—took immediate charge of writing the rules for a new international monetary system (IMS), which is the part of the global financial system involving the exchange of national currencies. For the postwar IMS, the basic framework was explicitly negotiated at Bretton Woods in 1944. One product of these negotiations was the creation of a modified gold standard, one in which the U.S. dollar was fixed to gold at the rate of $35 to one ounce. There is nothing novel about the gold standard: it has been used and adopted many times throughout history. Importantly, though, it has always failed. The same was true for the postwar IMS, which abruptly collapsed in 1971 when President Nixon declared on national TV that the gold window was closed. The United States’ inability, or steadily decreasing willingness, due to high costs, to maintain the gold-exchange system (GES), as it was also known, tells us that states—even ostensibly hegemonic states—do not have unlimited power in key structures of the global economy. This does not mean that the international monetary system, any more than the system for international trade, can somehow operate smoothly and efficiently without an overarching political framework or foundation. In the 1930s, there was no hegemonic power to maintain the international monetary system (one that also used a gold standard). As a result, the system collapsed, and was replaced by a series of relatively closed currency blocs (Helleiner 2005, p. 153), which exacerbated economic and financial problems around the world.

At the same time, a closer examination of the IMS (and the global credit system) tells us that there is a powerful economic logic at work, one that shapes and is shaped by the action and power of states and a variety of other key actors, especially transnational financial institutions (both public and private). Thus, understanding the global financial system quite predictably requires that careful attention be paid to the interdependent relationship between politics/power on the one hand, and monetary and financial forces on the other hand. It also requires that a number of key questions be addressed. Why have previous attempts to develop a stable international monetary system, especially through the establishment of gold standard, failed? What are the costs (and benefits) of a more flexible, but less stable system? What should the relationship between different national currencies be? Who or what controls the creation and allocation of credit in world financial markets? How is power distributed and exercised within the international or global financial system? These are all fundamental questions, the answers to which will shed considerable light on the dynamics of the global financial system.

With this in mind, the first part of this chapter will cover important background issues and key definitions.

 

The Global Financial System: The Basics

            The global financial system, as noted above, can be divided into two separate, but tightly inter-related systems: a monetary system and a credit system. The international monetary system, in the most general sense, is defined by the relationship between and among national currencies. More concretely, it revolves around the question of how the exchange rate among different national currencies is determined. The credit system refers to the framework of rules, agreements, institutions, and practices that facilitate the transnational flow of financial capital for the purposes of investment and trade financing. From these two very general definitions, it should be easy to see how the monetary and credit systems are inextricably related to one another. These rather dry definitions, however, do not tell us very much. It is important to delve more deeply into each in order to establish a firmer basis for understanding the dynamics of the global financial system as a whole (and, it is important to add, between the global financial system and the international trading system). To begin, then, it will be useful to focus more sharply on the main components of the monetary and credit systems.

 

Exchange Rates and the Exchange-Rate System

Text Box: Figure 5.1. Yen to U.S. Dollar Exchange Rate, 1998–2013
 
Source: Data from U.S. Federal Reserve Bank of New York; image was generated by indexmundi.com at http://www.indexmundi.com/xrates/graph.aspx?c1=JPY&c2=USD&days=5475. 

Text Box: Table 5.1. Exchange Rates, Major Currencies (July 2013)
Currency Pair	Value (rounded)
EURO (€)-U.S. Dollar ($)	1.3
U.S.$-Japanese Yen (¥)	98.1
British Pound (£)-U.S.$ 	1.5
EURO-Japanese Yen	130.3

            In a primer on exchange rates, Robert Bartley (2003), writing for the venerable Wall Street Journal, began with the following sentence, “The American political elite knows almost nothing about exchange rates. Worse, much of what it does know is wrong.” If Bartley is right, then there is no shame in being a bit befuddled about the exchange-rate system. At the same time, a basic and accurate understanding of this system is not difficult to achieve. The simplest point is this: an exchange rate is the price of one national currency in terms of another. For example, take a look at table 5.1, which shows exchange rates for a few major currencies. The table tells us that, in July 2013, one U.S. dollar ($1) was worth 98.1 Japanese yen (¥), while one British pound (£) was worth 1.54 U.S. dollars. Presently, however, exchange rates tend to vary over time. If you look at a graph of the exchange rate between the dollar and yen (see figure 5.1), to cite one specific relationship, significant variations can be observed: in August 1998, one U.S. dollar was worth 145.8 yen; compared to the rate in July 2013, that is a difference of almost 50 percent. In other words, in August 1998, the yen was substantially “weaker” (the quotation marks are used because a weak currency is not necessarily a disadvantage). What does this mean in concrete terms? Well, say you have $2,000. In 1998, if you had traveled to Japan you could have exchanged that $2,000 for 291,000 yen, but in 2013 that same $2,000 (to keep things simple, disregard inflation) could be exchanged for only 196,000 yen. In short, you would have a lot less Japanese yen to spend in 2013. From a country perspective, Japanese exports in 1998 were considerably less expensive than today. In fact, for a long time—from 1949 until 1971—the value of the Japanese yen was even weaker at ¥360 to $1. Still more, the yen-dollar rate was fixed during this entire period. After 1971, the rate was readjusted as part of the Smithsonian Agreement, but in early 1973, the adjusted fixed rate was abandoned in favor of a floating exchange rate.

            In the foregoing paragraph, several terms were introduced: the exchange-rate system, the fixed exchange rate, and the floating exchange rate. These are all basic terms. An exchange-rate system (or regime) refers, in general, to the set of rules that govern the relative value of national currencies. Fixed and floating exchange rates represent two major types of exchange-rate systems. While the notion of fixed and floating exchange-rate systems suggests a dichotomous separation, in practice, exchange-rate systems exist on a continuum (Stockman 1999, p. 1484): at one extreme is the pure floating (or flexible) exchange rate, while on the other end is the pure fixed (or pegged) rate system. The fixed and floating-rate systems, in this regard, might be better seen as ideal types—that is, purposeful simplifications or abstractions not meant to correspond to all the actual characteristics of a particular case. Scholars use ideal types for analytical purposes, to help us more clearly see and understand the essential characteristics and features of specific phenomena. With this in mind, in a pure floating-rate system, the value of a currency is determined solely by supply and demand; a pure floating-rate system, in other words, exists only when there is absolutely no intervention by governments or other actors capable of influencing exchange-rate values through nonmarket means. These conditions, it should be noted, have never been met; there has always been some degree of government intervention in the determination of currency values. A pure fixed-rate system, on the other hand, is one in which the value of a particular currency is fixed against the value of another single currency or against a basket of currencies, or against another measure of value, such as gold or silver (or some combination thereof). In practice, pure fixed-rate systems can exist, but only on a short-run basis; adjustments are, in practical terms, inevitable. The postwar gold-exchange system, for example, lasted until 1971, but prior to its collapse, there were exchange-rate realignments in 1958, 1961, and 1967 (Stockman 1999, p. 1485).

            In between the pure fixed and floating exchange-rate systems, as already noted, are many variants. The IMF lists eight specific types or regimes, some with quite interesting names: (1) exchange arrangements with no separate legal tender, (2) currency board arrangements, (3) other conventional fixed peg arrangements, (4) pegged exchange rates with horizontal bands, (5) crawling pegs, (6) exchange rates within crawling bands, (7) managed floating with no predetermined path for the exchange rate, and (8) independently floating (there are also other variants). It is not necessary to go into a detailed explanation of each of these regimes (interested readers can find descriptions on the IMF website). The range of options, however, raises a question: is one type of exchange system better than others? The simple answer is no. The U.S. Treasury notes, on this point, that there is “probably no universally ‘optimal’ regime. Regime choices should reflect the individual properties and characteristics of an economy” (Appendix II, p. 1). Most scholars, including mainstream economists, agree.[18] But this takes us back to one of the key questions raised in the introduction—What should the relationship between different national currencies be? To answer this question, we need to recognize a truism in political economy: choices must be made, and typically getting more of something means giving up something else. This is called a trade-off. In the decision over whether to adopt a primarily fixed or floating exchange-rate system, the basic trade-off is easy to discern: stability versus autonomy. Fixed exchange-rate systems generally offer greater exchange-rate stability. This is important, in that it reduces the risks in international trade (because the prices of imports and exports will not fluctuate based on unanticipated changes in the exchange rate), and, in principle, reduces the risk of currency speculation. The price of greater stability, however, is less flexibility or autonomy in dealing with domestic economic issues. With a fixed exchange-rate system, in particular, governments have less freedom to use monetary policy (e.g., adjusting interest rates, expanding or contracting the money supply) to manage the domestic economy. The basic problem is encapsulated in the Mundell-Fleming model (see figure 5.2).

 

Figure 5.2. The Mundell-Fleming Model

 

Text Box:  
Image. Robert Mundell. Permission to use granted under the terms of the GNU Free Documentation License, Version 1.2.
Back in the 1960s, Robert Mundell (who won the 1999 Nobel Prize in Economics) and Marcus Fleming argued that, when a small country tries to maintain a fixed exchange rate in a world of perfect capital mobility (keep in mind that basic economic models often use simplifying assumptions to highlight key points of concern), money stock becomes exogenous. In other words, the stock of money is determined by other variables; in practical terms, this means that monetary policy is rendered completely ineffective as a stabilization policy instrument (Fan and Fan 2002). In addition, the two economists theorized that governments could not simultaneously have an independent monetary policy (i.e., control of the money supply and interest rates), a stable exchange rate (via a fixed or pegged system), and free capital movement. It is possible to achieve two of these objectives at the same time, but not all three; this has been labeled the “impossible trinity” (also known as the “trilemma”). O’Brien and Williams (2007) explain it this way:

 

Governments have to choose their priorities. For example, if a government favours capital mobility to attract investment, then it must choose between a fixed exchange rate which will facilitate trade and investment and an autonomous monetary policy which will support domestic economic conditions. If it chooses a fixed exchange rate, interest rates must support that policy by providing investors with returns which will keep their money in the country. If interest rates do not support the currency and instead target domestic concerns, such as unemployment, investors may move money out of the country, putting pressure on the exchange rate. The exchange rate comes under pressure because investors sell the currency as they move their money into other currencies and assets. This creates excess supply of the currency, lowering its value. (p. 213)

 

            It is important to understand that the Mundell-Fleming model was proposed at a time when most countries had a fixed exchange-rate regime; thus, the model was prospective. By the 1980s, however, as more and more countries shifted to floating-rate regimes, the model’s predictions were, by and large, confirmed. This is one reason for its widespread popularity and acceptance. At the same time, as with almost all economic models, there have been criticisms. Most of these center on the simplifying assumptions of the model; nonetheless, the Mundell-Fleming model has done a very good job of withstanding the test of time (and empirical evidence).

 

 

            Beyond the basic trade-off between stability under a fixed-rate regime and autonomy under a floating-rate regime, a number of more specific advantages and disadvantages can be identified. For example, for a country heavily reliant on exports, such as China, a fixed exchange rate can be used to keep the value of the country’s currency low relative to other currencies. This effectively increases the competitiveness of the country’s exports on a generalized basis, and thus encourages stronger exports and stronger economic growth for the national economy. Fixed exchange rates also encourage greater and more consistent foreign investment: outside investors do not have to worry that the value of their investments will fluctuate based on the value of the local currency; thus, they are more likely to invest. In principle, a fixed or pegged currency can also help to lower inflation rates, again, because there are fewer concerns that the local currency will unexpectedly lose or gain value. For developing economies, in particular, fixed exchange-rate systems are, in a somewhat counterintuitive way, far easier to maintain than a floating exchange-rate system: floating systems generally require stronger financial institutions and more mature markets to properly maintain (Haekal 2012).

The disadvantages, however, can be quite severe. In particular, over time, fixed exchange-rate systems can lead to major distortions in the underlying value of the currency. If this happens, investors and other with financial interests in an economy may suddenly lose confidence, and begin to withdraw their investments en masse (as they try to convert local currency holdings into, say, dollars or euros at the fixed or pegged rate). In this situation, governments may try to prop up the local currency by using foreign reserves. Invariably, this strategy fails, and the currency’s value collapses. The result is a massive financial meltdown, of which there are many real-world examples: Mexico (1995); Thailand, Indonesia, Malaysia, and South Korea (1997); and Russia (1998), to name just a few. In other words, the greatest advantage (i.e., stability) of a fixed exchange-rate system is, ironically, also its greatest weakness. (At the same time, one can argue that China’s re-adoption of a fixed rate helped it to avoid the financial turmoil afflicting most other countries at the time.) In addition to the potential for economic catastrophe, a fixed-rate system generally requires countries to maintain higher-than-average currency reserves, but this can result in inflation because it increases the supply of currency (i.e., the monetary base) in the economy.[19]

            One key advantage of a floating (or flexible) exchange-rate system is that it acts as an automatic stabilizer for the economy. For example, if external demand for a country’s exports decline, this will lead to a decline in overall output and an automatic depreciation in the value of the country’s currency. This, in turn, makes the country’s goods cheaper, which should (in principle) increase exports. In this regard, too, a floating system leads to an automatic adjustment in the balance of payments. As noted above, the floating-rate system gives countries more autonomy with regard to domestic monetary policy, especially in determining interest rates. In a fixed-rate system, domestic interest rates must be set at a level that will keep the exchange rate within a predetermined band; in a floating-rate system, this is not necessary. This allows a government, for example, to sharply cut interest rates during a recessionary period to spur domestic economic growth. The main disadvantages of the flexible exchange-rate system, according to Evrensel (2013), are three-fold. The first disadvantage is greater volatility (a point discussed several times already). The second disadvantage is the potential for too much use of expansionary monetary policy, and the third disadvantage of the floating exchange-rate system is that it does not, in the real world, live up to its reputation as an automatic stabilizer. The U.S. is a case in point: despite using a floating exchange-rate system, the U.S. has run large and persistent deficits in its current account.

 

Balance of Payments

            The selection of an exchange-rate system also has important implications for a country’s balance of payments, a point that will be discussed below. But first, a few words about the balance of payments. The balance of payments (BOP) is another basic concept; it refers to the method countries use to account for all of their international monetary transactions (this includes transactions for goods and services, as well as purely financial ones by individuals, businesses, and governments) over a specified period of time. “Every international transaction”, as the Federal Reserve Bank of New York (n.d.) explains it, “results in a credit and a debit. Transactions that cause money to flow into a country are credits, and transactions that cause money to leave a country are debits. For instance, if someone in England buys a South Korean stereo, the purchase is a debit to the British account and a credit to the South Korean account. If a Brazilian company sends an interest payment on a loan to a bank in the United States, the transaction represents a debit to the Brazilian BOP account and a credit to the U.S. BOP account” (n.p.). These transactions are further divided into two general categories: the current account, and the capital and financial account (the capital/financial account is sometimes divided into separate accounts). The current account is used primarily to mark the inflow and outflow of goods and services, but also includes earnings on investments, foreign aid, charitable giving, and wages paid to temporary (nonresident) workers. It is referred to as the current account because these transactions mark a short-term or one-time flow of payments or transfers. The capital and financial account is where all cross-border capital transfers are recorded. This includes the transfer of financial and nonfinancial assets such as stocks, bonds, securities (debt), foreign direct investment (FDI), official reserve transactions (e.g., financial assets denominated in foreign currencies or in Special Drawing Rights, also known as SDRs), land, factories, and mines. These are longer-term economic transactions.

            In discussing the balance of payments, it is easy to become confused. One major point of confusion stems from the tendency to speak of a “balance-of-payments deficit (or surplus).” Technically, the balance of payments is always in balance, or zero; that is, if the current account has a deficit, the capital account has an exactly equal surplus. (This is a basic accounting principle, although in practice, there is typically a statistical discrepancy.) Perhaps the easiest way to understand why this is the case is to consider what happens when a country runs a large current account deficit, which is typically the result of an imbalance between exports and imports. In this example, in order to pay for the imports it needs, a country (or economic actors within the country) may borrow money from a commercial bank or from an international financial institution; it could also receive foreign aid money or sell financial or nonfinancial assets. Or in the case of the United States, it could sell Treasury bonds or other government-based securities such as Treasury bills, known as T-bills, and notes (basically a Treasury bond is a type of long-term debt obligation; Treasury bills are short-term obligations; notes are medium-term obligations). These funds are recorded as a credit for the U.S. because other countries are effectively loaning money to the United States. In other words, what might otherwise be considered a liability or debt is a “credit” in the capital and financial account—which is another reason for confusion. It is important to note, too, that strong demand for U.S. securities (and U.S. securities are generally viewed as one of the safest, most secure investments available) may strengthen the relative value of the dollar, making U.S. exports less competitive (thereby worsening the U.S. current account deficit).

From the foregoing example, it should be apparent that the balance of payments is an important issue in international political economy generally, and for individual countries specifically. For the most part, countries see a capital and financial account surplus as a negative: to repeat, a surplus in the capital and financial account means that a country’s debits are more than its credits. More simply, this means that the country is a net debtor to the rest of the world. On the other hand, a country that runs consistent current account surpluses and capital and financial account deficits is a net creditor, which is viewed in very positive terms. China represents a good example of the latter case. For many years, China has been running huge current account surpluses—the figures for 2005 to 2011 are available in table 5.2. In 2008, to cite a particularly significant year, China’s current account surplus was a remarkable $420.6 billion. Primarily as a result of these massive and consistent current account surpluses, China has been able to accumulate a balance-of-payments reserve of more than $3.3 trillion in 2011 (in the balance of payments, the difference between the current account surplus and a capital/financial account deficit is made up for by an increase in reserves, which are foreign assets held by the central bank; reserves are considered a debit). To appreciate the significance of this figure, consider this: China’s reserves are almost three times that of Japan, which has the second largest reserves in the world at $1.3 trillion. China’s immense holding of foreign assets, it is important to understand, significantly enhances China’s power in all the major structures of the global system, but especially in the financial and production structures. Reserves are essentially “money in the bank,” which can be used to purchase anything that the government or country needs, from the best commercial technology to the most advanced military weaponry, and anything in between. Foreign reserves also provide China leverage as the major creditor economy in the world today. China is not on the same level as the U.S. was at the end of World War II, when the United States controlled 70 percent of the world’s monetary gold (Mundell 2012, p. 526), but it is certainly a central player in the global financial system. What this means will be discussed in more detail later in this chapter. For now, it is time to move on to a more substantive examination of the development of the global financial system. This takes us back to the Bretton Woods conference (as well as other key elements of the early postwar period).

Text Box: Table 5.2. China’s Current Account Surplus and Total Reserves, 2005–2011 (in billions U.S.$)

	2005	2006	2007	2008	2009	2010	2011
Current Account	132.4	231.8	353.2	420.6	243.3	237.8	136.1
Total Reserves	831.4	1,080.8	2,183.4	1,966.0	2,452.9	3,254.7	3,331.1

Source: World Bank, “World Development Indicators,” online database available at http://wdi.worldbank.org/ (all figures are rounded).

Constructing the U.S.-Led Postwar Global Financial System

            By now, you are quite familiar with (and perhaps weary of reading about) the Bretton Woods conference. Still, any discussion of the global financial system requires a further examination of its significance, and especially of the larger context and underlying relations of power that shaped the negotiations and agreements reached at Bretton Woods. With respect to the global financial system, one of the key elements of the negotiations at Bretton Woods, as noted above, was the creation of the gold-exchange system, or GES (or, less commonly, the par value system). Two other key elements were the creation of the International Monetary Fund (IMF) and the World Bank (or the International Bank for Reconstruction and Development [IBRD], as it is formally known). To go along with this financial framework, it is important to understand, were frameworks for cross-border trade and for international security. The framework for trade was discussed in depth in chapter 4, so no more must be said here. As for the security framework, suffice it to say that it was important for creating a strong and enduring basis for political stability and order, and was part and parcel of the exercise of hegemonic power by the United States.

It is clear that all of these elements of the postwar order were consciously designed—largely by the United States—as part of an overarching whole. It should be very easy to discern why the United States would willingly, even eagerly, take on this role: in a word, hegemony. Remember that hegemony as an analytical concept and tool has resonance in most political-economy or IPE approaches. Even more, from a theoretical perspective it is very hard—and, arguably, unthinkable—to ignore the significance of hegemony in explaining the major economic and political dynamics of the early postwar period (I emphasize “early” because this is the period in which U.S. power was at its apex, and therefore, it is the period in which hegemony was likely most important). This does not mean that the conceptualization and implications of hegemony are the same in the various theoretical approaches. They are not. In addition to the concept of hegemony, it is crucial to keep in mind the notion of two-level games, for even in the early postwar period, it is clear that domestic political-economic considerations were an integral part of decision-making within the United States (as well as other countries).

 That said, the primary topic in this chapter is the framework of the global financial system, and a good place to begin an examination of this topic is with the negotiations and agreements reached at the Bretton Woods conference of 1944.

 

 

American Power, Bretton Woods, and the Postwar Global Financial System

            Even before the end of World War II, plans were in the works to construct a specific postwar order premised primarily on American power and interests. One of the first orders of business was to re-establish economic and financial order, which was designed, at least in part, to avoid the mistakes of the past. Two contrasting mistakes were, first, the overly rigid gold standard of the 19th century, which did not allow governments to effectively manage domestic economic issues. The second was the disastrous experiment with floating exchange rates in the 1920s and 1930s: one of the problems with the 1930 system was that it encouraged countries to engage in “competitive devaluations” in order to gain a temporary advantage in international markets (Gilpin 2002). Part of the solution, then, was the gold-exchange system, which was designed to provide the best of both worlds. Specifically, the GES was meant to provide the stability of a fixed exchange system by establishing a set value for the U.S. dollar relative to gold (and other currencies relative to the U.S. dollar or gold), but also have the flexibility of a floating system via an “adjustable peg.” Under this exchange regime, participating countries were obliged to declare a specific value for their currency, known as a par value, or peg; they were also required to intervene in currency markets to limit exchange-rate fluctuations with a maximum margin, or band, of one percent above or below parity. (The par value concept, it is worth emphasizing, was originally used to define the Bretton Woods system, which is why it is common to hear people say that the Bretton Woods system collapsed in 1971.) At the same time, all countries retained the right to alter their par value to correct a “fundamental disequilibrium” in their balance of payments (Cohen 2001). In principle, this meant that governments could devalue their currency (beyond the one percent band), but quite unlike the prewar period, devaluations were subject to oversight by a third, supposedly disinterested or impartial, party—the IMF. The IMF, in other words, was given the authority, albeit not unlimited, to approve or reject requests for currency devaluation. This was, in an important respect, a remarkable development in the world of global finance, but it likely could only have happened with U.S. leadership. To see this, consider that, in practice, the IMF was often bypassed in favor of negotiations between the U.S. government and the affected government(s).

