In the last chapter, we offer what we believe is a new hypothesis for the industrialization of South Korea and Taiwan: A primary force driving not only the emergence, but also the divergence of these two economies is the development of intermediary demand growing out of the U.S. retail revolution, to which Korean and Taiwanese firms were able to respond successfully, but in different ways. This claim will certainly be controversial because the existing explanations for East Asian industrialization consistently ignore factors relating to demand. To be more specific, existing explanations do not consider the interconnectedness of global markets in explaining industrialization. They do not consider how changes in the growth and organization of one economy, the U.S. economy, can shape the growth and organization of other economies, the Asian economies. The possibility of this proposition is implicit in our revised Walrasian perspective that we outlined in Chapter 1. Insofar as markets are interconnected, then a change in one market should result in changes in other markets. Within this perspective, theoretically speaking, markets should not be confined artificially to only markets within a national economy. Rather, the test of the extent of interconnectedness is empirical: Are markets actually linked in fact, and if so, with what consequence for all markets so linked?

We demonstrated in the last chapter that changes in the organization of retailing in the United States is directly related to changes in the composition of international trade between the United States, on the one hand, and South Korea and Taiwan, on the other.

In her book, Regional Advantage, AnnaLee Saxenian (1994) makes an argument for “two models of industrial systems – the decentralized regional network-based system and the independent firm-based system.” The two models come out of her case studies of two regional economies in the United States, one centered on high technology industries along Route 128 in Massachusetts and the other centered on high technology industries in Silicon Valley in Northern California. An astute ethnographer and not a formal modeler, Saxenian uses these case studies inductively to make a theoretical point: Differences in economic organization create decisive differences in the course of economic development.

Generalizations reached from two contrasting ethnographic studies do not provide much evidence for a theory, but anecdotally the cases are very persuasive. In the 1960s, the East Coast region with its mammoth, vertically integrated corporations, such as IBM, RCA, Digital Equipment Corporation, Wang, Honeywell, and General Electric, pioneered the move into high technology industries and quickly built a formidable technological and corporate advantage over firms elsewhere in the rest of the country. “By 1975,” says Saxenian (1994, p. 19), “the technology complex along Route 128 employed close to 100,000 workers and was poised for a decade of explosive growth.”

On the other side of the country, in the area around Stanford University south of San Francisco, the 1960s had also been a decade of expansion into high technology industries. Unlike the East Coast mega-corporations, the firms in the Silicon Valley were mainly small and narrowly focused.

Local Economies in an Age of Global Capitalism

In his new and very thoughtful book, Principles of Economic Sociology, Richard Swedberg (2003, p. 54, his emphasis) makes the following point: “It seems clear that economic sociology should set capitalism at the very center of its analysis since this is the dominant way of organizing the economy – legally, politically, and socially – in today's world.” Although the point may be obvious, Swedberg (2003, pp. 63–5) notes that very few sociologists actually build an economic sociology on this fact:

Although Swedberg singles out economic sociologists, we would add that many economists are guilty of the same oversight: The narrowed concentration on the firm and equating the firm with economic organization in general obscures the very workings of capitalism.

This book began as a study of the business groups in South Korea and Taiwan, but has grown into something much more. Business groups – affiliations of firms, usually with some degree of common ownership – have been a favorite topic of study among a number of economists (who have had a principal interest in the keiretsu in Japan, but also the groups found elsewhere in Asia) and economic sociologists (including one of the authors), as well as political scientists and area specialists. In economics, the traditional explanation for these groups has been that they are a response to market failure; because the market for capital or entrepreneurial skill or some other asset does not function well within the economy at large, business groups allocate this scarce resource among affiliated firms, thereby substituting managerial initiative for market mechanisms. In political science, rather than being a function of market processes, these groups are explained as being the creation of government mandates, expressed by preferential policies toward business groups and the entrepreneurs who establish them. In sociology, the explanations also downplay purely market processes, but make these groups the outcome of background institutional environments in which political and social institutions place parameters on how economies operate.

On the surface, these various explanations have little in common. Obviously, they are all shaped by the disciplinary gaze of the analysts and the countries they observe.

Why do business groups in South Korea and Taiwan, as well as the organization of economies in which they are embedded, differ from each other? Why should Korea have moved along a path toward an economy where so much production, especially in the export sectors, is concentrated in only a small number of vertically integrated chaebol? Why should Taiwan have taken a very different path, one where business groups firms assemble parts made by unaffiliated firms or where they occupy upstream and service niches in an economy pulled along by downstream demand, mainly from small and medium-sized firms? In the language of our model, why was the “high concentration” equilibrium set selected in Korea and the “low concentration” set selected in Taiwan? The answers to these questions are the subject of Part II.

