Race for the Exits: The Unraveling of Japan's System of Social Protection.

Witt, Michael A.

Race for the Exits: The Unraveling of Japan's System of Social Protection. By Leonard J. Schoppa. Ithaca: Cornell University Press, 2006. 247 pp. $65.00 (cloth).

Race for the Exits is a book about structural inertia. Drawing on Albert Hirschman's dichotomy of exit versus voice, it argues that Japan has seen limited institutional change in key areas of the political economy because it has experienced neither enough voice to push through political reforms nor enough exit to shock the political system into reforms. Instead, moderate levels of exit involving the most disaffected actors have reduced pressure for change. The book adduces supporting empirical evidence from two constituencies taken to be bearing a heavy burden under existing institutions: large firms and women.

Race for the Exits presents its argument in nine chapters. After an introduction containing an overview of argument and contents of the book, Chapter 2 lays the theoretical foundations of the work. Chapter 3 reviews the characteristics of Japanese "convoy capitalism." Chapter 4 describes recent socioeconomic changes and attendant exit responses by firms and women. Chapters 5 through 7 explore policy responses to these exit responses, highlighting how the moderate nature of the exit responses has hampered policy changes. Chapter 8 argues that change is possible if actors choose to exert voice because they cannot exit, or if actors can exit rapidly and at low cost. Chapter 9 presents a summary and scenarios for change in Japan.

The book thus seems to make the opposite argument of what one would expect from the title. The evidence introduced suggests that in the areas that represent the empirical core of the book, there is no "race" to the exits--an expression that evokes images of large numbers of defections. On the contrary, it is for lack of a race for the exits that the Japanese system of social protection has not undergone more radical change.

The main conceptual contribution of this book lies in a point already developed by Hirschman but subsequently largely forgotten in the literature: where the cost of exit makes it a choice available only to a relatively small number of actors, exit can reduce overall pressure for institutional change. This point is well taken and is a good departure point for further theory building. Future research would seek to draw and build on the wide literature on the topic that has emerged since Hirschman published his classic almost forty years ago, especially in studies of management and sociological institutionalism. This literature has evolved a picture that is more fine-grained in that it encompasses a wider range of behavioral responses and presents predictions concerning their impact on institutional change.

Applying finer-grained theory to empirical phenomena such as those presented in this book may help reduce the strain on the construct of "exit," which under the given model accommodates a wide range of behavioral responses. These include disparate behaviors such as firms relocating plants to foreign countries; women deciding not to marry or have children in order to pursue a career; women quitting the labor force to take care of their children; or large firms negotiating private electricity tariffs with utilities. The common denominator of these actions, as the book notes, is their "uncoordinated individual" (p. 7) nature. But do these disparate forms of responses affect pressure for change in the same way? For instance, while reluctance to get married may well, over the long term, build pressure for change because its effects are incompatible with existing formal institutions, the same seems unlikely for women quitting their jobs, which is consistent with those institutions. Exploring these issues would be a worthwhile undertaking in the further evolution of this stream of research.

In this context, future works might study in depth an interesting variation in outcomes between the two main case studies: While firms tried to exercise at least some voice, young women are "conspicuous for their absence from the contemporary women's movement" (p. 160). This suggests a combination of no voice and limited exit, something that Hirschman's model as presented in the first chapter does not seem to accommodate. Surely there are Japanese women in their early twenties with a strong preference for both career and family. What keeps them from organizing, and what revisions, if any, does this require of the underlying theory?

Further research may also explore in detail other variables that can affect the political dynamics noted. For instance, how does the complexity of changes required compare across the case studies offered in the book? The two cases of successful reforms--long-term care insurance and financial sector reforms--may be relatively less complex and thus comparatively easy to design and execute. By contrast, both of the main cases would seem to require far-reaching, systemwide reforms with linkages across many issue areas. If true, this may affect the pace of reforms.

A quibble is that in some cases one would like to see a stronger exposition of how the evidence supports the point being made. For example, Table 4.1 shows evidence of high costs in Japan. However, the underlying data come from a period of endaka in Japan's real effective exchange rate, so part of the effect is likely to be due to macroeconomic reasons. Taking this into account, how much of the effect remains? Similarly, of the three industries to have shown the greatest propensity to relocate abroad because of high cost (p. 85), two--textiles and farm products--are declining industries in Japan, so one would expect them to disappear from there no matter what institutional reforms are introduced. Electronics, while not declining, is sensitive to labor costs in assembly, which is why most electronics firms in the world outsource assembly to China. Taking these considerations into account, how much of the posited structural effect remains?

Overall, Race for the Exits speaks to an important issue. It contributes interesting insights, but in particular, it provides a good foundation for further theoretical and empirical exploration.

Michael A. Witt

Economics and Political Science Area

INSEAD