            The IMF, which formally began operations on March 1, 1947, had much more to do than simply approve requests for currency adjustments. Indeed, it was meant to play a (even the) key role in the postwar global financial system. Thus, while set up as an ostensibly neutral international financial institution, the IMF was clearly meant to represent U.S. interests and power first and foremost, and the interests of the other major capitalist countries (the developed economies) secondarily. This can be seen, more concretely, in how decision-making power within the IMF was designed, a point discussed in chapter 3. Again, voting power is determined by what the IMF calls a quota. A quota (or capital subscription) is the amount of money that a member country pays to the IMF; it is the price of admission, so to speak, and a central component of the IMF’s financial resources. The quota is supposed to reflect the relative size of a country’s economy; in reality, however, this has never been the case (Bird and Rowlands 2006). China, for example, has the second largest economy in the world, but still has a smaller quota than France, Germany, Japan, and Great Britain. Even more interesting (or telling), China’s quota is only about 35 percent larger than Saudi Arabia’s quota, despite the fact that China’s economy is about 12 times (or 1,200 percent) larger (in terms of GDP). This is important, because quotas also determine the number of votes each member has. All members are automatically entitled to 250 basic votes, plus one for each SDR 100,000 of quota (an SDR is a special type of monetary currency reserve created by the IMF): in practical terms, the allocation of 250 basic votes means almost nothing. On this point, just consider that, in 2013, the U.S. quota (number of votes) was 421,961; compare this to, say, Chad’s 1,403 votes (IMF 2013). When the IMF was first formed, the United States had almost 35 percent of all votes, while the other developed countries controlled more than 40 percent. To further ensure decision-making control, the countries with the five largest quotas were given permanent seats on the IMF’s executive board (composed of 24 total members). In addition, important decisions require a supermajority of 85 percent, which means that the largest members effectively have veto power. None of this should be surprising, especially from a political-economy perspective, which tells us to pay close attention to the question of how power shapes the economic system, whether domestically or internationally.

            Beyond the issue of voting power, the quota system within the IMF was primarily meant to provide a stabilization fund. The IMF’s stabilization fund—which was partly, but not coincidentally, modeled on the U.S. Exchange Stabilization Fund, or ESF (established in 1934)—provided a pool of money available at the international level. This money was loaned, on a short-term basis, to countries suffering from temporary balance-of-payments problems (e.g., a current account deficit). The loans were meant to provide a type of safety valve so that governments would not be tempted to resort to unilateral devaluations of their currencies in an effort to reduce their current account deficits. The stabilization fund, to some extent, worked hand in hand with the IMF’s authority to approve currency devaluation requests. Specifically, if the IMF opposed devaluation, it could not directly prevent a country from devaluating its currency. After all, the IMF had no enforcement arm, no IMF “police force.” Instead, the IMF was authorized by the Articles of Agreement to bar that country from drawing from the stabilization fund. Unlike the par value system, the stabilization function of the IMF not only survived, but has also, over time, come to play a larger and quite significant role in the global financial system. In this regard, it is important to understand that the basic purpose behind the stabilization fund and IMF lending has changed. As the IMF itself explains it:

 

the purpose of the IMF’s lending has changed dramatically since the organization was created. Over time, the IMF’s financial assistance has evolved from helping countries deal with short-term trade fluctuations to supporting adjustment and addressing a wide range of balance-of-payments problems resulting from terms of trade shocks, natural disasters, postconflict situations, broad economic transition, poverty reduction and economic development, sovereign debt restructuring, and confidence-driven banking and currency crises (IMF n.d.).

 

The IMF and Conditionality

            Significantly, along with these general adjustments in the purpose of IMF loans have come other changes as well. One of these changes centers on the recipients. Originally, the IMF was designed to provide assistance to industrialized countries, and for the first two decades of its existence, more than half of IMF lending went to these richer economies. Since the 1970s, however, the vast majority of recipient countries have been from the developing world (although with the global financial crisis beginning in 2008, the IMF again began providing loans to European countries). A related, but more important change has been a stricter and more expansive application of conditionality. Conditionality, in the broadest sense, refers to any condition or requirement attached to the receipt of a loan, as in, “In return for our willingness to extend this loan to you, you are required to meet the following conditions...” When the IMF was first established, there was no reference to conditionality; nor was it written into the Fund’s original Articles of Agreement. The concept, instead, was first introduced in 1952, but not formally incorporated into the Articles until 1969 as part of the First Amendment (Buira 2003, p. 3). One of the reasons for the lack of conditionality in the beginning was Britain’s very strong resistance to the idea. Britain’s negotiating team, led by John Maynard Keynes, was explicitly instructed not to accept any wording that would even suggest conditionality—the fear was that doing so would invite the “evils of the old automatic gold standard” (cited in Moggridge 1980, p. 143). The U.S. was willing at the time to accede to Britain’s position. Still, as soon as the opportunity arose, the U.S. used its position of advantage to push through its original desire to attach conditions to IMF loans. This original desire was summarized by Keynes, who pointed out that the U.S. wanted the IMF to have “wide discretionary powers” and the ability to exercise “influence and control over the central banks of member countries” (cited in Buira 2003, p. 2).

After conditionality was first incorporated into the IMF, it was generally limited to monetary, fiscal, and exchange policies—that is, conditions that were directly related to balance-of-payments issues. But beginning in the 1980s, conditionality began to extend well beyond balance-of-payments issues to “encompass structural change in the trade regime, pricing and marketing policy, public sector management, public safety nets, restructuring and privatization of public enterprises, the agricultural sector, the energy sector, the financial sector, and more recently to issues of governance and others in which the expertise of the Fund is limited” (Buira 2003, p. 19). The IMF, in short, began to encroach more deeply and significantly into issues of state sovereignty. Not surprisingly, this has made the practice of conditionality a deeply controversial and profoundly political practice, with many critics charging that the IMF has become little more than a tool—an extremely effective one—the United States uses primarily to enforce its will on the rest of world. The IMF, of course, steadfastly refutes the notion that it is primarily or even partly a political tool used by the U.S. or other powerful countries. There are certainly valid points made by all sides of this debate[20]; still, it is difficult to dismiss the argument that asymmetries in political and economic power play a central role in how the IMF deals with countries. As Paul Volcker, former chairman of the Federal Reserve, once put it, “When the Fund consults with a poor and weak country, the country gets in line. When it consults with a big and strong country, the Fund gets in line” (cited in Buira 2003, p. 4). Chapter 7 will provide a more in-depth discussion of conditionality and its impact on developing economies.

 

The Final Push: The Importance of Hegemony

            It should be fairly clear—almost beyond doubt—that the U.S. played a key role in constructing the postwar global financial system. However, building a system and ensuring that it actually works are two very different things. For the postwar global financial system and the GES/par value system specifically, it was immediately apparent that the U.S. needed to do more than forge an agreement on the framework and rules governing that system. In fact, the system created by the United States was not implemented until many years after the final agreement was reached. The problem was clear: in the early postwar period, the major European economies did not have sufficient foreign exchange reserves to implement fully the gold-exchange or par value system. The system, to remind you, required participating countries not only to set a par value, but also to intervene in financial markets to prevent their currency from falling or rising one percent or more outside a preset band (based on the par value). Without sufficient foreign exchange reserves, this was a risky policy to adopt. Why? Because the war-decimated European economies needed to conserve as much foreign currency as possible for reconstruction and other basic needs of the domestic economy (Oatley 2012, p. 216). As a result, most of the European governments were unwilling to adopt the convertibility rules the system required; that is, they were not willing to allow to the free convertibility of domestic currency—francs, deutschmarks, lira, etc.—into dollars or gold. The British, it should be noted, did implement convertibility for a short period in 1947, primarily because it was a condition attached to the Anglo-American Loan Agreement of 1946: in return for a much-needed loan of $3.75 billion (plus an additional $1.2 from Canada), Britain was required to restore the convertibility of the pound sterling (this was an early form of conditionality, which is discussed below). Less than a month after restoring convertibility, however, Britain’s foreign reserves were drained of $1 billion; this forced the British government to suspend convertibility of the pound sterling.

The solution was quite simple, but politically very difficult (see figure 5.3, “Selling the Marshall Plan,” for further discussion): the U.S. would need to export U.S. dollars to Europe at a hitherto unprecedented level. Thus was born the Marshall Plan (or, as it was officially called, the European Recovery Program), which transferred $13 billion to European allies between 1948 and 1951, on top of an equal sum already provided to Europe in the years following the end of the war. While $13 billion does not sound like much today, at the time it constituted 5 percent of U.S. GDP ($258 billion), albeit spread over several years. The Marshall Plan helped the European economies to rebuild, and also eventually stimulated a strong flow of private capital from the U.S. to Europe, such that European governments were able to accumulate large foreign reserves by the end of the 1950s. The primary vehicle for the disbursement of Marshall Plan money was the World Bank, which was originally set up to aid in the (immediate) reconstruction of Europe. The end result was exactly what the U.S. had hoped to achieve: a financially, economically, politically more stable and stronger Europe. With this stability and strength, the European governments were then willing to finally implement the Bretton Woods system in 1959—only for it to fail a little more than a decade later (discussed in the following section).

The key point: the Bretton Woods system was very much a product of American power, and more specifically, of America’s hegemonic power. The system may not have ever been implemented were it not for the capacity and willingness of the U.S. government to use its vast resources to ensure implementation of the system by beholden, but constrained allies. The use of U.S. resources, moreover, was certainly not limited to the Marshall Plan and postwar reconstruction; the U.S. also built and funded an expensive security framework. Outside of Europe, too, the U.S. was very busy: Japan’s postwar reconstruction was another “responsibility” taken on by the United States. At the same time, hegemony did not determine what the outcome(s) would be. Domestic politics, both within the United States and in countries throughout the world, played out according to their own logic (recall the discussion of two-level games). This is par for the course when discussing political-economic issues. It is also important to emphasize that the system was very much a social construction. For the U.S., in particular, a new reality of internationalism, as opposed to isolationism, had to be defined. For the objective fact of American economic and military dominance does not necessarily translate into a globally oriented policy requiring major expenditures for the rebuilding and defense of other countries. In addition, this new reality required the elevation of ostensibly power-neutral international financial institutions (i.e., the IMF and the World Bank) to quasi-sovereign status. The IMF especially was made into the authoritative (albeit not unchallenged) voice for the entire international financial system. This privileged position allowed the IMF to translate its views—particularly on critical issues such as balance-of-payments problems—into reality. Barnett and Finnemore (2004) argue, for example, that, beginning in the 1950s, the IMF developed a monetary approach to balance of payments that essentially defined the “problem as one in which both the cause of and the solution to balance-of-payments problems lay in the deficit state” (n.p.); this became the justification for conditionality and for the IMF’s steadily increasing intervention in the domestic affairs of member states.

 

Figure 5.3. Selling the Marshall Plan

 

Text Box:  The Marshall Plan played a critical role in ensuring the early success of the Bretton Woods system, and also in stabilizing America’s overall postwar framework. From the beginning, however, opposition to the plan was intense. As Barry Machado (2007), writing for the George C. Marshall Foundation, explains it, most of the opposition came from the Republican Party, which generally saw the Marshall Plan as unnecessary and wasteful, and certainly not necessary for America’s prosperity or long-term security. Howard Buffett (the father of Warren Buffet, the second richest person in the United States in 2013), who represented Nebraska’s Second Congressional District in 1948, was a particularly strident foe. He labeled the plan “Operation Rathole,” and condemned the “barrage of propaganda ... drench[ing] this country” as the Truman administration attempted to gain support for its passage (p. 15). It is important to understand that this was a time when the U.S. had yet to throw off its long-standing and deeply embedded preference for isolationism.

Given the level and degree of opposition, passage of the plan was far from certain. Indeed, the Truman administration engaged in an “intensive five-month campaign of discussion, debate, and persuasion” that ultimately won broad public endorsement for the Marshall Plan. Of course, as Machado also notes, “the Truman administration’s willingness to concede a great deal to the
concerns and biases of Congress in jointly crafting the final, compromise version of the ERP bill secured additional votes. Congress was always actively engaged in the process of revision, and the enabling legislation bore numerous congressional fingerprints: the program would not be run out of the State Department, its director would be a respected businessman from the private sector, appropriations would be for one year only with annual reviews of how money was spent, guidelines and safeguards for disbursing funds would be imposed, aid would be denied to governments which went Communist, counterpart funds would be required, and American shipping would be employed” (p. 21).

 

Image: One of a number of posters created by the Economic Cooperation Administration (ECA), an agency of the U.S. government, to sell the Marshall Plan. The image is in the public domain, since it was prepared by an officer or employee of the U.S. government as part that person’s official duties.

 

Why Did the Bretton Woods System Fail?

            Given the discussion thus far, it may seem odd that a key element of the Bretton Woods system, the par value system (or the GES), collapsed after a relatively brief period of time. After all, the U.S. largely set up the system it wanted (including in the areas of trade and security), and as the hegemon, used its resources to considerable effect. The reason the system failed, however, is not difficult to discern. It was even predicted. As early as 1960, Robert Triffin argued that the gold-exchange system was inherently flawed since, for it to work, it depended on the U.S. being reliably able to convert dollars into gold. As long as dollars remained relatively scarce, as was the case in the 1950s, the problem was moot. This was because countries that had dollars would spend them on goods and services. As the European economies recovered, however, and as they began building their foreign reserves (composed mostly of U.S. dollars), the dollar gap turned into a dollar glut. This meant, in part, the supply of dollars had begun to exceed the supply of gold held by the United States: in fact, by 1959, U.S. gold holdings and foreign dollar holdings had already reached rough parity. As the situation deteriorated, with a larger and larger “dollar overhang” (the dollar overhang is simply the amount by which U.S. dollars held overseas exceeded U.S. reserves of gold), foreign countries and other holders of U.S. dollars began to lose confidence in the dollar itself, and specifically in their ability to freely convert dollars into gold.

The solution to the dollar overhang would have been for the U.S. to reduce the amount of dollars in circulation overseas. But this was problematic for two reasons. First, reducing the supply of dollars would have entailed a significant cutback in domestic spending and an increase in interest rates, neither of which the U.S. government was willing—or, as hegemon, able—to do. The U.S. would have also had to stop running a deficit in its current account. This led to the second, perhaps more significant, problem: by closing the current account deficit, the U.S. would have effectively reintroduced a shortage of the world’s de facto reserve currency; the result would have been a liquidity shortage and a contractionary spiral, perhaps resulting in a worldwide recession. The dilemma, to recap, was simple: Keep spending, and sending dollars to the rest of the world, and erode confidence in the dollar and the Bretton Woods system. Stop spending and eliminate the current account deficit, and risk a global recession and instability. Despite widespread recognition of the dilemma from an early stage—Triffin testified before the U.S. Congress on this very issue in 1960—a viable solution could not be found, although there were several efforts made. One of these was the creation of the London Gold Pool in 1961, in which the U.S. and seven European countries agreed to cooperate in defending a gold price of $35 an ounce through coordinated interventions in the London gold market. The London Gold Pool collapsed in 1968. Another effort, based on Triffin’s suggestion, was the creation (in 1969) of an entirely new reserve unit—the SDR, or special drawing right. This effort, however, also failed to avert the U.S. decision to end the GES just two years later.

Richard Nixon pursued a third effort when he took office in 1969. His approach was to place the blame squarely on other governments, especially Germany and Japan, both of which had begun to run current account surpluses (specifically trade surpluses) and accumulate foreign reserves starting in the mid-1960s (see table 5.3). The Nixon administration pressured its West European allies and Japan to more actively support the dollar in the foreign exchange market, and to reduce their trade surpluses by importing more from the United States (Eichengreen 1996, p. 130). This policy, too, worked for a short time, but the American allies had their own domestic concerns and interests, and were unwilling to accept too much pain in order to appease the United States. Japan, in particular, was in the midst of its phenomenal postwar economic rise, and was ill-disposed towards slowing its export boom. Indeed, in 1971—the same year Nixon suspended convertibility of the dollar into gold—Japan achieved its largest trade surplus ever. That year, too, Japan’s reserve holdings shot up by $10.8 billion, an astronomical sum relative to the previous years (in 1970, for example, the increase was $903 million). As the dollar overhang increased, confidence sank. This fueled speculative attacks against the dollar (speculators were betting on a major devaluation of the U.S. dollar), which made it even more difficult for foreign governments to support the United States. In May 1971—to cite the most egregious case, at least from the standpoint of the U.S.—Germany formally left the Bretton Woods system, because it was not willing to devalue the deutsche mark any further to support the dollar. The “Nixon shock” (as it was dubbed), therefore, was a surprise only in the sense that he failed to consult with any American allies before he suspended convertibility (Nixon also imposed a 10 percent tax on imports).

 

 

 

 

Table 5.3. Trade Balance and Reserves for Germany and Japan, 1963–1969 (all figures in millions U.S.$)

 

 

1963

1964

1965

1966

1967

1968

1969

Germany

 

 

 

 

 

 

 

     Trade Balance

2,402

1,300

2,956

5,252

5,726

5,676

5,133

     Change in Reserves

--

-133

-617

418

329

1,332

--

Japan

 

 

 

 

 

 

 

     Trade Balance

-166

375

1,901

2,275

1,160

2,529

3,699

     Change in Reserves

37

121

108

-33

-69

886

605

 

Source: OECD Economic Surveys: Germany, various years; OECD Economic Surveys: Japan, various years.

 

The collapse of the GES, or par value system, tells us that economic forces and processes cannot be ignored. It also tells us that hegemonic power has clear limits. The United States could not simply dictate the outcomes it wanted, although its influence was certainly felt, and felt quite deeply. On this point, though, it bears repeating that the expansion and stability of international or global capitalism in the postwar period required new and strong frameworks—for finance, trade, and security. Without these frameworks it is not difficult to imagine that postwar economic recovery within the capitalist world would have taken much longer, been much more uneven, and remained dangerously unstable, just as it was for much of the first part of the 20th century. Creating these frameworks, even if imperfect, was therefore a critical task. But it was also a task that only a hegemonic power could effectively take on—although many scholars argue that the Nixon shock marked the end of American hegemony. This tells us, too, that an understanding of the postwar global financial system requires an examination of the intersection and interaction between politics and economics. Thus, just because the GES or the Bretton Woods system collapsed does not mean that, all of a sudden, politics and power no longer mattered. The shift to a floating or flexible exchange-rate regime (among the major economies), more specifically, did not mean that pure market forces had inserted themselves into and completely taken over the global financial system. This certainly was not the case, and even if it had been, it is important to remember that a free market rests on a political edifice. Indeed, the end of the GES did not even entail an immediate switch to a floating system; instead, for a couple more years, the global financial regime was based on an adjustable peg system. The floating exchange regime emerged in 1973.

Still, what exactly did the transition to a floating or flexible exchange-rate regime mean for the global financial system? What were the implications of this transition for the global economy and for individual countries? The next section will address these questions.

 

The Floating World

            Different exchange regimes, as you know, each have their advantages and disadvantages; there are always trade-offs. One of the most salient trade-offs, at the most general level, is between (relative) financial stability and instability, or volatility. In the 1970s, however, a significant degree of stability was maintained, in part because governments (and their central banks) continued to intervene in foreign currency markets to prevent overly large swings in the value of their currency. Moreover, there were no significant incidents of competitive devaluation, which might have sparked a trade war among the major economies. In fact, international trade continued to grow strongly throughout the 1970s, and cross-border investment began to take off since, without the need to maintain fixed rates, countries were more willing to loosen capital control regulations. Japan, for example, liberalized outward investment regulations for its banks in the early 1970s, which helped set the stage for an unprecedented expansion of outward or foreign investment beginning in the 1980s. In short, everything seemed to be going along smoothly. However, there was one contingent, but very important, development: the drastic increase in oil prices, the first of which took place in 1973–74, as a result of an embargo carried out by the Organization of Petroleum Exporting Countries (OPEC) against Western countries for their support of Israel during Yom Kippur War. (OPEC also reacted to the end of Bretton Woods by announcing that oil would be priced in terms of gold, rather than the U.S. dollar.) The details of this complex event are beyond the scope of this chapter. Suffice it to say that the embargo had a major impact worldwide, but also had the unintended effect of maintaining the status of the dollar as the world’s top currency: because of the drastic increase in the price of oil, billions more dollars flowed into OPEC states—more than these states could spend. This immense windfall of oil money—dubbed petrodollars—was deposited into Western banks, and then “recycled” largely in the form of loans to developing countries, which (ironically) were starved for hard currency as a result of the oil-price increases.