In the opening chapters of this book, we covered a number of answers that various writers have given to explain the presence of business groups, rather than the differences among them. These answers usually represent one or a combination of either strong-state, market-failure, or institutional explanations. Although we do not deny the importance of political policies or of problems with capital markets in developing economies, we believe that such factors should be seen as proximate causes, as occurrences or events that trigger economic changes, often extremely important ones, without actually dictating the organizational medium of that change.

“In many successful late-industrializing countries in the 20th century … business groups with operating units in technologically unrelated industries have acted as the microeconomic agent of industrial growth” (Amsden and Hikino, 1994, p. 112). This is Alice Amsden's conclusion, presented here in an article with Takashi Hikino, but also found in many of her other writings (2001, 1989, 1985). She shares this opinion with many political economists (Cumings, 1985, Wade, 1990, Fields, 1995, Evans, 1995, Woo-Cumings, 1991, 1999), who, like herself, have been instrumental in making the strong state interpretation the most prominent explanation of Asian development. A number of economists (Leff, 1976, 1978, Aoki, 1984, 1988, Jorgensen, Hafsi, and Kiggundu, 1986, Ghemawat and Kanna, 1998, Khanna and Palepu, 1999, 2000a, b, c), while willing to concede the potential role of government policy, instead list market failures and transaction cost savings as the primary reasons for the emergence of business groups. Not to be outdone, sociologists (Granovetter, forthcoming, 1995b, 1994, Chung, 2001, 2003, Whitley, 1990, 1992, Gereffi and Wyman, 1990, Orrù, Biggart, and Hamilton, 1997, Hamilton, Zeile, and Kim, 1989) see business groups as embodiments of social networks and institutions rather than as outcomes of political or economic processes. That scholars from such diverse disciplines have zeroed in on business groups from different disciplinary angles testifies not only to their actual importance in Asian economic development, but also to their theoretical ambiguity.

The model of business groups described in Chapters 3 and 4 is formally developed in Feenstra, Huang, and Hamilton (2003), which is reproduced in part here, along with the mathematical Appendix from that paper.

The business groups model is a natural extension of the monopolistic competition framework used in industrial organization (Dixit and Stiglitz, 1977, Spence, 1976) and international trade (Helpman and Krugman, 1985). In this framework there are large numbers of firms, each producing a unique product. Although it is normally assumed that the firms operate independently, we shall allow groups of firms to jointly maximize profits, and such a group of firms is called a “business group.” Equivalently, we can think of a business group as a multi-product company, that chooses both the range of upstream and downstream goods to produce, and their optimal prices. Helpman and Krugman (1985, pp. 220–2) recognized that the monopolistic competition model had the potential to include economic organization in their discussion of “industrial complexes,” but this idea was not pursued further in the trade context; instead, the upstream and downstream linkages between firms became a building block of the new models in economic geography (Krugman, 1991, 1996). The equilibrium concept we use is closest in spirit to the work in industrial organization by Perry (1989, pp. 229–35), though also anticipated by the early work of Caves (1974).

In Chapter 3, we refer to investigations by the Korea Fair Trade Commission dealing with differential pricing practices within Korean chaebol. Several such cases are summarized subsequently.

Case 1. Goldstar Cable Ltd.

Source: The Korea Fair Trade Commission (1994), “A Case on the Illegal Internal Transactions of Goldstar Cable Ltd. Co.,” Proceedings of Fair Trade Commission Decisions, No. 94-242.

According to the source, Goldstar Cable, an affiliate of Lucky-Goldstar group, favored its affiliates over non-affiliate firms in trading various commodities in 1993. It sold its products to affiliate buyers at much lower prices than to non-affiliate buyers (see Table B.1). It also preferentially treated its affiliates by buying their products at significantly higher prices than from other firms (see Table B.2). Table B.1 shows that Goldstar Cable sold electrical wires to Goldstar, its affiliate, at below-market price by 9.6–29.6 percent; it sold high voltage cables to Kukje Cable, its affiliate, at lower prices than to non-affiliate buyers by 1.1–9.1 percent; it sold electrical chillers to Lucky Engineering, its affiliate, at lower prices than to non-affiliates by 16.7–35.9 percent; it sold electrical connectors to Goldstar, its affiliate, at significantly lower prices than to the other non-affiliate buyer by 21.9–24.8 percent; and it sold coated wires to Kukje Cable, its affiliate, at lower prices than to non-affiliate buyers by 4.5–15.9 percent.