Text Box: Figure 5.4. Average Crude Oil Prices, 1939–2008 (in U.S.$ at 2008 prices)

 

Source: BP Statistical Review of World Energy. Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
            While the short-term effects of the switch to a floating or flexible exchange-rate regime seemed to cause little disruption, the longer-term effects have been much less innocuous. In some respects, the predictions of the fiercest pessimists have come to fruition. To wit, the global financial system has become extraordinarily volatile, unstable, and dangerous; it is increasingly fueled by speculation at a level that likely few people—even those familiar with the problems of the 1920s and 1930s—ever imagined was possible. Not everyone, of course, has been surprised by this. One scholar who has written extensively on the growing problems of the global financial system is Susan Strange, whose work was discussed in chapter 1. Strange (who passed away in 1998) wrote two fascinating books on the subject, Casino Capitalism (1986) and Mad Money (1998); although dated, both deserve careful reading by anyone interested in international political economy, and especially in the dynamics of the global financial system. In Casino Capitalism, Strange explained how a series of steps—decisions and nondecisions on the part of major state actors—laid the groundwork for a global financial system increasingly characterized by speculative activity; indeed, Strange asserted that global finance had become very similar to a game of chance, or gambling. Even more, Strange argued that the entire global financial system had become one huge casino, but unlike in regular gambling, the entire world is forced to play or at least bear the consequences of a losing bet.

One of the most important decisions leading to all this, according to Strange, was the “extreme withdrawal by the United States from any intervention in foreign exchange markets” after 1971 (Strange 1998, p. 6). In addition, at the domestic level, decisions by the U.S. government to begin a process of deregulation in the financial sector “freed” banks from their traditional activities of, well, banking (i.e., taking deposits and making loans); commercial banks became investment banks and increasingly started using their own capital in the global financial “casino.” In this regard, though, it might be better to call banks massive “institutional gamblers.” Strange, it is fair to say, was quite prescient, especially in light of the collapse of the U.S. housing bubble in the mid- to late 2000s, which was rife with all sorts of creative, but extremely risky investment vehicles. Still, not everyone would agree with Strange’s basic analysis, as Strange herself readily acknowledged: she devoted a whole chapter in Casino Capitalism on divergent views (chapter 3).[21] One thing, however, is clear: speculative activity has increased significantly, a point that is partly evidenced by the radical increase in currency trading since the early 1970s: in 1973, daily foreign exchange trading averaged around $15 billion; by 1998, this figure had grown to $15 trillion. As Helleiner (2005) points out, the latter figure vastly exceeds the amount needed to service regular international trade and investment flows (p. 161), which strongly suggests that most currency trading since the 1990s has had little to do with basic or nonspeculative international economic activity.

 

Global Finance in the 1980s and 1990s: The Decline of Hegemony

            Recent developments in the global financial system are of obvious relevance to this chapter, but before covering these, it will be useful to go back to the 1980s and 1990s. This was a crucial period for a number of reasons. One particularly salient reason is this: the 1980s and 1990s marked a period of hegemonic decline, or at least erosion, in terms of the economic position of the United States. Japan and Germany, in particular, emerged as major economic competitors, and the formation and development of the European Union as a whole created a significant economic and political counterweight to the United States. In addition, this was a period in which significant parts of the hitherto marginalized developing world began to function more independently. OPEC is one prominent example, but there was also a more general (albeit far from radical) shift in the relations of power between the developed and developing world. As with OPEC, the shift was most evident in cases where developing countries were able to leverage their power through bargaining coalitions and other forms of collective action. In general, the decline or erosion of hegemony tells us that new modes of cooperation (e.g., a more significant role for multilateral institutions) would be necessary, but also that cooperation and coordination would be more difficult to achieve. The flipside to this would be the increasing likelihood of significant conflict between and among major state actors, but also among an increasing array of transnational and nonstate actors. To a significant extent, this is exactly what transpired.

            The decline of U.S. economic dominance was clear. In a two-and-half-year period between 1980 and 1982, the U.S. experienced its deepest recession since the Great Depression, and despite a recovery beginning in 1983, the country began to run historically large current account deficits. In 1981, the U.S. had a current account surplus of $6.3 billion, but in the following years the surplus turned into deficits of $8.3 billion (1982), $40.8 billion (1983), $101.5 billion (1984), and $124.3 billion (1985) (all figures cited in OECD 1986). This was after decades of a near balance in the current account for the U.S. Not coincidentally, as the U.S. current account deficit became larger and larger, the current account surplus for Japan and Germany grew dramatically, primarily because of growing trade imbalances. The trade imbalance, however, was not the only reason for the U.S. deficit. As the hegemon (even if in decline), the United States, from the very beginning of the postwar period, played and continues to play the role of “defender in chief” for the capitalist world (or in terms of the Cold War rhetoric still used today, the “Free World”). In other words, the U.S. not only built, but also continued to maintain, the postwar security framework at a very high cost throughout the 1980s and 1990s. Even at the end of the first decade of the 2000s, the United States pays for more than 22 percent of NATO’s budget, which was $430.4 billion in 2010, and contributes another $84.1 million (21.7 percent) to NATO’s civilian budget. In addition, the U.S. funds 22.2 percent of the NATO Security Investment Program (NSIP) at highly variable amounts: in 2009, the figure was $330.9 million and in 2010, $197.4 million (Benitez 2011).

As long as foreign governments were willing to finance U.S. spending, however, there was nothing to prevent the overspending from continuing. In fact, in the first part of the 1980s, President Reagan encouraged greater domestic spending by increasing the military budget; Reagan also lowered taxes (thus putting more money into the pockets of ordinary U.S. citizens), which increased the national deficit. But to make sure that foreign governments and other investors would continue to buy U.S. debt, the Reagan administration had to maintain high interest rates—which, ironically, were originally put in place to slow domestic inflation and reduce demand. High interest rates attracted foreign investors because they offered an attractive return on investment. This policy worked: capital flowed into the U.S. Yet, this also caused the U.S. dollar to strengthen dramatically: between 1980 and 1985, the dollar appreciated 50 percent. But a stronger dollar meant less competitive exports, and, of course, a worsening trade balance. The ins and outs of this process can get confusing. The key point, though, is fairly clear: by the mid-1980s, the United States had lost its standing as the world’s omnipotent economy. The U.S., instead, had become a debtor country with increasingly uncompetitive firms. Calls for protectionism began to ring out in the halls of Congress. American autoworkers—with their bosses’ approval—began to smash Japanese cars, a symbolic act, but one that reflected American industrial decline rather than American power. (See figure 5.5, “Demons in the Parking Lot.”) A “Buy American” campaign gathered steam, and the domestic political situation within the U.S. was becoming untenable as a broad coalition of automobile and heavy equipment manufacturers, high-tech companies, grain exporters, labor unions, farmers, and others put pressure on the U.S. government. Things seemed very bad for the once unchallengeable hegemon. There was, however, a very bright spot for the U.S.—namely, the financial service sector. The 1980s saw the beginning of tremendous growth in this industry: between 1980 and 2006, the industry’s share of total U.S. GDP went from 4.9 percent to 8.3 percent (Greenwood and Scharfstein 2013). Moreover, wages in the financial sector shot up: in 1980, the typical financial services employee earned approximately the same wage as workers in other industries; by 2006, however, financial services employees were earning an average of 70 percent more (cited in Greenwood and Scharfstein 2013, p. 4). The increasingly impactful financial services sector in the United States would come to play a very important role in the global financial crisis in the latter part of the 2000s.

Figure 5.5. Demons in the Parking Lot

 

The following passage is an excerpt from the book Buy American: The Untold Story of Economic Nationalism, by Dana Frank (2000). It gives a palpable sense of the economic difficulties the United States began to experience in the 1980s:

 

The scene was like something out of a Batman cartoon. The first man lifted the sledgehammer, planted his feet wide apart, sighted over his left shoulder, and THWACK! sank it deep into the car with a grunt of pleasure. The crowd roared. The next man paid his dollar, swaggered up to the car, hefted the sledgehammer back and forth ... and suddenly WHAM! swung it into the car’s front corner....

 

The object of their aggression was a Toyota, its assailants members of the United Auto Workers at a union picnic in the 1980s. If the ILGWU’s [International Ladies’ Garment Workers’ Union] “look for the union label” song burned itself into the memories of millions of TV watchers in the 1970s, even more emblematic of Buy American campaigns by the 1980s was the image of an unemployed auto worker in Detroit smashing a Japanese car. (p. 160)

 

            While the difficulties faced by the United States in the 1980s were multifaceted, the switch to a floating exchange-rate regime certainly played a key role. Unlike in the interwar period, however, the United States did not follow the path of Smoot-Hawley; instead, the U.S. government opted for negotiations with the other G5 nations: the United Kingdom, France, West Germany, and Japan (see figure 5.7, “From the G5 to the G20,” for a discussion of why only these countries initially met, and for a discussion of the evolution of the G5 into a larger “club”). In September 1985, ministers of finance and central-bank governors from the five countries met at the Plaza Hotel in New York City. The representatives all wanted to stifle the threat of protectionism, and therefore were able to reach an agreement that resulted in the gradual devaluation of the dollar against the yen, deutsche mark, and franc (see figure 5.6): over the following two years, the dollar depreciated by about 40 percent in relatively orderly fashion. The agreement is known as the Plaza Accord. This did not turn the U.S. current account deficit into a surplus, but it did almost immediately lead to a rise in U.S. exports: in 1986, exports were up 8.2 percent; this was followed by increases of 13.8 Text Box: Figure 5.6. Foreign Exchange Rates, Major Currencies Before and After the Plaza Accord

 

Source: http://www.research.stlouisfed.org/. Permission is granted to copy, distribute, and/or modify this document under the terms of the GNU Free Documentation License, version 1.2, or any later version.
percent, 18.3 percent, and 11.0 percent in 1987, 1988, and 1989 respectively (OECD 1991, p. 19). In 1987, the G5 (plus Canada and Italy) met again at the Louvre in Paris, France. This time, though, they wanted to put the brakes on the depreciation of the dollar. In addition, the seven industrial powers also wanted to improve coordination across a range of fiscal and macroeconomic policy areas beyond exchange rates, a task that was largely achieved, at least on paper. France agreed to reduce its budget deficits by one percent of GDP; Japan agreed to stimulus expenditures and tax cuts; Germany agreed to reduce public spending, cut taxes, and keep interest rates low; and the U.S. agreed to reduce its fiscal deficit and cut spending. There was also an attempt to rejigger the exchange-rate regime by introducing a target zone; this was a variant of the fixed-but-adjustable exchange-rate system (instead of the +/- one percent band of the par value system, this one used a +/- 10 percent band). Compared to the Plaza Accord, the Louvre Accord, as it is known, was far more ambitious. Unfortunately, but not surprisingly, the Louvre Accord did not work. A degree of stability was achieved for the first eight months, but after that the agreement was largely abandoned. Domestic economic and political concerns in the major countries, especially Germany, Japan, and the United States, trumped concerns for international cooperation (this helps underscore, once again, the value of two-level-game insights on the interconnectedness between international and domestic politics). Indeed, disagreements among the countries—particularly those between Germany and the U.S.—were, at times, quite intense. (At the same time, the value of the dollar did, for the most part, remain stable for a long period of time following the negotiations in Paris.)

 

 

 

 

 

Figure 5.7. From the G5 to the G20

 

Why did the Plaza Accord originally include only five countries? While it is difficult to answer this question definitively, it is worth re-emphasizing a point made in chapter 2: in the early postwar years, the international system was thoroughly dominated by exclusive “clubs” of like-minded, economically powerful countries. The first of these clubs, according to Gordon Smith (2011), emerged in the aftermath of the collapse of the par value system. The first meeting occurred on March 25, 1973, when finance ministers from Britain, France, Germany, and the U.S. met and formed the Library Group, which was named after the venue of the initial meeting in the White House Library. A few months later, Japan was invited to join the club, and the Group of Five, or G5, was born. The G5 continued to meet on a periodic basis for the next decade or so, although Italy (in 1975) and Canada (in 1976) were invited to join the group, too, which led to the creation of the G7. (Canada, it should be noted, was added primarily to provide more geographic balance to the group—since Italy was added, the United States argued that Canada should be included as well.) The delayed inclusion of Italy and Canada, however, set them outside the core membership of the club. This is the most likely reason only five countries were included in the Plaza Accord meetings. Indeed, in the initial follow-up meeting at the Louvre, Italy and Canada were excluded; it was only after an agreement was reached among the five original members that the two latecomers were invited. This did not sit well with the Italian finance minister, who ended up leaving the meeting midway (Oba 2007). To avoid a recurrence of hurt feelings, subsequent meetings all included the expanded list of seven member countries. After the breakup of the Soviet Union, Russia was invited to participate, first on a selective basis and then as a full-fledged member in 1997. Thus, the G7 became the G8. Russia’s inclusion, it is useful to add, was something of a risk. The genesis of the club stipulated that all members, despite differences in policy, shared the same basic characteristics and values—i.e., they had to be market-based liberal democracies. The hope was that, by allowing Russia to join, it would eventually embrace the same set of values (Smith 2011).

 

 

As Smith (2011) notes, the exclusivity and elitist character of the club, however, became increasingly hard to justify over time, both for practical and ideological reasons. By the late 1990s, then, an effort—led by Canadian finance minister Paul Martin and U.S. Treasury secretary Lawrence Summers—was begun to expand the membership. One proposal called for increasing membership to 33, while another proposed expanding to 22 members. In the end, a slightly smaller number was selected; thus, in December 1999, the G20 was established. This did not mean, however, that the G8 would be disbanded; instead, the G20 was made into a separate (but not necessarily equal) club, while the G8 continued to hold its own, exclusive meetings. This, too, became less and less tenable, especially with the economic rise of China. Understanding that the unrepresentative nature of the G8 was becoming a burden, the chair of the 2005 G8 Summit, UK prime minister Tony Blair, invited five additional countries to the meeting: China, Brazil, India, Mexico, and South Africa. The 2007 summit regularized the relationship between the “G8+5,” but the new format was perceived as insulting. Paul Martin (the Canadian prime minister) nicely summed up the issue:

 

the image of Hu Jintao, the president of China, and Manmohan Singh, the prime minister of India —leaders of the two most populous countries on earth, quite possibly destined to be the largest economies on earth within our lifetimes—waiting outside while we held our G8 meetings, coming in for lunch, and then being ushered from the room so that we could resume our discussions among ourselves, is one that stayed with me…. Either the world will reform its institutions, including the G8, to embrace these new economic giants, or they will go ahead and establish their own institutions. (cited in Smith 2011)

 

            In sum, the evolution of the G5/G7 is a microcosm of changes in the global political economy. It reflects the exercise of power in the global economy, and how changes in the distribution of power may compel, but do not always immediately lead to, institutional changes.

Image. A group photo of G7 finance ministers and central-bank governors during meetings at the U.S. Treasury Department on April 11, 2008.

 

Source: IMF Staff Photograph/Stephen Jaffe. The photo is in the public domain.

The Rise of Neoliberalism: A New Social Reality

            Underlying the concrete problems with the foreign exchange regime was a significant shift in thinking about the nature of the relationship between the state and market. In chapter 1, the debate between Keynesians and followers of Hayek was discussed; up until the early 1970s, Keynesians were the clear winners (after Nixon took the U.S. off the gold standard, for example, he is reported to have said, “I am a Keynesian now” [New York Times, January 4, 1971]). But just a few years later, things had started to change, although a change had been in the works for some time: put most simply, neoliberalism was on the rise. The intellectual guru most closely associated with the rise of neoliberalism, however, was not Hayek (see figure 5.8, “Masters of the Universe and the Birth of Neoliberalism”) but instead was Milton Friedman, who was then teaching at the University of Chicago (the term the Chicago boys is sometimes used to refer to the neoliberal movement led by Friedman and a group of like-minded economists). No doubt, a big reason for the changes that were beginning to take place stemmed from a worsening economic environment. As David Harvey explains it, by the end of the 1960s, “[s]igns of a serious crisis of capital accumulation were everywhere apparent. Unemployment and inflation were both surging everywhere, ushering in a global phase of ‘stagflation’ that lasted throughout much of the 1970s. Fiscal crises of various states (Britain, for example, had to be bailed out by the IMF in 1975–76) resulted as tax revenues plunged and social expenditures soared. Keynesian policies were no longer working” (12). Actually, it is more accurate to say that the signs of serious crises were in many places, but not everywhere. Among the major economies, Japan and Germany were doing fine, while the U.S. and the United Kingdom were suffering from serious financial strains. Thus, it is no surprise that the shift in thinking was most evident in the latter two countries, with President Reagan and Prime Minister Margaret Thatcher leading the charge.  

 

 

Figure 5.8. Masters of the Universe and the Birth of Neoliberalism

 

Daniel Stedman Jones (2012), in his book Masters of the Universe, points out that neoliberalism is an essentially transatlantic—as opposed to primarily American—phenomenon. Indeed, the origins of neoliberalism (which he defines as “the free market ideology based on individual liberty and limited government that connected human freedom to the actions of the rational, self-interested actor in the competitive marketplace” [p. 2]) can be traced to the 1920s, and the Austrian and German Freiburg (also known as ordoliberal) schools. The most prominent neoliberals in this early period were all Europeans: Friedrich Hayek, Alexander Rüstow, William Röpke, Jacques Rueff, Michael Polanyi, and, more prominently perhaps, Ludwig von Mises (p. 6). In this early period, however, the appeal of neoliberalism remained limited. After the end of World War II (which marks the beginning of the second phase of neoliberalism), the mantle of neoliberalism was taken up again, and the most stalwart advocate was Friedrich Hayek, part of the Austrian school, who founded the Mont Pelerin Society (MPS) in 1947. The society (which first met at Mont Pelerin, near Montreux, Switzerland) was devoted to the “exchange of ideas between like-minded scholars in the hope of strengthening the principles and practice of a free society and to study the workings, virtues, and defects of market-oriented economic systems” (https://www.montpelerin.org/).

 

The MPS became a hub for neoliberal thought, attracting, as it did, the leading neoliberal thinkers of the day, including the American economist Milton Friedman. Friedman, in fact, became an important bridge between the first and second phases of neoliberalism and “between the concerns of the predominantly European founding figures ... and a subsequent generation of thinkers, mainly though by no means all American, located especially in Chicago and Virginia” (pp. 6–7).

 

From the 1950s to the early 1980s (until the rise of Thatcher in the UK and Reagan in the U.S.), the influence of the MPS and of neoliberalism slowly, but inexorably, grew. It was toward the last part of this period that neoliberalism developed into a recognizable group of ideas and a veritable movement (p. 7). The third and current phase, according to Jones, includes neoliberalism’s ascendance during the Reagan and Thatcher period, during which time neoliberal ideas moved from the periphery to the center of the global political economy, as they became embedded in domestic national politics throughout the world, and into many global institutions, including the IMF, the World Bank, and the WTO (p. 8).

 

            What happened, however, was more than a simple shift in thinking; instead, it was, at least to some analysts, a revolutionary change in understanding the common-sense relationship between the state and market, or among the state, market, and society. Following the ideas of Antonio Gramsci, Harvey (2005) tells us that common sense—defined as “the sense held in common”—is a social construction; it is the product of “cultural socialization often rooted deep in regional or national traditions” (p. 39). Common sense helps to define our social realities. Prior to the 1970s, the common sense was embedded liberalism, which was premised on an understanding that market processes and entrepreneurial and corporate activities should be grounded or embedded in a larger social context. In this view, the regulation and management of the market, and the active protection of society through, for example, social welfare programs, was taken for granted. That common sense has been replaced by neoliberalism, which tells us most generally that markets need to be liberated from state intervention and that society needs to be made more open to the logic of the market, rather than to be protected from markets. How this transition in common sense was achieved is a complicated story and too much to cover in this chapter.[22] For our purposes, it is enough to say that the ideological transition to neoliberalism was accompanied by a series of concrete policy changes: privatization, marketization (of areas not yet governed by market processes—see figure 5.9, “Marketization of the U.S. Military”), deregulation, tax cuts, the reduction of social welfare programs, and so on. In the global financial system, deregulation has been the most salient and far-reaching change: it has “freed” capital and led to significant financial innovations, which have not only produced denser and more sophisticated global interconnections than ever before, but also new kinds of financial markets based on securitization, derivatives, and all manner of futures trading: neoliberalization has meant, in short, the financialization of everything (Harvey 2005, p. 33).

 

 

 

Figure 5.9. Marketization of the U.S. Military: An Example

 

Marketization, most simply, is the process of turning anything into a product that can be sold in markets. Consider the military. For a long time, the military was considered to be outside the domain of markets (even if the separation was never entirely complete). In the United States, however, military functions have been increasingly marketized since the early 2000s. When the U.S. defeated the Iraqi army in 2003, for example, at least one out of every ten people deployed in the theater of conflict—doing work that traditionally had been done by soldiers—was an employee of a private security company. In the occupation that followed, this number increased: by May 2004, there were more than 20,000 private security personnel in Iraq employed by some 25 different private security companies (Avant 2006), one of the best-known of which was Blackwater USA. Indeed, Blackwater employees even provided security for top U.S. officials in Iraq during the occupation: L. Paul Bremer, who served as the administrator of the Coalition Provisional Authority of Iraq following the 2003 invasion (until June 2004, when sovereignty was transferred from the U.S. government back to the Iraqi government). Blackwater was paid $21 million to guard Bremer over an 11-month period, but the company also had a five-year $320 million contract with the U.S. State Department to provide security for U.S. officials in conflict zones around the world. In 2006, Blackwater won the contract to protect the U.S. embassy in Iraq, which is the largest American embassy in the world (http://www.corpwatch.org/section. php?id=210).

 

Image: Iraqi Prime Minister Ayad Allawi (left), Ambassador L. Paul Bremer, and President Sheikh Ghazi Ajil al-Yawar make their farewells after a ceremony celebrating the transfer of full governmental authority to the Iraqi Interim Government, June 28, 2004, in Baghdad, Iraq. Although not indicated, it is likely the Blackwater personnel are trailing behind the three leaders.

 

Source: U.S. Air Force photo by Staff Sgt. Ashley Brokop. The image is the work of a U.S. government employee, taken as part of that person’s official duties. As a work of the U.S. federal government, the image is in the public domain.

 

Neoliberalism generally, and the deregulation of the financial sector (domestically and globally) more specifically, produced, in the views of many observers, breathtaking results. By the mid-1990s, all the economic problems of the 1970s were nothing but distant memories, especially in the United States and the United Kingdom. But one can argue that the positive effects were much more widely felt as well throughout much of the developing world, including, most prominently, in China. In the United States, almost all the credit was given to Reagan’s neoliberal vision. Consider how Peter Ferrara, writing for Forbes magazine, described the success of neoliberalism:

[Reagan’s] economic policies amounted to the most successful economic experiment in world history. The Reagan recovery started in official records in November 1982, and lasted 92 months without a recession until July 1990, when the tax increases of the 1990 budget deal killed it. This set a new record for the longest peacetime expansion ever, the previous high in peacetime being 58 months. During this seven-year recovery, the economy grew by almost one-third, the equivalent of adding the entire economy of West Germany, the third largest in the world at the time, to the U.S. economy. In 1984 alone real economic growth boomed by 6.8%, the highest in 50 years. Nearly 20 million new jobs were created during the recovery, increasing U.S. civilian employment by almost 20%. Unemployment fell to 5.3% by 1989 (2011, n.p.).

 

Crises and the Global Financial System

            Not all was well, however. Particularly on the credit side of the global financial system, serious problems were beginning to emerge. The first hint of trouble began in the 1970s, but the real difficulties began in the early 1980s, with the Mexican debt crisis of 1982. This was shortly followed by a string of debt crises throughout Latin America, leading to what has been labeled the “Lost Decade” for the entire region. Moving into the 1990s, the bad news kept coming: Japan suffered from an asset price bubble, which collapsed in 1990, while Mexico experienced another crisis, this one centered on its currency, which was aptly called the Mexican peso crisis (1994–95). These were followed by the Asian financial crisis of 1997–98 (primarily affecting Thailand, Indonesia, Malaysia, and South Korea), the Russian financial crisis of 1998, and a string of crises in Brazil, Turkey, and Argentina at the end of the decade and into the first few years of the new millennium. Towards the end of the first decade of the 2000s, of course, the world was struck by a global financial crisis, but this time the principle victims were in the developed world, including the United States and much of Western Europe. Crises have become so common, in fact, that it might be said that the global financial system has been in a perpetual state of crisis since the early 1980s. All of this tends to support the analysis by Susan Strange, discussed earlier: the global financial system appears to have become extremely unstable and volatile. Whether it is a giant casino, or whether another metaphor provides a better description, it behooves us to take a closer look.

 

The Debt Crises of the 1980s

            Despite some success stories in the developing world, in the 1980s, a new type of problem emerged: an international debt crisis (more accurately, a series of debt crises striking a range of developing countries). The oil shocks of the 1970s, discussed earlier, helped to lay the groundwork for these crises by releasing vast sums of money into international financial markets—money that had to go somewhere in order for banks to earn profits. Much of this money found its way into the developing world, some of which was used to finance oil imports (although, contrary to conventional wisdom, only parts of the developing world were severely affected by this), and some of which was used to invest in industrialization. This sounds reasonable, but it was—at least according to some scholars—a recipe for disaster. First, a brief discussion of some details is in order. Between 1972 and 1981, external debt in the developing world increased six-fold to around $500 billion. The most indebted countries—in terms of total external debt as a proportion of GDP—were primarily in Africa and Latin America. In 1982, for example, Brazil and Mexico had, respectively, external debt obligations of $93 billion and $86 billion with debt-to-export ratios of 447.3 percent and 311.5 percent. (The debt-to-export ratio is the total amount of debt in comparison to the total annual exports; it provides a rough measure of the capacity of a country to repay its external debt obligations; anything over a 100 percent is high.) The extremely high debt-to-export ratios, as well as high debt-service ratios (see table 5.4) underscore an important and fairly obvious point: very much like in the housing bubble in the United States—also referred to as the subprime mortgage crisis—a lot of the loans made to developing countries in the 1970s were subprime (subprime loans are those given to borrowers with a tarnished or limited credit history).

 

 

Table 5.4. External Debt and Debt Ratios for Selected Countries in Latin America, 1982 (U.S.$ billions)

 

 

Total Debt

Debt-Export Ratio (%)

Debt Service Ratio (%)

Argentina

$43.63

447.3

50.0

Brazil

$92.99

396.1

81.3

Chile

$17.31

335.9

71.3

Colombia

$10.30

204.3

29.5

Mexico

$86.02

311.5

56.8

Peru

$10.71

255.9

48.7

Venezuela

$32.15

159.8

29.5

 

Source: World Bank, World Debt Tables, 1992–93, Vol. 2: Country Tables (Washington, D.C., IRBD, 1992)

 

Text Box: Figure 5.10. Monthly Treasury Bill Rate (3-Month), 1970–1994

 

Source: FDIC (1997), p. 205. Original source is Haver Analytics.
All this bad lending eventually led to situations in which a number of countries could not pay back their debts. The first big case was Mexico in 1982, but there were clear harbingers well before that: between 1976 and 1980, Zaire, Argentina, Peru, Sierra Leone, Sudan, and Togo all experienced serious difficulties. However, Mexico’s announcement—on August 12th, 1982, Mexico’s minister of finance told the U.S. government and the IMF that it would no longer be able to service its debt (i.e., pay the interest and principal on the scheduled due date)—opened the floodgates. By October 1983, another 26 countries, owing a total of $239 billion, were forced to reschedule their debts (or were in the process of doing so), with many others to follow. Of the first 27 countries, 16 were located in Latin America, and the four largest—Mexico, Brazil, Venezuela, and Argentina—owed approximately $176 billion to commercial banks, including $37 billion to just eight U.S. banks. For these eight banks, it is useful to note, the $37 billion in loans accounted for about 147 percent of the capital and reserves at the time, meaning that they all faced the prospect of insolvency if the loans defaulted (all figures cited in FDIC 1997, v. 1, p. 191). The fact that the most serious problems began to appear in the early 1980s is tied to the manner in which the loans were structured: about two-thirds of outstanding developing-country debt was tied to a floating LIBOR, or the London Interbank Offering Rate (FDIC 1997, v. 1, p. 195). Thus, when interest rates shot up to record levels in the early 1980s (see figure 5.10)—a product of U.S. Federal Reserve efforts to curb oil-based inflation—debt-service costs to developing countries quickly became unmanageable.

So why were so many “bad” loans made? Needless to say, there are various competing answers, which largely parallel the arguments commonly heard about the subprime mortgage crisis in the United States. The main arguments can be neatly classified in the following manner: (1) irresponsible and greedy lenders; (2) irresponsible and corrupt borrowers; and (3) systemic problems (Cohn 2011, pp. 344–47). The first two arguments lay the blame on individual actors and strongly suggest that, without “bad actors,” there would have been no problem (which begs the question of whether deregulation of the global financial system was the main issue). These actor-centered arguments, however, are difficult to accept, at least as complete explanations, given the pervasiveness of the international debt crisis in the 1980s, and the fact that debt crises have become a lasting and recurring problem in the global financial system for the last three decades (more on this shortly). Thus, there are almost certainly systemic factors at play. In addition, we have to consider the issue of (structural) power and interests, and how political considerations more generally play a central role in the global political economy.

From a Marxist or neo-Marxist perspective, the explanation is fairly clear: the system is rigged to ensure that the developing world remains in a subservient and dependent position, and to maximize exploitative processes for capital accumulation among the core economies. The foreign debt regime, in particular, is designed so that the developing countries never get out of debt, and never develop significant economic autonomy. To see this, consider the fact that the commercial bankers knew well in advance that continued lending to the developing countries was unsustainable. In 1969, for example, the journal of the American Bankers Association warned, “Many poor nations have already incurred debts past the possibility of repayment.... International loans, even if made on ‘businesslike’ terms, have a way of getting lost unless they are repaid out of the proceeds of additional loans” (emphasis added; cited in Payer 1991, p. 69). The last part of the statement is quite telling—and prophetic—because this is precisely what happened. In Latin America, the region began exporting capital to the core economies in 1982, despite receiving a continuing inflow of capital from Western banks and other sources. This included loans from the eight largest U.S. banks, which were threatened with insolvency by the Mexican and related debt crises in the early 1980s: from 1983 to 1989, their lending to developing countries (not just Latin America) averaged $52.9 billion a year for a total of $370.5 billion (FDIC 1997, v. 1, p. 197; author’s calculations). Latin American countries received $49.6 billion in loans in 1983, yet they ended up transferring $66.3 billion (in interest and amortization) back to the West that same year. Payer (1991) estimates that, for Africa, the net transfer became negative in 1984, and for East Asia, 1986 (p. 14). If this were just a temporary or short-term phenomenon (or limited to a small set of countries), it would not necessarily mean much. But this has not been the case: it continued through the 1990s and into the 2000s. In 2007, a peak net transfer of financial resources from the developing world to the developed was reached: approximately $881 billion (UN DESA 2011, p. 69), although this figure did decline to “only” $557 billion in 2010.

 

 

Table 5.5. Net Transfer of Financial Resources to Developing Economies, 1998–2010 (billions U.S.$)

 

 

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

Developing Economies (All)

-41

-128

-194

-164

-210

-302

-379

-597

-807

-881

-876

-545

-557

Africa

2.9

1.6

-31.7

-16.4

-4.2

-16.1

-34.5

-79.4

-108

-101

-99

2.9

-35.3

East and South Asia

-130

-140

-123

-121

-149

-176

-183

-265

-386

-530

-481

-427

-533

Western Asia

34.5

2.7

-35.3

-29.7

-23.2

-46.4

-76.3

-144

-175

-144

-222

-48.4

-112

Latin America and Caribbean

41.5

7.7

-4.2

2.5

-33.6

-64.3

-85.4

-111

-138

-106

-73.5

-72.1

-56.1

 

* Figures are rounded Figures for 2010 are estimates

Source: UN DESA (2011, p. 71), based on IMF, World Economic Database, October 2010 and IMF, Balance-of-Payments Statistics

 

Table 5.6. External Debt Stock of Developing Countries and Key Ratios, 2005–2010 (billions U.S.$)

 

 

2005

2006

2007

2008

2009

2010

Total External Debt Outstanding

2,514.1

2,675.3

3,220.5

3,499.2

3,639.6

4,076.3

Ratios

 

 

 

 

 

 

    External Debt to GNI (%)

26.6

23.9

23.2

21.0

22.4

21.0

    External Debt to Exports (%)

75.9

66.1

65.6

9.3

77.0

68.7

 

Source: World Bank Debtor Reporting System and IMF; cited in World Bank (2012), p. 2.

 

 

These figures, combined with the statistics on total outstanding debt (see table 5.6), are astounding. They tell us that the developing world—barring truly radical changes in the global economy—will never be able to pay off or even pay down its outstanding debt. They also tell us, and even the harshest critics of Marxism might have to agree, that the developing world will likely forever remain unhealthily dependent on capital and markets in the developed world. Even more astounding, perhaps, is that the proposed solutions have only made the problem worse for the vast majority of developing economies (the figures in table 5.6 are clear testament to this). The most damaging “solution,” again from a neo-Marxist perspective, is obvious: liberalization, and more specifically the imposition of the neoliberal model on the developing world. This was manifested in the so-called Washington Consensus, which describes a set of neoliberal economic policy prescriptions that became the standard reform package used by the IMF and the World Bank in dealing with the debt crisis in the developing world. It was first applied to Latin American countries, and included trade liberalization, currency devaluation, liberalized capital markets, privatization, deregulation, fiscal austerity (i.e., a reduction in government spending), and tax reform. These policies also became the basis for IMF conditionality. The logic was simple: by liberalizing their economies and practicing greater fiscal discipline, developing countries would reduce their current account deficit, grow their economies faster, and be in a much better position to service and eventually pay off their external debt. This has clearly not been the case. Thus, while both critics and advocates of the Washington Consensus (or of conditionality) have plenty of examples they can trot out to support their respective positions, there is almost no doubt that the debt problem—and other indications of financial instability—are continuing to grow in the developing world.

Politically, however, the debt and currency crises in the developing world have proven to be quite successful: they have, according to critics, provided the IMF (and powerful actors within the U.S.) a useful pretext for expanding neoliberalism to most of the world. Thanks in part to conditionality, the promise of neoliberal globalization has come closer and closer to fruition, as virtually the entire world is now integrated into global markets for capital and trade. Of course, conditionality is only one mechanism used to advance neoliberal globalization—there are also emulation effects, as countries, such as China, see the voluntary embrace of some neoliberal policies as beneficial. Whether this is for good or bad, of course, is subject to intense and probably never-ending debate. Liberal economists tell us that liberalization—despite the inevitable ups and downs—maximizes the efficient use of resources, and therefore maximizes prosperity for the greatest number of people. Critics of the liberal view, neo-mercantilists and Marxists, tell us that liberalization is ultimately about the unequal distribution of power in the world, whether exercised by states-as-actors or by class actors, and is designed to maximize exploitation. Who is right? While the discussion here has admittedly been slanted toward the latter view, one thing is certain: power, politics, and economics always intersect in the global political economy.

Text Box: Figure 5.11. Current Account Balance: the U.S., Japan, Germany, and China (percentage of GDP)
 
Source: UNCTADstat, http://unctad.org/en/Pages/Statistics.aspx

Trade Imbalances, the U.S. Housing Bubble, and the Global Financial Crisis

            For a long time, at least from the perspective of many people who live in core economies, the debt and currency crises in the developing world could be looked at with a fair degree of detachment. The developing world, after all, was different. Developing economies still had a lot to learn, and still had much to do to fully develop their markets. Their economic difficulties, in short, were symptomatic of immature market systems. To be sure, there have been plenty of significant financial crises in core economies in the postwar period—Black Monday in the UK (1987), the savings-and-loan crisis in the U.S. (1989–91), Black Wednesday in the eurozone (1992–93), the Long-Term Capital Management bailout (1998), and the dot-com collapse (2001), among others—but these have, to some extent, been viewed as anomalies. That is, they have been viewed as relatively isolated one-off events, different in character than the debt and currency crises afflicting the developing world in the 1980s and 1990s. To the extent that core countries suffered from long-term, debilitating problems—such as Japan experienced throughout the 1990s and into the 2000s—it was simply because, argued advocates of liberal economic theory, those countries had not embraced market principles strongly or closely enough. The global recession of 2008–2012 (including the sovereign wealth or eurozone crisis), however, has helped to dispel the notion that market liberalization offers a panacea. Indeed, the global recession has caused many analysts and observers (although certainly not all) to question the “common sense” of neoliberalism.

            In retrospect, there were plenty of warning signs, not the least of which was the string of smaller, but still significant financial crises sweeping through the entire world (in developing and developed economies alike) for several decades. But there were more specific indicators, too. In the United States, the current account—which showed a surplus in 1991, but then quickly went back to a deficit—steadily worsened over the rest of the decade and into the mid-2000s, reaching a record high of 6 percent of GDP in 2006. For the U.S., large (even massive) current account deficits do not always lead to immediate problems. The basic reason, discussed several times already, is clear: the attractiveness of U.S.-issued securities (i.e., bonds, notes, and T-bills) means that the United States has easily been able to finance its current account deficits through foreign savings. This is a luxury few other countries have. There is, however, an important flip side to the large and persistent U.S. current account deficit: equally large current account surpluses in countries that are major trading partners of the U.S. These surpluses, in turn, typically lead to a significant amount of savings, which is partly indicated in the growth of foreign exchange reserves. This was quite evident in the case of China, which ran a current account surplus of $353.2 billion in 2007, and increased its total reserves to $2.1 trillion (as discussed below, there were other reasons why China’s foreign exchange reserves suddenly shot up after 2005). That same year, Japan had foreign exchange reserves of almost $1 trillion, while Germany’s was just below $900 billion. In other words, just as with the OPEC oil crisis (or oil-pricing boom, depending on your perspective), there was a lot of excess or surplus cash lying around in certain parts of the world. Much of this was, to repeat, used to finance the U.S. current account deficit. On the surface, this was a win-win situation: both the U.S. and its major trading partners clearly benefited from what appeared to be a completely reciprocal financial relationship. Still, the very large and persistent deficits in the U.S. were problematic, not just for the United States, but also for the global financial system as a whole.

            Within the United States, the deficits of the 2000s led to renewed calls for protectionism. Unlike in the 1980s, though, the primary target of groups in the United States was China, which was most prominently labeled a currency manipulator, and was accused of recklessly violating the norms of international trade. Tensions between the two countries, as McKinnon (2012) describes it, increased significantly during the decade, with the U.S. senator Charles Schumer at one point threatening to impose punitive tariffs of 27.5 percent on all Chinese goods through a cosponsored bill (p. 38)—although this particular measure was later determined to be illegal under WTO standards (and, ironically, because China joined the WTO in 2001, the U.S. could not unilaterally impose protectionist measures against China). Nonetheless, the Chinese government finally relented in July 2005, and began to allow a slow appreciation of renminbi (RMB) against the dollar at about 6 percent a year until July 2008 (p. 36). This policy shift led to a sudden influx of foreign capital into China as investors bet on a stronger RMB in the future (p. 37)—this was another reason for China’s huge increase in foreign exchange reserves. The stronger RMB, however, did not result in a decline in China’s current account surplus; instead, it continued to climb, reaching $800 billion in 2010. It is apparent, then, that exchange-rate policy, by itself, does not necessarily correct for current account imbalances within the context of a globalized financial system (MacKinnon and Schnabl 2012). There is, in short, no easy fix to the trade and current account imbalances that have plagued the U.S. since the mid-1990s.

            What does any of this have to do with the global financial crisis? Oatley (2012) provides a simple explanation: “The connection between imbalances and the financial crisis lay in the flow of cheap and plentiful credit from surplus countries to the United States at an unprecedented rate. The ability to borrow large volumes at low interest rates [after the dot-com bubble, the U.S. Federal Reserve cut short-term interest rates from about 6.5 percent to 1 percent] created credit conditions that typically generate asset bubbles” (p. 237). In the U.S., the bubble emerged in the real estate market, and first began to appear in the early 2000s. Table 5.7 provides some basic statistics, but the numbers are very clear: between 2001 and 2006, new mortgage originations totaled $18 trillion, an average of $3 trillion a year. In 1990, by contrast, new mortgage originations were only $459 billion, and in 2000, $1.14 trillion. The doubling, trebling, and (in 2003) quadrupling of mortgage lending over the space of a few years suggests that there was a vast and seemingly unlimited reservoir of “cheap” money available. But for mortgage lenders and others with access to those funds, there was a problem: the traditional customer base for home mortgages was much too small to absorb all that cheap money. Most simply, this is what led to a lowering of underwriting standards (as well as to a significant increase in refinancing). That is, to maximize profits, lenders had to dramatically expand their pool of customers; they did this by lending to homebuyers who had generally not had easy access—or any access at all—to the mortgage market. The lowering of underwriting standards is evidenced in the growth of subprime and Alt-A, or limited documentation, loans (see table 5.7 for a definition of Alt-A loans): from 2001 to 2003 nonprime (that is, subprime and Alt-A) loans constituted less than 16 percent of all mortgage originations, but in 2004, that figure more than doubled to 37 percent; in 2006, virtually half of all new loans were nonprime. In addition to lowering underwriting standards, the average difference in mortgage interest rates between nonprime and prime mortgages declined from 2.8 percent in 2001 to just 1.3 percent in 2003—this was almost completely contrary to past norms, since the decline in the gap between nonprime and prime loans coincided with increased risks in terms of the creditworthiness of borrowers (Bianco 2008, pp. 6–7).

 

 

Table 5.7. U.S. Mortgage Originations, 2001–2006 and Selected Years (in billions)

 

 

Mortgage Originations (Total)

Subprime

Alt-A

HELOC

% Nonprime

1990

459

N/A

N/A

N/A

N/A

2000

1,139

N/A

N/A

N/A

N/A

2001

2,215

120

60

130

14

2002

2,885

185

67

170

15

2003

3,945

310

85

220

16

2004

2,920

530

185

355

37

2005

3,120

625

380

365

44

2006

2,980

600

400

430

48

 

Notes: An Alt-A loan is a type of nonprime loan; it falls between the prime and subprime loan classifications. Borrowers typically have clean credit histories, but other risk factors are present, including (1) high loan-to-income ratio; (2) high loan-to-value ratio (e.g., these loans typically have minimal down payments), and (3) inadequate documentation of the borrowers income. HELOC stands for home equity line of credit.

 

Sources: Figures for 1990 and 2000 are from U.S. Census, “Table 1194. Mortgage Originations and Delinquency and Foreclosure Rates,” available at http://www.census.gov/compendia/statab/2012/tables/12s1194.pdf; figures for 2001–2006 are from Inside Mortgage Finance, cited in Acharya, Richardson, Neeuwerburgh, and White (2011), p. 36.

 

The lowering of underwriting standards was only part of the problem. After 2003, the market began to shift from “financing mortgages with regulated securitization to using unregulated securitization” (Levitin and Wachter 2012, p. 1182). Mortgage securitization—combining mortgages into one large pool and then marketing different tiers of this pool as separate securities backed by the cash flow from the original loans—has long been a common practice. Until 2003, though, the vast majority of mortgage securitization was done through the government-sponsored entities known as Fannie Mae, Freddie Mac, and Ginnie Mae.[23] This began to change as banks and other financial institutions invented new and “exotic” types of mortgage-backed securities; this, in turn, led to the creation of credit default swaps (or credit derivative contracts), which were designed to transfer the credit exposure of fixed income products (e.g., mortgage-backed securities) between parties. Ostensibly, credit default swaps reduced or insured risk—which is why they were sold by insurance/financial companies, such as AIG (American International Group)—by providing credit protection to the buyer. But this protection was only meaningful as long as the real estate bubble could be maintained. All bubbles, however, eventually burst, and when the housing bubble did, the consequences were not only immense for the United States, but also for the entire global financial system.