Most theories of economic organization, regardless of discipline, involve a sleight of hand. Theorists begin by assuming the existence of decision-making individuals. They then provide these actors with inner motivations: desire for gain, for power, or for social honor and reputation. Driven by these motivations, economic actors are then set in motion. They plot strategy; they use guile. They act on their interests; they interact in trusting ways. Seeking to maximize, they also respond to incentives and constraints put in place by powerful people, such as state planners or heads of state banking systems or the CEOs of the largest firms. Whatever these actors do and however they respond shape the calculations and subsequent actions of others. Assuming all similarly motivated individuals act more or less alike, economic theorists then posit an orderly, organized economy, conceived, for example, as a capitalist economy composed of independent and competitive firms. When theorized in this fashion, economic organization is pulled, like a rabbit from a hat, out of aggregated individual decisions.

Attempts to induce societal level organization from individual actions are common enough in every social science. In sociology, anthropology, and political science, theorists often, in a single bound, make the same leap from individual behavior to social and political structure. In these disciplines, however, the reverse trick is equally widespread: The inner motivations and actions of individuals are produced, as if by magic, from descriptions of the whole.

If our Walrasian view of economic organization is correct, then business groups can be thought of as causative forces in economic systems that also cause them. Before testing this idea empirically, we needed first to develop a theory of business groups. To this end, in the last chapter, we constructed a highly stylized economic model consisting of upstream sectors producing intermediate inputs and downstream sectors using those inputs to produce final consumer goods. In the model, manufacturing firms decide whether to buy intermediate products at zero markups from a firm within a group, or at full markups from unaffiliated firms. Solving this model based on pricing decisions of firms in general equilibrium reveals multiple equilibria in the form of two distinct and economically stable solutions to business group integration in an organized economy: a high and a low concentration set of groups.

We can think of these two stylized economies as both consisting of interconnected networks of firms where authority and economic power interact in the context of a price structure. In one solution, the few very large business groups organize an economy in which they remain the stable elements, pushing other groups and independent firms into niches that the large groups do not dominate. In the other solution, many business groups coexist by occupying different distinct upstream or downstream niches and by trading with unaffiliated firms. The idealized model, therefore, predicts that business groups, which represent concentrated and authoritatively held assets, organize themselves quite differently across economies.

Using our theoretical model of Chapter 3, we demonstrated in Chapter 4 that a drop in demand can move the economy from one equilibria (that is, structure of the business groups) to another, leading to a string of bankruptcies in the process. We argued that such a string of bankruptcies precipitated the financial crisis in Korea in 1997. In this appendix, we supplement that theoretical demonstration with two empirical arguments related to the financing of the business groups in Korea. While these arguments fall outside the narrow confines of our Walrasian model, they nevertheless offer insight into the sources of the crisis there.

First, we argue that the bankruptcies before November 17, 1997, are predicted well by the excessively high debt/equity ratios of the groups. In contrast, the bankruptcies after November 17 cannot be explained by the overall debt/equity ratios, but rather, by the excessively high levels of short-term debt of these groups. In other words, the bankruptcies before November 17 show every indication that the capital market was working as it should, whereas the bankruptcies after November 17 show the characteristics of a financial panic, in which banks are not willing to roll over short-term loans regardless of the performance of their debtors.

Second, we explain how the interaction between the bankruptcies of chaebol and the precarious structure of the financial system combined to create the financial crisis during the last quarter of 1997.

In his review of the book Trust: The Social Virtues and the Creation of Prosperity, by Francis Fukuyama, Robert Solow makes the following remarkably candid comment:

Unfortunately, the book does not meet Solow's criterion of proof, and he concludes:

To convince economists that organizational structure matters, it is necessary to point to objective measures of economic performance that are affected. We have argued at length that the business groups in South Korea and Taiwan are different, and that these differences are rooted in the exercise of authority and market power in the two countries. We have also shown that the predictions of the authority/market power model – in terms of the size, vertical integration, and horizontal diversification of business groups – match up remarkably well with contrasting structure of the groups in South Korea and Taiwan: this was our first test of the model, in Chapter 4.