The most salient effects centered on the dramatic decline in housing prices in the U.S., which by mid-2009 had fallen 33 percent from the peak. But the decline in housing prices was only the tip of the iceberg. Not surprisingly, home foreclosures shot up, unemployment increased dramatically, and U.S. stock markets (e.g., the Dow Jones, NASDAQ, and the S&P 500) plunged. The decline in U.S. stock markets then led to significant declines in stock markets around the world. More importantly, because firms and investors in many other countries, especially in Europe, Japan, and China, were heavily invested in mortgage-backed securities and credit default swaps—international investors owned one-third of U.S. mortgages in some form (Cox, Faucette, and Lickstein 2010, p. 4)—the collapse of the U.S. housing bubble had a direct effect outside the United States. Even more, when the prices for mortgage-backed securities fell, investors could not, as Randall Dodd (2007), writing for the IMF’s Finance and Development magazine, explained it, “trade out of their losing positions” (n.p.). This meant that they had to sell off other assets—especially those with large unrealized gains, such as emerging market equities—to meet margin calls or to offset loses. This further exacerbated stock market declines worldwide; it also led to a temporary decline of currency values in emerging markets. All this turmoil in the global financial system—which included the bankruptcy of several large financial firms, including Lehman Brothers—resulted in a free fall of global credit markets, which undermined economic growth on a global scale and made massive bailouts a “necessity.” Among the companies that were bailed out was AIG, which played a key role in creating the crisis: the U.S. government committed $182 billion to AIG’s rescue, although the amount that AIG actually used was $68 billion. Altogether, $640 billion was disbursed through various rescue packages (see table 5.8); of that amount, bailed-out companies returned $367 billion to the U.S. Treasury (Kiel and Nguyen 2013).

 

Text Box: Table 5.8. Breakdown of Bailout Funds (Outflow and Inflow)

	Disbursed (in U.S.$ billions)
Banks and Other Financial Institutions	$245
Fannie Mae and Freddie Mac	$187
Auto Companies	$79.7
AIG	$67.8
Toxic Asset Purchases	$18.6
State Housing Programs	$2.68
Mortgage Modification Program	$2.35
Small Business Loan Aid	$0.368
FHA Refinance Program	$0.050
Total	$640.3
	
	Returned 
Funds
Refunds	$376
Dividends	$152
Interest	$1.81
Warrants	$9.42
Other Proceeds	$19.4
Total	$549.9

Source: Kiel and Nguyen (2013), ProPublica. Available at http://projects.propublica.org/bailout/main/summary.

The Political Economy of the Global Financial Crisis

            The full story of the U.S. housing bubble and its fallout is far too complex to cover here in any depth, so the aim in this section is to highlight—largely in outline form—the political-economic aspects of the crisis. (For readers interested in a more detailed treatment of the housing bubble and subsequent financial crisis see, for example, Schwartz [2009] and Wachter and Smith [2011].) The first is the most general: the crisis clearly involved a complex interaction between political and economic forces and between state and nonstate actors. Power, moreover, was clearly and significantly diffused throughout the global financial system. Thus, while the decisions of state actors mattered—e.g., the decision by the U.S. Federal Reserve to lower interest rates after the dot-com crisis was something that only the U.S. state had the power to do—so did decisions made by nonstate actors. The decisions of large banks and financial firms obviously were instrumental in the crisis; the privileged position they occupied in the global finance structure made it possible for them to have a fundamental impact on the dynamics of the system. It is important to recognize, however, that firms are made up of individuals, who also have the capacity to exercise power (especially in the knowledge structure), both as individuals and as part of larger organizations. The credit default swap, for instance, was the creation of a single person—Blythe Masters, who as an employee of J.P. Morgan, pitched the idea of selling credit risk to the European Bank of Reconstruction and Development in 1994 (Romm 2010).[24] While someone else may certainly have created a similar financial product sometime later, the fact remains that the credit default swap had to be created through individual action. Moreover, the initial timing and success of Masters’s innovation was critical: it allowed for the credit default swap to become an important part of the global financial system before the housing bubble started to form. Borrowers, too, played a necessary role: the creation of mortgage-backed securities and credit default swaps would have had little impact if millions of individual borrowers did not actively seek out new mortgages. Agency, in short, clearly mattered in the process leading to the global financial crisis.

Added to this mix is the issue of moral hazard, which can be most simply defined as a situation in which one party in a transaction can make a decision about how much risk to take, while someone else bears the cost if things go badly (Krugman 2009). The housing market in the period leading up to the collapse was rife with moral hazard. AIG’s use of credit default swaps (CDSs), in this regard, likely played a key role because, first, credit default swaps allowed the financial institutions making poor-quality loans to transfer risk to another party—i.e., those willing to buy the CDS. Second, the parties buying the CDS assumed that AIG was “too big to fail”; thus, they were able to transfer the risk of their investments to the American taxpayer (for further discussion see Dowd 2009). It is crucial, however, that agency (and the exercise of power) always be understood in context. This leads to the second point, which is that the U.S. housing bubble and subsequent global crisis was made possible by a specific type of financial regime, one that rested on a solidly political foundation and was the product of a profoundly political process. On this point, keep the earlier discussion of global neoliberalism in mind: to repeat, global neoliberalism did not arise automatically, but had to be painstakingly constructed by both state and nonstate actors. An important aspect of the neoliberal regime was deregulation, the freeing up of capital and nonstate financial actors from meaningful regulatory oversight and control. Thus, while the causes of the housing bubble and global financial crisis are manifold,[25] there is little doubt that the radical shift to unregulated securitization was a major factor. This shift, on one level, was simply a reflection of efforts by financial firms (important nonstate actors) to maintain their earning and profit levels. Significantly, another type of firm (or nonstate actor), the credit-rating agencies such as Standard & Poor’s, Moody’s, and Fitch Group, played a key role as well: without their say-so, which was represented by a triple-A (or similarly high) rating, a large percentage of the new types of securities used to finance subprime mortgages would likely not have been sold. The credit-rating agencies, in other words, had tremendous capacity to regulate risk (even though they do not possess formal regulatory powers), and their collective unwillingness to exercise this power responsibly almost certainly exacerbated the conditions that led to the bubble and its ultimate collapse. Of course, the credit-rating agencies were acting within the context of a regulatory environment that allowed them to profit directly from providing “generous” (that is, overly optimistic and even spurious) evaluations to the firms they were regulating. Thus, at another level, the housing bubble might not have happened (at least to the extent that it did) in a different domestic regulatory environment, nor would its effects have been as far-flung in a different global financial regime.

Granted, these are rather large generalizations, but it is easy enough to see that, for the last 30 to 40 years, deregulation and privatization have been virtual movements within the United States, and through much of the world’s financial markets. In the U.S. specifically, a key decision was the reinterpretation of the Glass-Steagall Act, put in place after the Great Depression. Glass-Steagall prevented institutions that were “engaged primarily” in banking activities from dealing in securities of any kind, and vice versa (Sherman 2009, p. 8). In 1986, the Federal Reserve qualified the original restriction and ruled that a bank could derive up to 5 percent of gross revenue in investment banking; a few months later, Alan Greenspan—an outspoken advocate of deregulation—was appointed chairman of the Federal Reserve. He used his three decades as chairman to render the Glass-Steagall Act effectively obsolete (Sherman 2009, p. 9). This was no accident: Greenspan held a strong belief in neoliberalism in general, and more specifically believed that the “inherent incentive structures” and self-regulating nature of free markets made the system “fireproof”; this belief, as Jones (2012) put it, “was based on the view that the self-interest of financial institutions would effectively substitute for the rigorous external regulation of financial markets because it would prevent banks from overexposure to high-risk strategies” (p. 339). Overweening faith in neoliberal principles, in this regard, played a key role in both deregulation and the housing bubble. Again, there is much more to the story of deregulation, but the gist is clear enough: deregulation was a purposeful, political process.

            The fallout from the crisis helps underscore the third point, which is that states still must play a significant role, even, or especially, in the context of the global neoliberal order. While not a few analysts argue that doing nothing would have been the best response on the part of national governments, that was a near-impossible choice as the fallout from the crisis began to reverberate throughout the world. States or national governments could not afford to ignore the structural power (and political influence) of giant financial firms, nor could they risk the potentially calamitous damage to the global financial system by failing to act. Thus, while a handful of firms (e.g., Bear Sterns, Lehman Brothers, Countrywide Financial, IndyMac, Washington Mutual, and Wachovia, among others) could be allowed to go bankrupt, or more typically, acquired by other firms, the private financial sector as a whole—both domestically and globally—had to be rescued. Significantly, only states had the capacity and interest to carry out this rescue operation. And the state-led rescue effort was massive. Table 5.8, “Breakdown of Bailout Funds,” gives a clear indication of the resources that were devoted by the U.S. government to stave off financial collapse: $640 billion of actual disbursed financial assistance, and much more on paper. AIG was accorded extraordinarily special treatment because its involvement in credit default swaps was intimately connected to the financial viability of a host of other large financial firms—Goldman Sachs, Morgan Stanley, Bank of America, Merrill Lynch, and dozens of European banks (Greider 2010). Tellingly, in early efforts to rescue AIG, as William Greider (2010) explains it, the U.S. government attempted to coordinate with the private sector (i.e., the government wanted the private sector to cover most of the costs of saving AIG), but the banks rebuffed these efforts, naturally preferring a bailout using primarily public funds—a nice expression of their dominant positions in the financial structure even in the midst of a full-blown crisis that the industry, most everyone agrees, was primarily responsible for creating in the first place. Neo-Marxist analysts, it should be noted, would not be surprised at all by this situation, and would perhaps offer a sardonic smile. After all, they tell us, states and their agencies have always represented dominant class interests. Indeed, it is hard not to give some credence to this point of view. On the other hand, it is equally easy to argue that state actors were not simply doing the bidding of “big capital,” but were instead acting in their own interests by trying to save the system they created. The key point, to reiterate, is nonetheless clear: states played a critical role in ensuring that the crisis did not spiral out of control.

            This leads to the fourth and final point. While, in the United States, the response to the collapse of the housing bubble and the ensuing financial crisis was a domestic affair, the global crisis that followed underscored the importance of international cooperation—even if limited and imperfect—and of the general framework for greater cooperation in the postwar period. To be sure, it was a relative lack of international cooperation (regarding current account imbalances) that helped to lay the groundwork; but unlike the Great Depression, the global financial crisis did not ultimately turn into a global conflagration. This was at least partly due to what John Lipsky (2010), First Deputy Managing Director for the IMF, described as the “unprecedented anti-crisis measures [implemented through enhanced international cooperation, which] included the largest ever coordinated counter-cyclical budgetary actions, rapid and massive rate cuts by major central banks and their provision of unprecedented sums through currency swap lines to support global market liquidity” (n.p.). In addition, according to Lipsky, the largest economies acted in concert to provide “large increases in the resources available to international financial institutions—including a tripling of the resources available to the IMF, among other effects helping to cushion the poorest countries from the brunt of the crisis.” International cooperation was facilitated through G20[26] summit meetings, the first of which was held in Washington, DC, in November 2008, followed by a second meeting in London in April 2009 (in addition, there was a great deal of less formal communication between government officials throughout the crisis). Importantly, the G20 was created in 1999 in response to the financial crisis of that decade, in recognition of the fact that the growing economic power of so-called emerging economies could no longer be ignored or marginalized in discussions of global economic issues. Since then, the G20 has been considered a major mechanism for international economic cooperation. The effectiveness of international cooperation during the global financial crisis can certainly be criticized, but the critical point is that a practical and normative framework for cooperation existed in the first place, and that it was used for coordinating policy responses and helping to prevent the crisis from getting out of control. (The topic of global governance will be discussed in more depth in chapter 8.)

 

Chapter 5: Conclusion

            This chapter has covered a lot of ground, but has also left a lot more ground essentially uncovered. This is unavoidable. The global financial system is immensely complex and far ranging; to cover it adequately would require several stand-alone volumes. The goal of the chapter, however, is simply to provide a basic and useful framework for understanding and interpreting the global financial system and major processes and events within that system from a political-economy standpoint. At a minimum, this means recognizing that the relationship between markets and states—in an increasingly globalized financial system—is complex and increasingly reciprocal. The complex and reciprocal relationship between states and markets also tells us that no one actor or set of actors is all-powerful. That is, power in a globalized financial system is also diffused among state and nonstate actors. This diffusion of power can be extremely messy, as different actors with divergent, oftentimes conflicting interests, endeavor to achieve their goals in concert with or in opposition to others. But this is what the study of international or global political economy is all about.


Chapter 6

 

 

 

 

Transnational Production, Foreign Direct Investment, and Economic Development

 

 

 

Introduction

            The previous two chapters focused on two major elements of the international or global political economy: trade and finance. Both these areas have seen major developments during the 20th century, especially in the latter half of the century. There is a third, integrally related element that has undergone equally dramatic change and development since 1945—namely, the production or manufacturing system. More specifically, what was once a largely disconnected agglomeration of domestically based production systems has become increasingly linked on a globalized or transnational basis. To be sure, as with trade and finance, cross-border activity in manufacturing has taken place for a long time. The era of colonialism, which goes back many centuries, provides good examples of early endeavors to create transborder or transnational production networks. In particular, colonies were economically—typically, in a highly exploitive manner—with the imperial economies. Today, however, transnational production networks are immensely more complex and immensely larger in scale and scope than at any other time in history. They are arguably less exploitative, too, although many, if not most, Marxist and neo-Marxist analysts assert that the contemporary system of transnational production remains extremely and even necessarily exploitative. The question of whether transnational production is exploitative will be discussed later in this chapter (and in the following chapter). For now, suffice it to say that, as with the transnational production system as a whole, the issue of exploitation is complex and difficult to untangle.

In addition to the question of exploitation, any examination of the development of the contemporary transnational production system must deal with a range of issues and factors. One of the most important of these revolves around the basic building block of any such system— namely, the firm, and more specifically, the transnational corporation (TNC). TNCs, as noted in chapter 1, have become ubiquitous in the global economy, with as many as 82,000 such firms, along with more than 810,000 affiliates. To better appreciate the increasing significance of TNCs, consider the phenomenon of intra-firm trade (which, most simply, can be defined as the cross-border flow of goods and services between parent companies and their affiliates, or among affiliates). According to Lanz and Miroudot (2011), it is likely that intra-firm trade accounts for a significant share of total world trade.[27] In the case of the United States specifically (the only country that keeps detailed statistics on intra-firm transactions), intra-firm trade accounted for 48 percent of imports and 30 percent of exports in 2009 (p. 12). Although the figures for the U.S. are likely higher than for most other countries—simply because the U.S. has a larger number of major TNCs—the level of intra-firm trade suggests that it has become a major part of the global economy.

The predominance of TNCs from the United States underscores another important issue, which is simply that major TNCs are concentrated in the developed world. Unsurprisingly, for most of the postwar period, the largest and most economically powerful TNCs came from, or were almost exclusively based in, the United States and a handful of mostly Western countries, including: the United Kingdom, France, Germany, Switzerland, the Netherlands, and Japan (the one major non-Western case). Significantly, though, while firms based in developed-world countries continue to dominate global networks of production, over the past several decades, firms from other regions have begun to emerge. In 2012, for example, among the 100 largest nonfinancial TNCs (based on foreign assets) were firms from Hong Kong SAR[28] (Hutchison Whampoa Limited), mainland China (CITIC Group and China Ocean Shipping), Taiwan (Hon Hai Precision Industry), Mexico (América Móvil and Cemex), Russia (VimpelCom Ltd.), and Brazil (Vale)—a total of eight (UNCTAD 2013, Annex Table 28). Five years earlier, in 2007, there were five TNCs from the so-called developing world in the top 100 (two of these were based in South Korea and one in Malaysia); and in 1995, there were only two (one company based in South Korea and one in Venezuela). The capacity of TNCs from the developing world to break into the upper echelons of global production is, to many analysts, a significant development. At the same time, there is little doubt that global production—as well as cross-border trade and investment—continues to be dominated by a fairly standard list of firms from a small number of advanced capitalist economies. From a political-economy perspective this raises important questions. Is this continuing dominance a product of market dynamics primarily, or does it reflect a highly unequal and thoroughly embedded distribution of power in the global political economy? If so, what are the implications of an entrenched “hierarchy of positions” (Philips 2013, p. 32) in the global production structure (and what have the consequences been for the past five or six decades)? It is also important to answer the question, why has so much production become transnational in the first place? There is a relatively obvious answer to this last question, but there are also less obvious answers that have to do with political-economic, rather than primarily economic dynamics. Finally, for the purposes of this chapter, there is one more question that needs to be addressed, one that has to do with a different, but closely related issue. The question is this: How do the structure and processes of transnational or global production affect the prospects for economic development in the developing world?


Figure 6.1. A Profile: The Rise of Hon Hai Precision Industry

 

Although the name might not ring a bell for most readers, Hon Hai Precision Industry is perhaps the dominant player in the global electronics industry. Better known by its trade name, Foxconn, in 2013 the company was the largest electronics manufacturer in the world. It is best known as the main supplier for Apple products (including the iPad, iPhone, and iPod), but it also produces products for Amazon (Kindle), Sony (PlayStation), and Nintendo (Wii), as well as many other companies. Altogether, the company manufacturers about 40 percent of all consumer-electronics products sold throughout the world (Duhigg and Bradsher 2012).

 

The company was founded in 1974 (in Taiwan), as a supplier of electronic components primarily to Western-based firms. (On this point, it is useful to note that, for companies in the developing world, their first connection with richer, brand-name firms is typically based on an unequal relationship in which the developing-country firm supplies products for the developed-country firm.) For many companies in the developing world, the supplier relationship is tenuous, for as costs begin to rise domestically—which was certainly the case for Foxconn in Taiwan—the dominant firm or firms will often relocate to lower-cost locations. Foxconn, however, was able to avoid this problem by relocating its own production base from Taiwan to other countries, most prominently China. In 2013, Foxconn employed approximately 1.4 million workers in China in 13 factories. One of the company’s factory locations in China is a veritable city, with upwards of 450,000 employees. Foxconn also has factories in Brazil, Hungary, Slovakia, the Czech Republic, India, Japan, Malaysia, and Mexico. For a fuller and critical discussion of Foxconn global expansion, see Chan, Pun, and Selden (2013).

 

Image: A Foxconn factory in the Czech Republic.
Source: Nadkachna. Permission granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2.

This chapter will address all of these questions. First, however, it is important to address a very basic question—namely, what is transnational, or global, production? This question will be answered in the following section. In the same section, too, a number of other basic terms and concepts will be introduced, as well as some important facts and figures.

 

 

 

 


Transnational Production: Definitions, Concepts, and Basic Data

Text Box: Figure 6.3. A Modern Container Ship
 
The container ship seen above is the CMA Christophe Columb, an Explorer-class containership named after Christopher Columbus (but built by a South Korean company). When it was delivered in 2009, it was the largest passenger-carrying container ship in the world, measuring 365 meters long by 52 meters wide (about the size of four American football fields laid end to end). The ship can carry up to 13,300 twenty-foot equivalent containers. In 2013, however, there were even larger container ships in production. Indeed, the same Korean company that built the Christophe Columb, Daewoo Shipbuilding, planned to deliver ten 400-meter-long Triple-E-class ships in 2013 and 2014, with a capacity of 18,000 twenty-foot containers each. These ships will travel 184 kilometers using 1KWh of energy per ton of cargo.

Image Source: Huhu Uet. Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2.
Text Box: Figure 6.2. Transportation and Communications Costs Indexes, 1920–2000
 
Source: Copied from “The Geography of Transport Systems,” available at http://people.hofstra.edu/geotrans/eng/ch1en/conc1en/priceevol.html.
            Transnational, or global, production is relatively easy to understand, at least in its basic form. Most simply, it is a type of production in which different parts of the overall production process for a particular product take place across different national territories. To repeat an earlier point, this sort of production has been going on for a very long time, as even the simplest manufactured products often require inputs—especially raw materials or natural resources (Castro n.d.)—from other territories. Still, prior to huge advances in transportation (beginning in the late 1800s and continuing through the 20th century) and in communications, transnational production tended to be based on the necessity of sourcing a material that was only available from certain areas. One reason was fairly obvious: since transport and related costs were very high, it generally made more economic sense—given a choice—to source materials locally rather than globally.

Over the past century or so, however, the cost differential between producing locally and globally has not only equalized, but has begun to tilt in favor of transnational production. Consider, for example, the cost of air transport (expressed in constant U.S. dollars, using 2000 as the base year): the per-ton-kilometer price of transporting goods by air fell from $3.87 in 1955 to under $0.30 in 2004 (Hummels 2007, p. 138). Interestingly, there was no comparable decline in ocean shipping (during the postwar period), but as the World Bank (2008) notes, price trends for ocean shipping “do not factor in the total cost of door-to-door transportation” (p. 179). In this view, improvements in ocean shipping—most notably, the invention of containerized shipping—did lead to a dramatic decline in overall shipping costs. Thus, according to the World Bank, “[i]n 1956 the loading of loose cargo cost $5.83 a ton. When containers were introduced in that year, the loading cost was less than $0.16 a ton. So the main savings came from lower intermodal transfer costs. [That is, containerization] ... allowed goods to be packed only once and shipped over long distance using maritime, rail, and road transport” (p. 179).[29] Telecommunication costs decreased equally dramatically. In 1930, for example, a three-minute telephone call between London and New York cost $245 (in 1990 U.S. dollars); the same call in 1990 was $3 (UNDP 1999). By 2005, this had been reduced to $0.25, but with the advent of VoIP (voice over Internet Protocol) around 2004, international calls, including streaming video calls, have been reduced to essentially nothing.

There is, it is important to emphasize, another very important side to the equation, which is the cost differential that exists between different parts of the world for labor, land, and other localized inputs. In her well-read book, The Travels of a T-Shirt in the Global Economy (2009), Pietra Rivoli shows how it is cheaper to grow cotton in the United States, ship it to China, have Chinese cutters, spinners, knitters, and stitchers manufacture millions of t-shirts, and ship them back to the United States, than it is to have U.S. workers make the same t-shirts using U.S.-produced cotton. Another somewhat minor, but still telling, example is the production of boxed lunches (called bento in Japanese) that are produced, cooked, and packaged in California using local ingredients, and then shipped to Japan for sale to commuters on Japan’s biggest railway, the JR line (which also runs Japan’s famed bullet trains). “This innovative concept,” California trade secretary Lon Hatamiya stated in May 2000, “will introduce a California-grown, -prepared and -packaged product to one of ... [California’s] most lucrative foreign markets.” Despite the transport costs—the bento lunches must be transported 5,000 miles in refrigerated containers—the California-made lunches sell for about half the price of those made in Japan (Tempest 2000).

It is not, it is also important to understand, just cost differentials between and among different localities that drive transnational production: transnational corporations globalize their production processes for a variety of reasons. One of the main reasons that Japanese car companies—e.g., Toyota, Honda, Mazda, and Nissan—began to relocate plants to other parts of the world, and especially to North America, was to alleviate growing protectionist pressures. In the early 1980s, in particular, the U.S. government imposed voluntary export restraints (VERs) on Japanese-made cars: under this arrangement, Japanese automakers as a group were limited to exporting 1.68 million units to the United States from 1981 to 1983, and 1.85 million units for 1984 and 1985. Not coincidentally, Cornstubble (1998) argues, the major Japanese car manufacturers all began producing cars in the United States almost immediately. In fact, by 1994, the Japanese automakers were selling more U.S.-made cars than Japan-made cars in the United States (n.p.). There are other reasons as well for the shift to global production, especially for the production of capital intensive, durable goods. First, it reduces the risks of currency shifts, and second, it provides companies a better and more sophisticated understanding of local market conditions (Womack, Jones, and Roos 2007, pp. 211–212). Saxenian (2002) points to still another reason: the development of “transnational technical communities” and, more specifically, of transnational entrepreneurs who act as a “mechanism for the international diffusion of knowledge and the creation and upgrading of local capabilities” (p. 1). These are technically skilled and well-educated individuals, in immigrant communities, who travel back-and-forth between their home countries (e.g., India, Taiwan, or China) and other, more advanced capitalist countries, such as the U.S., to take advantage of entrepreneurial opportunities. They are able to bring their knowledge, skills, and know-how back to their home countries, and create businesses that complement and fit into existing global-production networks. Embedded in this notion of transnational technical communities, it is useful to note, is a strong cultural element. Transnational entrepreneurs, in other words, are not just motivated by profit opportunities, but also by their personal, familial, and cultural connections to their home countries. (See figure 6.4 for further discussion.)

 

Figure 6.4. Transnational Technical Communities: Silicon Valley and the Argonauts

 

In Greek mythology, the Argonauts were a band of adventurers who sailed with Jason in search of the Golden Fleece. The new Argonauts come from Asia: they first ventured to the U.S. or other rich Western countries (where many studied engineering and/or other highly technical subjects) to escape from harsh economic and/or political conditions, but as economic and political conditions in their homelands improved over the decades, many have ventured back “to take advantage of their experience in and linkages with leading high-tech regions” (Sternberg and Müller (2007, p. 1). In so doing, they have forged concrete economic linkages between their new and old homelands, and (to some extent) have upended the traditional relationship between so-called core and peripheral economies. This has been made possible, in part, by the “fragmentation of production and the falling costs of transport and communication [which] allow even small firms to build partnerships with foreign producers to tap overseas expertise, cost savings, and markets” (Saxenian 2006, p. 103).

 

A key part of this process is the back-and-forth, or circulatory, movement of people between locations. The new Argonauts do not simply return to their homelands, but instead maintain strong connections between locations; at the same time, they may develop a whole new network of domestic and transnational connections. Thus, they may return to Taiwan or Shanghai to begin a new firm, but this new firm will immediately become part of a broader network with ties to investors in China and the U.S. (or other countries); with personal, professional, and governmental connections in both countries; with cross-border customer and vendor relationships; and so on. The figure below, reproduced from Sternberg and Müller (2007, p. 11), provides an illustration of the type of transnational network created around “returned entrepreneurs.”


Description: Macintosh HD:private:var:folders:3r:pz8t0txj4lvcmd_nlwkbdfzc0000gq:T:b1m9sW:LittleSnapper.png
RIS = Regional Innovation System
NIS = National Innovation System

 

One salient result of the general shift to global production has been the creation of the so-called global factory, which is set up to seamlessly exploit differential opportunities that arise in fabrication, assembly, quality control, R&D, design, technology, regulatory environments, marketing, and so on, in locations around the world. Firms that have this capacity, it is important to emphasize, generally are among the largest, most economically powerful firms in the world: among the top 25 TNCs (measured in terms of foreign assets), 20 are also among the top 100 largest firms in terms of total revenue (see table 6.1, “Rankings of the World’s Largest Firms by Foreign Assets and Total Revenues”). Their sheer economic size, in turn, allows global factories to, among other things, purchase potentially competitive firms in host countries, thereby extending their reach in key markets and dominance over key technologies. Their economic size, too, gives them immense purchasing power—often verging on monopsony (i.e., a market situation in which there is only one buyer)—such that they can almost dictate the prices they will pay (Buckley 2009, p. 137). The rise and development of global factories, therefore, is an extremely significant phenomenon, and one that necessarily impacts the dynamics of the global political economy. Before discussing the impact of global factories, it will be useful to discuss another essential element in the shift toward global production—namely, foreign direct investment, or FDI.

 

Table 6.1. Rankings of the World’s Largest Firms by Foreign Assets and Total Revenues

 

Ranking by:

 

 

Foreign assets

TNI*

Global 500

Corporation

Home economy

1

79

22

General Electric Co

United States

2

32

1

Royal Dutch Shell PLC

United Kingdom

3

22

4

BP PLC

United Kingdom

4

77

10

Toyota Motor Corporation

Japan

5

28

11

Total SA

France

6

45

2

Exxon Mobil Corporation

United States

7

8

**

Vodafone Group PLC

United Kingdom

8

62

33

GDF Suez

France

9

61

8

Chevron Corporation

United States

10

64

12

Volkswagen Group

Germany

11

51

17

Eni SpA

Italy

12

1

71

Nestlé SA

Switzerland

13

71

52

Enel SpA

Italy

14

48

16

E.ON AG

Germany

15

4

**

Anheuser-Busch InBev NV

Belgium

16

6

70

ArcelorMittal

Luxembourg

17

29

47

Siemens AG

Germany

18

36

64

Honda Motor Co Ltd

Japan

19

92

**

Mitsubishi Corporation

Japan

20

98

**

EDF SA

France

21

73

21

Daimler AG

Germany

22

67

89

Deutsche Telekom AG

Germany

23

66

**

Pfizer Inc

United States

24

40

69

BMW AG

Germany

25

42

82

Telefonica SA

Spain

 

* TNI, the Transnationality Index, is calculated as the average of the following three ratios: foreign assets to total assets, foreign sales to total sales, and foreign employment to total employment.

** Not in top 100 (but may be ranked in Global 500, positions 101 to 500)

Global 500 is the annual rankings compiled by Fortune magazine of the 500 largest companies, by revenue, in the world. The list is available at http://money.cnn.com/magazines/fortune/global500/2012/full_list/index.html.

 

Source for foreign assets and TNI ranking: UNCTAD (2013).

 

FDI is, in the most general terms, investment made in a company or entity based in one country by a company or firm based in another country. A more specific definition is provided by the World Bank, which defines FDI as follows: “the net inflows of investment to acquire a lasting management interest (10 percent or more of voting stock) in an enterprise operating in an economy other than that of the investor. It is the sum of equity capital, reinvestment of earnings, other long-term capital, and short-term capital as shown in the balance of payments” (n.p.). What makes FDI direct is the fact that firms (or investors) making such investments actively participate in managing the companies or factories in which they invest. This contrasts with portfolio investment (or indirect investment), which typically involves purchasing shares of a firm’s stock or corporate bonds, but does not entail any management interest or responsibility for operations. The key point for our purposes, however, is this: FDI serves as a useful proxy for measuring the level of transnational or global production because it tells us how much direct investment companies are making outside their home countries.          

 

 

Table 6.2. Trends in FDI, 1913–2004 (FDI as a Percentage of GDP)

 

 

1913/14

1930s

1950s

1970/71

1980

1995

2003/04

Developed Countries

Outward Stock of FDI (as percent of GDP)

France

23

10

--

5

--

25

37

Germany

11

5

--

3

4

10

31

Japan

11

47

--

2

2

5

8

Netherlands

82

28

--

35

25

47

94

United Kingdom

49

18

9

17

15

28

65

United States

7

8

4

8

8

18

17

Developing Countries

Inward Stock of FDI (as percent of GDP)

Avg. for All Colonies

42

61

35

14

--

19

--

Postwar Averages

 

 

 

 

 

 

 

   Latin America

--

--

--

--

4

12

38

   Asia

--

--

--

--

4

12

24

   Africa

--

--

--

--

8

15

32

 

Sources: Table reproduced from Velde (2006), p. 5

 

 

Interestingly, the relative level of FDI was higher in the early 1900s than for the rest of the 20th century; as late as 1995, outward FDI, as a percentage of GDP, was still below what it was at the beginning of the century (see table 6.2). This does not mean that production was especially globalized a century go. It was not. Instead, it reflects the then-urgent need on the part of wealthier countries to develop sources of raw materials and natural resources for their rapidly industrializing economies. This type of FDI, by the way, is referred to as resource-seeking FDI. As Velde (2006) points out, in 1913, two-thirds of world FDI was flowing to developing countries (p. 6)—mostly colonies—and almost all of that investment was for exploiting natural resources and building the railways needed to transport these commodities back to the West (p. 7). The situation is dramatically different today. Now most FDI “is amongst developed countries, and only a quarter of FDI is going to developing countries” (Velde 2006, p. 6); moreover, there has been a marked shift towards efficiency-seeking and strategic asset-seeking FDI. The aim of efficiency-seeking FDI is to increase competitiveness by lowering production costs (this is also referred to as offshoring). Strategic asset-seeking FDI aims at “advancing a company’s global or regional strategy into foreign networks of created assets like technology, organizational abilities and markets” (Wadhwa and Reddy 2011, p. 220; citing Faeth 2009). A fourth form of FDI is market-seeking FDI, which is based on gaining access to local or regional markets, whether for primarily economic or political reasons (e.g., Japanese automobile companies were motivated to invest in the United States because of VERs, while American auto-manufacturing investments in China are primarily based on the advantage of being located in a major and rapidly expanding consumer market for automobiles: both are examples of market-seeking FDI). While resource-seeking FDI has certainly not disappeared, efficiency-seeking and strategic asset-seeking FDI are particularly important for creating the global factory and, by extension, a complex transnational-production structure.

Over the past few decades, the total stock of FDI (which includes investment in both manufacturing and services) has grown dramatically. In 1990, the inward stock of FDI was $2,078.3 billion, while outward stock was $2,091.5 billion. By 2000, those figures had grown to $7,511.3 billion and $8,025.8 billion respectively, and by 2012, the respective figures were $22,812.7 billion and $23,592.7 billion.[30] In other words, over a period of a little more than two decades, the stock of both inward and outward FDI increased more than ten-fold, or 1,000 percent. By 2012, too, as a percentage of GDP, the outward stock for developed countries reached 42.8 percent (up from 11.9 percent in 1990), while the inward stock increased to 33.4 percent from 8.9 percent. For developing countries, the increase in inward stock went from 13.4 percent in 1990 to 30.4 percent in 2012 (all figures cited in UNCTAD 2013[31]). Combined with the equally rapid growth in the number of TNCs, these figures point to a significant reorganization of the global economy. In other words, the dramatic rise in FDI—lead by TNCs—is reorganizing the global economy by expanding and deepening the integration of national economies; this represents a very different phenomenon than simple or shallow cross-border exchange via trade (as discussed in chapter 4). TNCs are creating new and more complex linkages that have, more than ever before, tightly coupled the world economy.

            The term TNC was covered in chapter 1, and there is definition in the glossary, but let us look at another basic definition. UNCTAD (n.d.) provides a useful definition on its website: “Transnational corporations (TNCs) are incorporated or unincorporated enterprises comprising parent enterprises and their foreign affiliates. A parent enterprise is defined as an enterprise that controls assets of other entities in countries other than its home country, usually by owning a certain equity capital stake” (http://unctad.org/en/ Pages/DIAE/Transnational-corporations-(TNC).aspx). This is a fairly standard definition, and one that largely suffices (although there are some scholars who disagree).[32]

            The rise of the TNC is also strongly connected to two related phenomena: strategic alliances between TNCs (or other enterprises) and full-scale fusions, usually through cross-border takeovers (Scholte 1997, p. 437). Strategic alliances comprise a range of cooperative arrangements between legally separate (and occasionally competing) corporations; nonetheless, some of these arrangements involve a high degree of coordination and collaboration. Consider, for example, Samsung Corporation—a major Korean electronics manufacturer—which has had (or still has) strategic alliances with, among others, Apple, Nokia, Alcatel, Sony, IBM, Sun Microsystems, Matsushita, Qualcomm, NEC, and Microsoft. Samsung’s erstwhile alliance with Apple is particularly interesting, as the two global companies compete head-to-head in the markets for smart phones and tablets. Apple relied heavily on Samsung to produce proprietary chips (and high-resolution display screens) for the iPhone and iPad—proprietary chip making is an extremely expensive and difficult undertaking, which only a few corporations are capable of doing today (the other major players are Taiwan Semiconductor Manufacturing, Intel, and GlobalFoundaries [Vance 2013]). When Samsung agreed to make the chips Apple needed, therefore, Apple readily agreed, although it required a level of cooperation that was unusual between two companies that were, in many respects, direct rivals. Not surprisingly, then, this alliance also led to a great deal of friction, as Apple accused Samsung of imitating its designs for smartphones and tablets; in 2011, Apple sued Samsung. Later, Samsung countersued Apple; all the while, however, Apple continued to use chips made by Samsung, and will continue to do so until at least 2015 (Lessin, Luk, and Osawa 2013).

Another useful example is Tesla Motors, which produces electric vehicles (EVs). The automobile industry is notoriously difficult for new companies to enter—largely because any new company must compete against huge global firms with a mature infrastructure, established supply and distribution systems, and strong ties to customers and political entities (Burroughs 2012). One way to overcome the “liability of newness” is to develop strategic alliances, which is exactly what Tesla Motors has done. Tesla’s alliances include suppliers, R&D experts, and original equipment manufacturers (or OEMs), including Daimler and Toyota—two automotive behemoths. The Daimler alliance, according to Stevan Holmberg, has been particularly important for Tesla, since it “represented an endorsement by a premier automotive manufacturer that further enhanced and verified for the broader market Tesla’s competencies, technologies, and ability to deliver results”; importantly, Tesla also produced battery packs and chargers for Daimler’s Smart fortwo car, or “Smart car” (quoted from Burroughs [2007], n.p.).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 6.5. Tesla’s Strategic Alliances

Description: Macintosh HD:private:var:folders:3r:pz8t0txj4lvcmd_nlwkbdfzc0000gq:T:EicqlR:Microsoft PowerPoint.png

Source: Image created by author based on Burroughs (2012). Original image available at this link: http://kogodnow.com/2012/09/how-tesla-used-strategic-alliances-to-power-green-products/.

 

            Full-scale fusions are most often reflected in mergers and acquisitions (M&A), which are accounted for in FDI statistics. In the 1980s and 1990s, there was a flurry of M&A activity, although the large majority involved domestically based acquisitions: in 1998, only a quarter of total merger volume (about $406.4 billion) involved a cross-border acquisition, but by 2007 that figure had grown to $1.02 trillion, or 45 percent of total volume (Erel, Liao, and Weisbach 2012, p. 1045).[33] With the onset of the global financial crisis, M&A activity declined precipitously: in 2009, the volume of cross-border deals was a relatively small $249.7 billion, much lower than the average annual amount over the previous 14 years. Still, compared to the early 1990s, not to mention the 1970s and 1980s, this was still a substantial amount. It is important to add, too, that every year there are a good number of megamergers—those involving at least $1 billion in investment. In 2012, for example, there were exactly 200 M&A worth $1 billion or more, the largest of which—the acquisition of GDF Suez SA, a French company, by the United Kingdom’s International Power PLC (UNCTAD 2013, Annex Table no. 17)[34]—was worth $12.9 billion. The scope and scale of cross-border M&A since the mid-1990s have, according to Kang and Johansson (2000/01), made them into one of the “fundamental mechanisms of industrial globalisation” (p. 6); indeed, the two economists assert that cross-border M&A have become even more important than greenfield FDI (greenfield refers to brand-new operations or factories).

Text Box: Figure 6.6. The Volvo S40: A Product of a Transnational Production System

 

Source: Richard Baldwin, PowerPoint presentation available at http://www.wto.org/english/res_e/reser_e/wkshop_nov10_e/baldwin_e.pdf

The increasing fusion of firms across borders suggests that the basic definition of transnational production as “a type of production in which different parts of the overall production process for a particular product take place across different national territories” is, perhaps, a bit lacking. Hveem (2007) extends the definition by emphasizing the systemic aspect of transnational production. Thus, Hveem writes that transnational production systems [TPS] are comprised of “geographically distributed but integrated and more or less coordinated activities that include production, marketing and distribution functions organized across national boundaries. The TPS are usually institutionalized in long-term arrangements coordinated through active socialization or governed under some degree of centralized overall control (or both)” (p. 1). Consider, for example, the Swedish automobile manufacturer Volvo. In a breakdown of one particular model—the Volvo S40—at least 38 major and minor components were manufactured in factories spread throughout the world: Slovakia, Japan, France, Norway, Brazil, Germany, the United States, Canada, Holland, the United Kingdom, and, of course, Sweden. The hood latch cable, for instance, was manufactured by Klüster in Slovakia; the amplifier by Alpine (Japan); the engine control unit by Borgwarner (USA); the turbo diesel by Sanden (Japan); the drive shaft by GNK/Visteon (USA); the air conditioner by Valeo (France), the doors by Brose (Germany), and so on (Baldwin and Thorton 2008). In addition, it is likely that each of these manufacturers had their own transnational system of production. Figure 6.6, “The Volvo S40: A Product of a Transnational Production System,” provides a more detailed breakdown.

 

 

 

 

 

Table 6.3. Trends in Cross-Border Mergers and Acquisitions (By Value and Number)

 

 

1990–1996 Avg.

1997–2003 Avg.

2004–2008

Avg.

2009

2010

2011

2012

Value (by Sales) in Millions U.S.$

$79,797

$426,270

$434,866

$249,732

$344,029

$555,173

$308,055

Number of Cross-Border M&A (Sales)

2,542

4,470

5,575

4,239

5,484

6,065

5,400

Cross-Border M&A, Avg. Value (in millions U.S.$)

$31.4

$103.4

$78.0

$58.9

$62.7

$91.5

$57.0

 

Source: UNCTAD, “Annex Table 14 - Value of cross-border M&A purchases, by sector/industry, 1990–2012”; and “Annex Table 15 - Number of cross-border M&A sales, by sector/industry, 1990–2012.” Available online at http://unctad.org/en/Pages/DIAE/World%20Investment%20Report/Annex-Tables.aspx. The author calculated the averages.

 

 

Explaining the Transnational Production Structure

            With the foregoing discussion in mind, it is now time to move to a more substantive discussion of the transnational production structure (or system). On first glance, it may seem obvious why transnational production has become such a significant phenomenon. Indeed, in the previous section, a number of major factors were already identified, and one of these, the drastic decrease in transport and communications costs, would likely be fingered by most casual observers as the most important. After all, the drop in transport and communications costs has clearly made transnational production much more economically efficient—this is also reflected in the rise in efficiency-seeking FDI. Certainly, the search for greater economic efficiency is part of the answer; in this respect, too, it is important to underscore the development of new and better technologies, as well as improvements in global finance (Strange 1994). All of these changes have made it easier and more profitable to build integrated production systems across borders.

In keeping with this general theme, there are a number of other economically based theories that seek to explain the growth and deepening of transnational production. Dicken (2003) provides a very helpful overview. He begins with a macro-level Marxist approach that focuses on the concept, “circuits of capital.” Marx himself identified three circuits of capital: commodity capital, money capital, and productive capital. The term circuit is used to emphasize how money or capital invariably circulates through the three interconnected, but distinct processes, each of which adds value to the original amount. This idea—that capitalism is a spiral-like system designed to make money and commodities more valuable at the end of the productive process than at the beginning—is actually quite simple, and even self-evident. Still, it is an important insight that can be used to help explain the globalization of capitalism in general, and the globalization of production more specifically. The basic circuit of capital works in the following manner: first, money (referred to as M) is used to pay for factors of production, or basic commodities, such as raw materials and labor (C). Second, through a productive process (P), these basic commodities are then transformed into brand-new commodities (C’), making them more valuable than the original cost. The new commodities are then sold; the profit or surplus value is turned back into money (M’), and the process repeats itself, ad infinitum. The basic “equation” used to express this ongoing process is M—C ... P ... C’—M.’

The reason this leads to globalization is equally simple. In the commodity-capital circuit, many raw materials need to be sourced internationally, which means that some level of cross-border activity must usually (albeit not necessarily) begin almost immediately. In addition, and more importantly, the sale of the newly produced commodities requires markets. Domestic or local markets are the obvious first choice, but these can become easily sated (especially as the productive process becomes more efficient). Thus, there is a strong and perhaps irresistible tendency toward cross-border trade. This is the first major phase in the internationalization of capitalism. Over time, as more surplus value is produced, new markets are also needed for investment capital; this leads to the globalization of the money-capital circuit (the second phase). The third phase is the globalization of the productive-capital circuit, which is the primary topic of this chapter. All three circuits, it is important to reemphasize, are part of an interconnected whole. This may all sound a bit arcane, but the main point is quite simple: capitalism is a constantly repeating process; as such, it has an inherent tendency to expand, first domestically, but then internationally or transnationally. In so doing, it inexorably connects and integrates national economies through trade, finance, and production. The globalization of productive capital (i.e., transnational production), more specifically, is a reflection of capitalism’s need to make the circuit operate at the highest possible velocity. “This requires”, as Murray (2006) explains it, “the development of space-shrinking and time-saving technologies which reduce the turnover time of capital. It is the development of such technologies, based on the capitalist imperative of maximizing profit, that ... has led to time-space compression and globalization as we know it” (p. 97; citing Harvey [1989]).

The advantage of the “circuits” approach, according to Dicken (2003) is that it emphasizes the interconnected and systemic character of trade, finance, and production (p. 210)—a very good lesson to keep in mind. At the same time, Dicken criticizes the Marxist approach for its inability to explain transnational production at a more specific level. Why are certain geographic areas chosen as sites for transnational production? After all, transnational production is not evenly distributed around the world; instead, it tends to concentrate in certain regions. Why are some kinds of organizational arrangements, or sector-specific decisions, made over others? What motivates firms to make the decision to engage in transnational production when they do? To answer these and similar questions, Dicken tells us that we must consider micro-level approaches. (By “micro-level,” Dicken means a firm-specific rather than a general system-level view.)

Dicken discusses a number of firm-specific arguments, the most comprehensive of which is John Dunning’s “eclectic paradigm,” which was first articulated in 1976 (see figure 6.7, “The Product Life Cycle and Transnational Production,” for a discussion of one other micro-level approach). Interestingly, Dunning did not consider his eclectic paradigm to be firm-specific: in a 1988 article, he asserted that it had “only limited power to explain or predict particular kinds of international production; and even less, the behavior of individual enterprises” (p. 1). Nonetheless, the paradigm provided a general explanation for why firms begin to engage in international and transnational production. As Dunning explained it:

 

In its original form, the eclectic paradigm stated that the extent, form, and pattern of international production was [sic] determined by the configuration of three sets of advantages as perceived by enterprises. First, in order for firms of one nationality to compete with those of another by producing in the latter's own countries, they must possess certain advantages specific to the nature and/or nationality of their ownership. These advantages sometimes called competitive or monopolistic advantages must be sufficient to compensate for the costs of setting up and operating a foreign value-adding operation, in addition to those faced by indigenous producers or potential producers (1988, p. 2).

 

            To put it in much simpler terms, before the decision to engage in transnational production is made by a specific firm, it must be reasonably clear that there is a sound economic basis for doing so. A second important and related condition, according to Dunning, is that there must be a compelling reason for a firm to not only locate production outside its home territory, but also outside the firm itself. Generally speaking, firms tend to internalize economic transactions in order to safeguard supplies of essential inputs, to ensure the quality of end products, to guarantee markets, to protect property rights, to spread the costs of shared overheads, and so on; thus, for a firm to transfer certain activities across national borders, there must be strong incentives (p. 3). On this last point, it is important to understand that Dunning’s model assumes that markets are imperfect—that is, that there are market failures. If there were not, then there would be no economically rational reason for firms to engage in transnational production. To understand the logic here, consider the issue of uncertainty. In real-world markets, firms cannot be sure that the intermediate supplies or other essential inputs they need for production will be available in the quantity and quality, and at the price, needed to ensure a consistent profit—especially if those goods are only available in certain foreign markets. It becomes rational, then, for firms to minimize or eliminate that uncertainty by taking control of the production of the goods they need by directly investing in the foreign markets in which those goods are located. Indeed, uncertainty is the major incentive for a firm to internalize factor or product markets (Dicken 2003, p. 205).

Text Box: Figure 6.7. The Product Life Cycle and Transnational Production

Product life cycle refers to the period of time over which a particular product—e.g., a videocassette recorder (VCR)—is first developed, brought to the market (sold), and finally withdrawn from the market. In addition to development, introduction, and withdrawal, there are two other stages: growth and maturity. In the introductory stage, the product may experience little competition in the marketplace, but (assuming that the product is successful) eventually new competitors will enter the market, eroding the position of the leading firms. Over time, the product becomes standardized and widely available: this makes earning a profit on sales of the product more and more difficult. Makers of the product must find ways to reduce costs. At some point, the product goes into decline, eventually becoming obsolete.

  
The significance of the product-life-cycle concept to transnational production is simple: in the early phases of a product’s life, production is typically centered in the originating country. Overseas demand for the product is met through exports. As the product matures, however, production tends to be (but is not necessarily) shifted across borders to lower-cost areas, or so that production can be closer to foreign markets. Dicken (2003) notes that the product-life-cycle concept explains only part of the process by which production is transnationalized (p. 204). 

Image source: Perreault, et al. (2008). Permission is granted to copy, distribute, and/or modify the image under the GNU Free Documentation License, ver. 1.2.
Finally, Dunning argues that there must be clear-cut locational advantages that make it profitable for the firm to use its assets in foreign as opposed to domestic locations. In other words, a location has to have something a firm needs that cannot be found in the firm’s home country. This could be as simple as low-cost labor or resources; it could also be access to a country’s consumer market. On this last point, it is important to emphasize that the choice of location may also be prompted by what Dunning refers to as “spatial market failure,” by which he means barriers to trade. As Dunning puts it, “historically the imposition of trade barriers has led to a lot of foreign manufacturing investment by [TNCs]” (Dunning 1988, p. 4). This last point provides a nice segue into another type of explanation. While Dunning and others recognize that political and institutional factors can play a role in the development of transnational production, the emphasis is primarily on economic factors and questions of efficiency. But economic efficiency is not the only—or necessarily the most important—factor driving the emergence and development of transnational production. Another factor, also mentioned above, hinges on state policy: to repeat an earlier example, it is clear that the proximate cause of the shift by Japanese automakers from primarily domestically based production to transnational production in the mid-1980s was the imposition of VERs by the United States. This, of course, is the same point made by Dunning; still, it is important to consider the issue in broader and more systematic terms.

 

The Political Context of Transnational Production: A Focus on the Auto Industry

            Previous chapters have emphasized the importance of the political framework within which all economic activity necessarily takes place (this framework shapes, and is shaped by, economic activity). The lesson has become almost banal in the context of this book; yet, it is still one that needs to be highlighted. Recall, for instance, that free trade—to the extent that it exists—reflects the outcome of complex processes and relations of power, all of which play out within domestic, international, and global structures. With this in mind, one useful way of explaining the emergence and development of the transnational production system is to bring the focus down to a concrete level by examining a particular industry: automobile production. The auto industry is a good industry to focus on since it encapsulates many of the key issues involved in the process of transnational production. To be sure, the transnationalization of automobile production has its own distinctive aspects, which means that the experiences of the industry are not wholly generalizable.[35] Nonetheless, the industry’s experiences are broadly instructive and important, particularly since automobile production remains an area of major economic significance.

Text Box: Figure 6.8. Ford Motor Plant, Windsor Ontario (1920)
 
Source: Canadian Postcard Company. The image is in the public domain in the United States because it was published before January 1, 1923.

To begin, it is useful to note that the automobile industry, as Dicken (2003) writes, “is essentially an assembly industry. It brings together an immense number and variety of components, many of which are manufactured by independent firms in other industries. It is a prime example of a producer-driven production chain” (emphasis in original; p. 355). Given the “immense number and variety of components” there has long been an important cross-border element to automobile production, in that certain materials—especially raw materials (e.g., rubber, glass, steel, and aluminum)—have always been sourced internationally. Still, production and assembly of the major (and high-value-added) components has tended to be located within the borders of a single country. Automobile production also has tended to be heavily concentrated in just a handful of major economies. In 1960, just six countries—the United States, Germany, the United Kingdom, France, Italy, and Canada (listed in order of production)—accounted for almost 92 percent of the total worldwide production of automobiles (OECD 1983). Significantly, Japan was not in the top six in 1960: it held the seventh spot, but was far behind Canada (that year, Canada produced 323,000 units, while Japan produced just 165,000). Canada’s inclusion on the list, it should be noted, reflects the limited degree of transnational production that had occurred prior to the 1960: both Ford Motor Company and General Motors (GM) began production operations in Canada in the early 1900s. Ford’s Canadian facility, however, was located just across the Detroit River in the city of Windsor, while GM acquired the Canadian company McLaughlin, which was located in Oshawa, Ontario (about 260 miles from Detroit). Both Ford and GM also established assembly operations in Europe, Latin America, Australia, and Japan in the 1920s. Not surprisingly, these early transnational operations were, according to Dicken (2003), “triggered primarily by the existence of protective barriers around major national markets as well as by the high cost of transporting assembled automobiles from the United States” (p. 378).

Despite the early entry into transnational production by Ford and GM, little changed for more than half a century. The cross-border operations of the two American companies remained generally limited (World War II, in fact, forced a significant drawback in U.S. cross-border operations, while Japan’s mercantilist policies completely shut U.S. companies out of the Japanese market even before the war[36]); and, in Europe and Japan, automobile producers remained firmly rooted in their local landscapes. In Europe, the situation began to change in the 1970s. A major reason for this change, according to Dieter (2007), can be attributed to the European integration process—that is, the establishment of the European Union (which began with the creation of the European Coal and Steel Community in 1951). For transnational production, the key element of the integration process was the creation of a single regulatory sphere, which did two things. First, it enlarged the space of business (i.e., it created a larger market). Second, it enabled the enlargement of the area available for sourcing of components without having to consider local content requirements, which were common among the European economies at the time (Dieter 2007, pp. 17–18). The second factor came to play a particularly prominent role with the collapse of the socialist regimes in Eastern Europe after the fall of the Berlin Wall in 1989. Within a few years, the EU forged a regional trade agreement—the Central European Free Trade Agreement (CEFTA)—that originally included Poland, Hungary, and Czechoslovakia (after the breakup of Czechoslovakia, the Czech Republic and Slovakia joined separately). The CEFTA had two main components. First, it completely eliminated tariffs among the parties in the agreement, and, second, it raised tariffs between the CEFTA and the rest of the world (van Tulder 2004).

Text Box: Figure 6.9. Fiat Auto Factory in Poland

 
Source: http://www.kocjan.pl/ • This file is licensed under the Creative Commons Attribution-Share Alike 3.0. Unported license.
The new political framework led very quickly to a significant, albeit not huge, inflow of auto-related FDI to Eastern and Central Europe, led by Western European companies. Poland was, by far, the major recipient. In terms of total FDI, Poland’s inward FDI stock in 2000 was $34 billion, compared to $23 billion in Hungary (which was second to Poland). By 2009, inward FDI stock in Poland had grown to $182 billion (cited in Zimny 2010, p. 9). The major European investors in Poland’s auto industry are Fiat, Volkswagen, Volvo, and MAN Nutzfahrzeuge (a German truck company); more recently, Japanese and Korean automobile companies have
invested in Poland. American car companies—Chevrolet and GM—also have operations located in the country. Indeed, except for one company (Solaris), the entire Polish car industry is based on foreign investments. Poland’s initial appeal was likely the country’s long experience producing cars during the socialist era: in 1989, the last year of socialist rule, the country produced about 289,000 units (during the socialist era, in fact, Poland was already producing cars for Fiat through a joint venture). After the CEFTA was implemented, but especially after Poland joined the EU, production in Poland ramped up. By 2004, Poland was producing over 600,000 units (passenger and commercial cars), and just four years later, in 2008, the country was nearing the one-million-unit production mark. (In more recent years, however, production in Poland has not only started to slump, but other Eastern European countries—the Czech Republic and Slovakia—have overtaken Poland.) In addition to finished automobiles, Poland also produces car engines, tires, and other parts. Significantly, almost all cars and auto parts produced in Poland are exported—about 98 percent—mostly to other European Union countries (Bulinski 2010, p. 3).

Text Box: Table 6.4. World’s Leading Motor Vehicle Producers

Country/Region	2012	1990	1960
China	19,271,800	509,242	22,574
European Union	16,240,476	n/a	n/a
United States	10,328,884	9,782,997	7,905,119
Japan	9,942,711	13,486,796	481,551
Germany	5,649,269	4,976,552	2,056,149
South Korea	4,557,738	1,321,630	--
India	4,145,194	362,655	51,136
Brazil	3,342,617	914,466	133,041
Mexico	3,001,974	820,558	49,807
Thailand	2,483,043	304,843	--
Canada	2,463,732	1,947,106	397,739
Other Selected Countries (World Rank 2012)
Czech Republic (15)	1,178,938	242,000	75,000
Indonesia (16)	1,065,557	--	--
Slovakia (19)	900,000	--	--
Poland (22)	647,803	347,975	37,000

Source: Organisation Internationale des Constructeurs d'Automobiles (OICA), various years (http://www.oica.net/category/production-statistics/)
In other regions, similar processes unfolded. That is, major steps in the transnationalization of automobile production were preceded by political integration, usually in the form of a regional trade agreement. In Latin America, to cite another example, the creation of Mercosur (short for the Spanish phrase for “Common Market of the South”), a regional trade agreement among Argentina, Brazil, Paraguay, and Uruguay (which was a product of the 1991 Treaty of Asunción), created the basis for increased FDI in South America. The Ouro Preto agreement of 1994, which reinforced Mercosur by establishing a customs union, further spurred FDI by providing greater political credibility for economic integration in the region. Indeed, almost immediately after the Ouro Preto agreement was signed, Fiat decided to set up a new production complex in Argentina with an investment of $600 million to produce 180,000 vehicles (Balcet and Enrietti n.d, p. 11). Balcet and Enrietti put this issue very bluntly: “Regional integration”, they write, “may be considered ... a necessary condition to the development of an intra-firm division of labour on a regional level” (p. 12).

In North America, the adoption of NAFTA had a similar effect. To be sure, as noted earlier, U.S. companies were already engaged in transnational production prior to the mid-1990s (NAFTA was signed in 1994), but NAFTA helped to accelerate a regional reorientation of the North American auto industry, from one that traditionally stretched east to west, emanating from Detroit, to one that now stretches southward from Detroit to the Gulf of Mexico (Moavenzadeh 2006). This reorientation has turned Mexico into a major center of automobile production: in 2012, light-vehicle production reached a high of 3.57 million units, which made Mexico the eighth largest producer of light vehicles in the world, ahead of Canada (in 1989, Mexico produced just 439,000 vehicles to Canada’s 984,000); Mexico is also the world’s fifth largest producer of auto parts. Most of the cars produced in Mexico are exported, and most of Mexico’s car exports (more than 68 percent) go to the United States (U.S. Embassy–Mexico City 2013). Asia has been somewhat of an exception, in that there are no major regional FTAs involving both developed- and developing-world economies, but the transnationalization of automobile production is still premised on the construction of a political framework. Consider the case of Thailand.

In the early 1960s, the Thai government was primarily concerned with protecting the Thai auto industry, and imposed high tariffs on imports (up to 60 percent). At the same time, the country needed foreign technology and know-how, so it promoted FDI—but limited auto-related FDI to joint-ownership deals. There were a variety of other policies designed to promote local industrial development, but they generally failed to generate significant investment. Over time, this led to a shift toward a more open investment policy, which was most clearly reflected in the country’s adoption of the WTO agreement on Trade Related Investment Measures (TRIMS), which Thailand was the first developing-world economy to adopt. The government also eliminated local content requirements and most other restrictions on FDI, including the joint-ownership requirements (import tariffs, however, remained at fairly high levels). Thailand also signed a series of bilateral FTAs, including deals with Australia (2005), New Zealand (2005), and Japan (2007). Although not immediately apparent, the FTAs with Australia and New Zealand, as well as the existing ASEAN free-trade area (originally signed in 1992), were important parts of Thailand’s ascendance as a major regional center of auto production. Toyota, in particular, exports most of its Thai-produced vehicles to ASEAN member states, Australia, and New Zealand (all data cited in Athukorala and Kohpaiboon n.d.).

Importantly, Thailand’s neighbors in Southeast Asia—e.g., Malaysia, Indonesia, and Vietnam—largely failed to follow suit. That is, Thailand got a significant head start in liberalizing the investment and trade environment for automobile production, and therefore was able to create distance between itself and neighboring countries (which otherwise might also have been attractive locations for investment). The result for Thailand was a stunning rise in domestic automobile production: Thailand, which produced no cars in 1960, had become the 9th largest car producer in the world in 2012 (producing 2.45 million units), just behind Mexico, and the 7th largest auto exporter (about 1 million units). Not surprisingly, however, Thailand’s “big three” manufacturers are all foreign—specifically, Japanese—companies: Toyota, Isuzu, and Honda. (Ford, Daimler Chrysler, and GM also have operations in Thailand.)

The transnationalization of production in the automobile industry, in sum, provides a near-ideal window through which to view the globalization of the production process. On the one hand, the economic imperatives (and benefits) behind the globalization (and regionalization) of production are quite clear. On the other hand, it is equally clear that the globalization and regionalization of automobile production invariably takes shape within a political framework. This political framework, to repeat a key point, is not merely a supplemental part of the process, but is instead an essential characteristic. In addition, while the transnationalization of automobile production has its own distinctive aspects, it is far from unique. To a significant extent, most manufacturing industries reflect the same basic dynamics.

 

The Auto Industry, Transnational Production, and Exploitation

Text Box: Figure 6.10. A Maquiladora Factory in Mexico (2007)

 
Source: Guildhammer • The copyright holder has released this image into the public domain.
            At the beginning of the chapter, I suggested that the increasing integration of the global economy, especially the linking of developed and developing world economies, may have made exploitation less serious today than in the past. On the surface, there seems to be some support for this view, as living standards in many of the more globally (or regionally) integrated developing economies—e.g., Mexico, China, Thailand, Poland, Brazil—seem to be rising. At the same time, some scholars argue that the expansion and deepening of transnational production is simply a repackaging of the same exploitative practices that have been going on for centuries under capitalism. Richard Vogel (2007), for example, asserts that the rise of transnational or global production reflects, at base, modern “capitalism’s relentless demand for cheap labor” (n.p.). In Vogel’s view, the globalization of production has created (or re-created) a hierarchical system of labor in which workers are divided, both domestically and internationally, into production tiers “that are paid grossly unequal wages and receive widely disparate employment benefits.” The ultimate goal or function of these global production chains “is to establish and maintain the lowest possible aggregate labor costs in order to maximize profits” (n.p.). This is not much different from previous eras, except that even workers in the core economies suffer from increasing exploitation. In the North American production system, for instance, midwestern autoworkers (a shrinking part of the overall workforce) in “Tier 1” jobs (these are jobs for the original equipment manufacturers [OEMs] in the main assembly plants) enjoy top wages, averaging about $26 an hour in 2007. U.S. autoworkers in the southern states, however, receive only about half that amount—$13.25 per hour on average. Tier 1 Mexican assembly workers, of course, are at the bottom of this scale, earning about $3.25 an hour, a trifling 13 percent of the wages of their counterparts in the U.S. Midwest (all statistics cited in Vogel 2007).

Tier 2 employees—those working for subsidiaries or primary contractors—earn much less, and have fewer benefits. Again, wages are generally based on location, with U.S. Tier 2 employees, most located in southern states, earning about $11 an hour, and Mexican workers earning an average of $1.75. According to Vogel, however, there are relatively few U.S.-based Tier 2 employees, since the proximity of much cheaper—i.e., much more exploitable—labor in Mexico encourages locating the production of most parts and components across the border. Although Vogel does not provide figures for Tier 3 workers—these are the people who clean and maintain office and production facilities, work in cafeterias, laundries, etc.—their wages are typically set at the minimum-wage level: in 2013, the minimum wage in Mexico was approximately $0.58 an hour. It is no accident, too, that many Tier 3 workers are undocumented immigrants who, because of their status, can be paid sub-minimum wages (with no benefits at all) in developed-country economies.

Text Box: Figure 6.11. Median Annual Earnings for U.S. Males
 
This graph shows a general decline in real wages for U.S. males; the upper line shows a relatively modest decline between 1964 and 2009 for full-time male workers. Significantly, though, only 66 percent of men held full-time jobs in 2009 compared to 80 percent in 1970. Taking both part-time and full-time jobs into account, the (median) real annual wage declined by $13,000 (28 percent) over the four decades between 1969 (the peak year) and 2009. 
Source: Hamilton Project at the Brookings Institution. Original chart at http://www.hamiltonproject.org/multimedia/charts/median_
annual_earnings_since_1964/
It is important to re-emphasize that this exploitative production system is part and parcel of a larger political process, one that is dominated by states and TNCs. The exploitation of Mexican workers, in particular, can be traced to the opening of the maquiladora program in 1964, which allowed foreign companies to set up and operate factories in Mexico free of duties, taxes, and other custom fees. NAFTA opened the door further to FDI, but made sure that workers’ labor (and other) costs would remain extremely low. The Mexican government, in particular, “does all it can to ensure that workers don’t unionize, or if they do that they join so-called ‘protection unions’ designed to assure the interests of plant owners and [to] keep wages low” (Johnson 2012, n.p). As a result, many if not most maquiladora workers make just enough to survive—typically no more than $7 to $9 a day. It would, of course, benefit the Mexican economy if the workers were paid more. If they could earn middle-class wages, their consumption of goods would increase, which would provide a significant boost to the local economy. But if workers make too much trouble, replacement workers from Mexico’s rural areas—where job prospects are even bleaker—can easily be brought in to replace them.

The Mexican government, however, is caught in a vise just as the workers are; if the government makes trouble by advocating for higher wages (for example, by imposing a minimum wage), demanding higher taxes and fees, or imposing regulations, the TNCs will threaten to move to another low-cost, less troublesome location. U.S. auto companies, in turn, have used their increasing reliance on the Mexican labor force to weaken, even “decimate” (as Vogel puts it), the unions in the United States. The United Auto Workers (UAW) union, in particular, is a shell of its former self: at its peak, it had 1.5 million members, but in 2012, membership stood at just 383,513 (UAW 2013). This has meant that wages even for Tier 1 employees in the Midwest have declined over the years. According to an analysis by Abby Ferla (2011), the entry-level wage (adjusted to 2011 dollars) for a UAW worker in 1961 was $18.97; by 1970, this had increased to $23.58. By 2007, however, the entry-level wage had dropped to $15.25. For longer-term employees, the situation was the same: the “maximum attainable rate” dropped from a high of $30.64 an hour in 1970 to $19.28 in 2011 (for workers hired after 2007). The decline in manufacturing wages in the auto industry, it should also be noted, cannot be disconnected from wages more generally, both in manufacturing and service-sector jobs. Critics of globalization (and transnational production, more specifically), point out, that real wages—even in the United States—have been on a steady decline over the past four decades or so. In one study by the Hamilton Project at the Brookings Institution (2011), the median income for all male workers in the U.S. declined by 28 percent between 1969 and 2009—the equivalent of a $13,000 drop in annual wages (see figure 6.11, “Median Annual Earnings for U.S. Males,” for more details).

The issue of whether transnational production exacerbates exploitation—about which there remains much vehement disagreement—raises a related and equally important question: has the globalization and increasing integration of the world economy created the basis for stronger, more dynamic economic growth in developing countries? Or is transnational production simply another mode of solidifying the division between the developed and developing world? It is to this question that we turn next.

 

Transnational Production, FDI, and Economic Development

            The relationship among transnational production, FDI, and economic development (in poor countries) is often oversimplified. On one side are advocates of neoliberal economic theory, who argue, quite assertively and unequivocally, that transnational production and FDI are almost entirely forces for economic progress. Consider the following statement by Anabel González, writing for the World Economic Forum and Global Agenda Council on Global Trade and FDI: “FDI is a powerful instrument for growth and development. Its relevance is enhanced today by its role as the crucial engine of growth, via global value chains, and by the critical need to increase investment flows to boost the global economy, create jobs, and promote knowledge and productivity enhancements” (p. 10). On the other side are writers such as Richard Vogel, whose work I discussed in the preceding section on exploitation and the globalization of the North American auto industry. Vogel, of course, is not alone. Critics of globalization and FDI argue, first, that the bulk of FDI does not flow to poor countries in the first place, but tends to concentrate in already wealthy economies. And, second, for poor or developing countries that do receive substantial FDI, their economies become seriously distorted—for example, they are made heavily reliant on external demand—and overly dependent on foreign companies that not only can leave at a moment’s notice, but that also tend to repatriate earnings back to their home countries. The result is precious little development (for an example of this type of argument, see Hart-Landsberg 2006).

            Looking around the world, it is almost assuredly the case that the truth lies somewhere in between these very general assessments. In Mexico, for example, the results are decidedly mixed: while FDI has created jobs and contributed to Mexico’s overall growth, it has not led to a turnaround for the country’s poor. This is reflected in the still extremely low wage levels in maquiladora factories and the largely unabated inflow of undocumented Mexican workers into the United States (which only slowed down after the collapse of the housing bubble in 2007—but only because there were fewer Tier 3–type jobs in the U.S.). Indeed, decades after the beginning of the maquiladora program and almost two decades after the implementation of NAFTA, Mexico continues to suffer from a very high level of poverty, with more than half the population (52.1 percent in 2012) living below the poverty line, a figure that is roughly the same as it was in 1992 (cited in Wilson and Silva 2013, p. 3). Similar stories can be found in other regions of the world. In Africa, a study by UNCTAD indicated that, with the exception of a few countries (Mauritius, Senegal, and Zimbabwe), the relationship between FDI and economic growth was either very weak or nonexistent (2005, p. 25). And while UNCTAD believes that FDI can play a constructive role on the continent, its overall conclusion is not favorable: “FDI seems to have reinforced a pattern of adjustment that privileges external integration [i.e., integration with world markets] at the expense of internal integration [i.e., development of strong linkages within domestic economies], typified by the establishment of enclave economies” (2005, p. 82). In Central and Eastern Europe and Asia, most studies have shown a generally positive relationship between FDI and economic growth, but there is clear evidence that the benefits from FDI, according to Hanson (2001) and others, tend “to be quite sensitive to host-country characteristics” (p. 23). The last part of the previous sentence is key. The effectiveness of FDI, to put it in slightly different terms, depends a great deal on the host country itself. However, it is not simply a matter of the host country having site-specific advantages, such as geographic proximity, cultural and linguistic affinity, a skilled and educated workforce, access to important natural resources, etc. These factors are certainly important in attracting FDI, but they have relatively little to do with the impact—positive or negative—that FDI will have on the host country as a whole.

            In determining the impact FDI (and the concomitant integration into a transnational production system) will have on the host economy, the neo-mercantilist (or statist) position may offer the best answer: much depends on political factors. Does the host country have sufficient leverage to ensure that FDI is used to benefit the domestic economy? Does the host country have the capacity to effectively make and implement public policies? Equally, does it have the capacity to mediate effectively between domestic firms and TNCs—or sometimes, to deal directly with TNCs—to wring the maximum benefits out of FDI? The relationship between a host country’s government and its own society is important as well. If state actors have little accountability and are generally unconstrained by domestic political arrangements—as often happens in authoritarian political systems—the benefits from FDI may accrue only narrowly to the economic and political elite. In short, when thinking about the impact of FDI it is important to keep firmly in mind that the goals and priorities of TNCs, governments, and societies are not only different, but also sometimes contradictory. In this view, the argument is straightforward: the only countries that will likely see a significant and broadly positive impact from FDI are those in which the state (1) has adequate leverage and capacity vis-à-vis TNCs, and (2) is focused on promoting national economic development—in a strategic and systematic manner—and enhancing the general welfare of its citizens. In this regard, UNCTAD (2005) asserts that the East Asian countries, especially South Korea and Taiwan (China should also be added to this list), offer the best examples.

            In East Asia, various policies were “employed to link FDI to a wider national development strategy, particularly in relation to upgrading and exporting; thus, in addition to clear ownership rights, guarantees against expropriation, EPZs (export processing zones) and fiscal incentives, such measures included reverse engineering of imported goods, technology screening, performance criteria, domestic content agreements, prohibited entry into infant sectors, and exchange controls” (UNCTAD 2005, pp. 55–56). Crucially, though, the UNCTAD report also acknowledges that “[s]trong and capable states are needed to bargain effectively with large firms” and with other interest groups, both domestic and foreign (p. 58). None of this is easy. Indeed, from the neo-mercantilist viewpoint, the primary problem is that most developing countries suffer from a serious lack of bargaining power vis-à-vis large TNCs. The reason is clear: in an era of globalized or transnational production, TNCs have an increasing capacity to exercise regulatory and labor arbitrage. Unless developing countries have something particularly valuable or unusual to offer, the ability of TNCs to locate production wherever the “best deal” is means that most developing countries have precious little leverage, much less power, of their own. (This can be partially mitigated through regional trade agreements, such as NAFTA or MERCOSUR, but even here state power is in play.) To better see the significance of state leverage and power, it would be useful to consider a specific case. One particularly good case to focus on, as suggested above, is China.

 

China’s Rapid Economic Rise and FDI

            The story of China’s economic rise, especially over the past two decades, is a complicated one—too complicated to cover in detail here. Thus, this section will examine, in a purposefully and extremely stylized manner, how China has dealt with FDI and transnational production, and why China has been successful in ensuring that FDI contributes significantly to the country’s economic growth and development. To begin, it is important to recall (from chapter 2) that China is still governed by the Chinese Communist Party (CCP), a highly organized, strongly interventionist political party that dominates all aspects of the Chinese state; the domination is so deep that China is said to be a party-state. Until 1978, the CCP presided over a centrally planned, command economy—the antithesis of a free market economy. Since then, however, the country has made a transition to a market economy, and the results have been stunning. In fact, since the transition, China has been one of the most dynamic and fastest-growing capitalist economies in


Table 6.5. China’s Average Annual Real GDP Growth Rates: 1960–2012

Time Period

Annual Growth
Rate (% per year)

1960–1978 (pre-reform)

5.3

1979–2008 (post-reform)

9.9

1990

3.8

1991

9.3

1992

14.2

1993

14.0

1994

13.1

1995

10.9

1996

10.0

1997

9.3

1998

7.8

1999

7.6

2000

8.4

2001

8.3

2003

9.1

2004

10.1

2005

9.9

2006

11.1

2007

13.0

2008

9.0

2009

9.1

2010

10.3

2011

9.2

2012

7.8

 

 

Sources: For the years prior to 2009, Morrison (2009, pp. 3–4); for the years 2009–2012, CIA World Factbook (https://www.cia.gov/library/publications/the-world-factbook/)

 

the world. In 2012, China had the second largest economy in the world with a (nominal) GDP of $8.2 trillion (in PPP terms, the CIA [2013] estimates China’s GDP at $12.6 trillion). Although still relatively poor in per capita GDP terms—China is ranked about 92nd in the world—the country’s economic growth rates, combined with relatively low fertility rates, suggest that it will move up quickly. Since 1979, China has grown at an unprecedented pace, averaging 9.9 percent between 1979 and 2008. Even during the depths of the global recession, from 2009 to 2011, China continued with high growth rates at an average of 9.5 percent per year, although in 2012, growth slowed to 7.8 percent (see table 6.5, “China’s Average Annual Real GDP Growth Rates”). For much of the rest of the world, by contrast, the period between 2009 and 2012 was marked by either negative growth or very low growth, with 2009 being a particularly bad year.


There is, it is important to re-emphasize, almost nothing laissez-faire about the Chinese party-state. It is deeply and pervasively interventionist. To the extent that China has freely operating markets, one can say with only slight exaggeration, it is because the state explicitly encourages and permits such activity. It is no surprise then that the party-state has played a key role managing FDI and China’s integration into global production systems. Particular attention has been paid to ensuring that FDI improves the competitive capacity of Chinese firms in areas that Chinese leaders consider to be key industrial sectors—which typically include one or several state-owned enterprises (or SOEs), discussed in more detail below. For a long time, for example, the Chinese state has published an “investment catalogue,” which lists specific areas in which FDI is “encouraged,” “accepted,” and “discouraged.” Investments in the first category have access to a range of preferences: tax subsidies, preferential access to land and labor, a simplified regulatory process, and so on. The offer of preferences, however, is generally contingent on the foreign firm’s willingness to transfer technology to Chinese firms. In the high-speed rail sector, for example, foreign firms were restricted to joint ventures and, as a condition of their investment, required to transfer important technology to their Chinese partners. The result? Chinese firms quickly absorbed the technology and then proceeded to compete directly against American, European, and Japanese firms for contracts outside of China (Sally 2011, p. 13). Another salient example is the steel industry. The Chinese government imposes strict guidelines on acceptable foreign investment. China’s “Iron and Steel Industry Development Policy,” for instance, lists the following criteria for foreign investors: “[they] must possess iron and steel technology with independent property rights and should have produced at least 10 millions tons of carbon steel or a least 1 millions tons of high-alloyed special steel in the previous year” (cited in Heiden 2011, p. 20). As in the high-speed rail sector, the Chinese state (through the National Development and Reform Commission) requires joint ventures in which the Chinese partner maintains at least 50 percent ownership in the firm. The state’s involvement, it is important to note, goes well beyond the issue of foreign investment: it also manipulates the prices of vital inputs; imposes export restrictions on important raw materials (such as coke) and semi-finished products; selectively promotes exports of high-value-added, technology-intensive products; and subsidizes outward investment by Chinese firms (Heiden 2011). The goal, of course, is to make China into the largest and most profitable steel producer in the world. The first part of this goal was achieved in 1996, when China surpassed Japan as the world’s leading steel producer; ten years later, China also surpassed Japan as the largest exporter.

There are many other aspects of China’s policy toward FDI—again, too many to cover here. Suffice it to say, then, that the Chinese state has been very successful in not only encouraging technology transfer, but also technology absorption. In the process, many Chinese firms have moved from “junior partner” to “senior partner” status in relatively short order. Of course, this is not uniformly the case throughout the Chinese economy, but the success stories are fairly common. Another reason for this is China’s policy of developing “national champions”; that is, very large companies—often SOEs (see figure 6.12, “State-Owned Enterprises in China,” for further discussion)—that have preferential access to bank capital, as well as to FDI. China’s success raises an important question: Is China’s experience with FDI easily copied? That is, can other states do what China has done? The short answer is—it depends. China is unusual; that is, it is not like the vast majority of other developing countries. It has something that TNCs need: vast, still-underexploited, and extremely valuable markets, both for labor and consumption. China’s exceptionalism gives the country’s state leverage and power that most other developing countries lack. It is partly, perhaps largely, for this reason that China can extract maximum benefits from FDI, and, to some extent, even dictate terms to the largest, most economically powerful TNCs. On this point, it is useful to consider the neo-Marxist perspective.

 

 


Figure 6.12. State-Owned Enterprises in China


State-owned enterprises, or SOEs, “are business entities established by central and local governments, and whose supervisory officials are from the government” (Szamosszegi and Kyle 2011, p. 6). SOEs, however, are only one of a number of types of state-controlled enterprises (SCEs). Thus, in addition to SOEs, there are joint-operation enterprises, limited-liability corporations, shareholding corporations (with the state owning the majority of shares), and public organizations. In 2010, there were 9,105 SOEs in China, and another 11,405 enterprises in which the state held a controlling share of the company (cited in Szamosszegi and Kyle 2011, p. 8). Despite the relatively small number of SOEs (and other state-controlled companies), their influence in the national economy remains significant. Accurate figures, unfortunately, are not available, but various analyses have put SOEs’ share of China’s GDP at between 30 percent and 50 percent. In addition, SOEs and SCEs account for as much as 48 percent of urban employment, and 54 percent of total wages paid to urban employees.

 

Equally if not more important, the Chinese state has designated defense, electric power, petroleum and petrochemical, telecommunications, coal, civil aviation, and shipping to be strategic industries, and equipment manufacturing, automobiles, IT, construction, iron and steel, nonferrous metals, chemicals, and surveying to be pillar industries. In strategic industries, the state has declared that SOEs or SCEs must maintain either sole ownership or absolute control, while in pillar industries a “strong control position” is required. All SOEs, but particularly those in strategic and pillar industries, receive preferential treatment from state-owned banks. For example, they have access to capital and favorable interest rates, or, if they are unable to repay their loans, their debts may be forgiven. In addition, some uncreditworthy SOEs are extended loans. (The statistics and other information cited here come from Szamosszegi and Kyle [2011].)

 

The greatest benefits are provided to the so-called national champions—that is, firms that are among China’s largest SOEs. These include:

§  China National Petroleum

§  Sinopec

§  China National Offshore Oil Company

§  Aluminum Corporation of China

§  China Minmetals

§  China State Construction and Engineering Corp.

§  China Ocean Shipping Group (COSCO)

§  China Communications Construction

§  ZTE Corp (telecommunications)

§  Lenovo (IT)

§  Haier (consumer goods)

§  CITIC

§  China Investment Corporation (a sovereign wealth fund)

In addition to their access to low-cost loans, the Chinese state helps to ensure the success of national champions by making sure they have access to cutting-edge, often proprietary technology. For instance, in return for access to the Chinese market, the Chinese government generally requires foreign firms not only to enter into joint ventures with Chinese manufacturers, but also to provide proprietary technology transfer—i.e., patents and trade secrets. For further discussion, see Hemphill and White III (2013).
 

 

In the neo-Marxist view, it is well understood that capitalism, to survive and prosper as a system, needs space for constant expansion. It is for this reason that China’s 1.3-billion-person economy has long been looked upon as a necessary part of global capitalism. Under strict communist rule, however, Chinese markets were closed off to the capitalist world. Of course, this did not last. Thus, when China began its transition to a market economy, TNCs were more than ready to take advantage. In the first decade or so, not surprisingly, there was some trepidation, but by 1992, once it had become clear that the Chinese leadership were thoroughly committed to the reform process, inward FDI began to ramp up. This is evident in the statistics. From 1982 to 1991, the net annual inflow of FDI remained relatively low, growing from just $430 million to $4.366 billion. In 1992, by contrast, FDI shot up to $11.15 billion, and then more than doubled to $27.5 billion in 1993. Between 1993 and 2002, FDI averaged almost $40 billion a year compared to the $2.26 per year average between 1982 and 1991 (all figures cited at http://data.worldbank.org/). By the late 1990s, China had become the second largest destination for FDI in the world, behind only the United States; in 2002, China (temporarily) passed the U.S., and since then the two countries have been neck and neck. Among late-industrializing countries, though, China has been, by far, the largest recipient of FDI. The most impressive growth began in 2005: that year alone, China attracted $117.2 billion in FDI, breaking the $100-billion mark for the first time (OECD 2012). In 2011, China broke the $200-billion mark with a total of $280 billion—about $23 billion more than the United States.[37]

The huge inflow of FDI to China reflects the importance of China to global capitalism. While it is certainly true that a great deal of FDI is meant to take advantage of low labor costs in China, it is also clear that TNCs are motivated by gaining a strong foothold inside China’s growing consumer market. China’s middle class plays a particularly important role in this regard, since it will be the main driver of increased and sustained consumption in the coming years. Consider, on this point, an analysis by Barton, Chen, and Jin (2013) of McKinsey and Company. They point out that China’s middle class (which is defined as households with income between $9,000 and $34,000 a year) has grown from just 4 percent of the urban household population[38] in 2000 to 68 percent in 2012; they project that this will increase to 75 percent in 2022—this is equivalent to 630 million consumers. Upper-middle-class households, in particular, are “poised to become the principal engine of consumer spending over the next decade,” both for China and, to a significant extent, the entire world. Barton (in a separate article) estimates that, in 2022, China’s middle class will be consuming goods and services valued at $3.4 trillion. “This,” he tells us, “will have enormous significance for U.S. businesses” and, by extension, for businesses from every core economy (2013, n.p.). Even now (in 2013), China is becoming a major market for relatively high-priced consumer goods: a case in point is Apple’s iPhone 5S, which broke sales records when it was released in September 2013. A big reason for the success of the 5S, according to industry analysts, was the Chinese market (Greenfield 2013). TNCs cannot afford to not have a foothold in the Chinese economy, and this is what gives the Chinese party-state tremendous leverage and power: it remains, for the time being, the principal gatekeeper into (as well as out of) China. While the Chinese economy is a market-based capitalist economy, it is decidedly not a free market.

To sum up: in assessing the impact of FDI, it is critical to consider, first and foremost, the power of the host country state in the global economy. Not all states (especially in the developing world), of course, are equally empowered, nor is the source of power always the same. For this reason, China is not alone in exercising influence over TNCs and FDI. At the same time, there are few developing countries in the same position as China (possible candidates might include India and Brazil); nor are there many states that have the internal capacity of the Chinese party-state (a possible candidate is Russia). This explains why China has been able to use FDI to such great advantage, and even become an economic juggernaut. 

            China’s economic ascendance, however, has not been all wine and roses. As in Mexico, integration into a global system of production has also meant integration into a highly exploitative global division of labor—a point also emphasized by neo-Marxist scholars. On this point, it is important to recognize that inequality in China has “increased steadily and inexorably” since the early 1980s (Naughton 2007, p. 217). The country’s Gini coefficient—a scale on which zero is perfect equality and 1.0 is perfect inequality—increased from 0.28 in 1983 to 0.447 in 2001. This is an unprecedented deterioration, and one that has turned a country that was once one of the most equalitarian in the world into one that is “now similar to the most unequal Asian developing countries, such as Thailand, 0.43, or the Philippines, 0.46” (Naughton 2007, p. 218). Thus, while it is true that a new, relatively prosperous middle class in China has emerged and is growing—along with the rise of a class of economic elite—a huge and almost assuredly permanent underclass of hyperexploited, low-skilled workers has also been created. Other analysts argue, however, that the picture is not quite so neat. The World Bank (n.d.) notes that China has made remarkable progress in reducing severe poverty within the country: since 1978, more than 500 million Chinese citizens have been lifted out of poverty, and the poverty rate has fallen from 84 percent in 1971 to a scant 13 percent in 2008 (as measured by the percentage of people living on the equivalent of U.S. $1.25 or less per day in PPP terms). Such results cannot be dismissed as unimportant; instead, they reflect a sea change, not only for China, but for the world as a whole, as half-a-billion people represents about 7.5 percent of the entire world population. Also, do not forget that China’s burgeoning middle class will comprise at least 45 percent of China’s population in 2022. The upshot is that there is no simple, black-and-white answer to the broader question: Does transnational production and FDI help the poor and less privileged segments of society? However, examining the question from a variety of theoretical perspectives will help you develop a better, more critical understanding of the issue.

 

Transnational Production and State-Firm Interactions

            The seeming strength of the Chinese state helps underscore a basic element of the global political economy: despite the increasing importance of TNCs and of the global production (and financial) structures, it is fairly evident that states still matter a great deal. This stands in sharp contrast with a common narrative of the past twenty plus years, which is that globalization has made the state increasingly irrelevant. The debate over the relevance of the state in an era of globalization, however, has not been terribly productive. This is true largely because the wrong question is being debated. That is, the question should not be, does the state still matter? Instead, the question should be, how has globalization (especially transnational production) changed the character of the state and the dynamics of its interactions with TNCs and other transnational actors? The discussion in the preceding sections also compels us to stop treating states as generic entities. There are, it is important to understand, a variety of states, with (1) widely varying degrees of internal capacity, competence, and coherence (political, economic, and military); (2) different policy interests, preferences, and choices; (3) divergent orientations and attitudes towards FDI, the market, and the world economy; (4) different political-regime types; (5) different levels of integration in the global economy and different levels of socioeconomic development; and so on. The list is quite long. All of these differences shape and even determine how states respond to globalization, and how effective their individual responses can be. To be sure, states share important characteristics, too, but their differences from each other can be, and often are, profound.

The case of China, for instance, represents a state with a high degree of internal capacity, competence,