Roundtable on Rick Doner, Gregory Noble, and John Ravenhill, The Political Economy of Automotive Industrialization in East Asia

Introduction

Research on the political economy of growth in East Asia from its inception was motivated by a focus on institutions. The early developmental state literature highlighted the role that government intervention played in rapid economic growth, but that literature centered largely on the question of which policies governments chose to implement. From Chalmers Johnson (1982) forward, political scientists have paid greater attention to political institutions—and particularly state institutions—as the enabling conditions for the growth process. Factors such as political autonomy, bureaucratic competence, and coordination with the private sector through business councils were seen as crucial to rapid capital accumulation, a relatively efficient allocation of resources, and corresponding increases in social welfare.

In contrast to this macrolevel focus on national-level institutions, a second strand of the early literature on growth was decidedly micro in its approach, honing in much more closely on economic or what might be called “local” institutions. These institutions served to coordinate or regulate a diverse set of state and private actors, including both foreign and local firms. This literature—which included its share of classics as well—typically took an industry or sectoral focus with particular attention being paid to the relationship between multinational corporations and networks of local suppliers.

This approach to the political economy of the region has continued in various guises, for example in a wide-ranging literature on international production networks. But as editors of the Journal of East Asian Studies, we have seen a decline in this qualitative, institutionally focused case study research on economic growth. Two related considerations thus motivate this roundtable on Rick Doner, Gregory Noble, and John Ravenhill's (2021) book The Political Economy of Automotive Industrialization in East Asia (hereafter DNR). The first is its return to these core micro-institutional questions. But second, the book invites scholars of East Asian political economy to consider how such studies fit with other research approaches to the region's development, including not only the earlier generation of sectoral studies, but also more recent quantitative research and the burgeoning revival of interest in a “new” or “open economy” industrial policy.

The argument in brief

DNR make four core arguments. First, they advocate for a more disaggregated approach to industrialization than is usually found in the development literature, and certainly in the macroeconomic literature on growth per se. They draw a core distinction between “extensive” and “intensive” growth. In the automotive sector, the former consists largely of the development of vehicle and components assembly and, in some cases, exports, controlled by multinational firms operating in global value chains. “Intensive” growth, by contrast, involves an increase in local value added as a result of an expansion in the supply of inputs from efficient indigenous producers with their own distinctive technical capabilities. Whereas economic diversification involving the expansion of auto assembly and exports is the hallmark of successful extensive growth, dense input–output linkages with competitive local producers characterize successful intensive growth. Innovation—typically associated with larger firms—is key to the development of these competitive local producers.

Second, DNR go beyond the developmental state literature in proposing that these different forms of industrial growth require different competencies at the national, sectoral, and even firm levels. These range from the well-known demands of financing capital formation and providing increasingly sophisticated physical infrastructure, but also to the supply of new forms of public goods—or what is often called “quality infrastructure”—such as testing and standards or technical training that expands firm-level capacities. Unlike most of the literature on the developmental state, DNR argue that an understanding of the reasons for the successful pursuit of either extensive or, more particularly, intensive strategies requires an investigation of sector-specific institutions.

Third, developing these various competencies poses institutional challenges, and not only for the state: public, private, and public–private partnerships are all in play. They argue—in a functional way—that capacities either “fit” (or “don't fit”) the demands of different types of growth. More specifically, the book proposes that extensive growth will be associated with institutions focusing largely on mobilizing capital and labor, whereas intensive growth requires institutions that promote the absorption of technologies new to a country's firms, or that can spur technological innovation within these firms.

Fourth, DNR go beyond a purely functional approach to these developmental challenges by addressing the question of institutional origins and the conditions under which such institutions are likely to arise in the first place. In the most explicitly political component of the book, DNR hypothesize that institutions with the capacity to promote intensive growth will emerge to the degree that political leaders are compelled to address external threats and domestic unrest but lack easy access to resources to do so. In such difficult circumstances—in effect, when constrained—leaders will be more likely to make the investment of scarce resources required to improve local industrial capacity through institutional innovation. They show this in particular with their contrast between the political economies of the resource-abundant Southeast Asian countries and their cases in Northeast Asia—China, Korea, and Taiwan—that were characterized by significant external threats and/or resource scarcity.

These arguments are played out over the course of the book. Chapter 2 reviews the lure and challenges of the global auto industry, and then establishes the book's core empirical puzzle: how to explain cross-national variation in response to these conditions within East Asia. DNR address this puzzle through a two-pronged comparative research design. The first, covered in Chapter 3, consists of theoretically informed, cross-case analysis. This “most-similar-different-outcome” design examines variation in the strategies and performance of seven countries pursuing automotive industrialization within the same region and the same producer-driven value chain. It is, of course, difficult to exclude the possibility that other factors, besides or beyond the presence of certain kinds of institutions and pressures on political leaders, account for the superior performance of, say, South Korea, relative to Malaysia. DNR thus turn to within-case analysis in Chapters 4–9. These case studies provide a finer-grained assessment of performance and, most crucially, trace the temporal evolution of policies and institutions relative to performance.

Theoretical touch points

We see several ways in which a renewed focus on sectoral analysis might prove useful for the study of the international and comparative political economy of the region.

Reinvigorating research on industries

DNR build on an earlier, robust body of sectoral research on East Asian growth. Much of this work was on large firms, especially in Japan, Korea, and Taiwan (Amsden 1989; Zysman 1983; Vogel 1996; Woo 1991; Chang 1994; Wade 1990; Mathews and Cho 2000), and others exploring the role of small firm networks especially in Taiwan (Amsden 1985; Kuo 1995). Some of this work was explicitly comparative (Haggard 1990; Weiss and Hobson 1995; Noble 1998; McKendrick, Doner, and Haggard 2000; Wong 2011; Yeung 2016).

Yet, at least in terms of explicitly political economy scholarship, this earlier tradition of industry studies, with the exception of work on China, for example (Thun 2006; Chin 2010; Hsueh 2011), has seen a relative decline in the English-language political economy literature. Several factors contributed to this shift, including the simple fact that the original subjects of the developmental state literature—Japan, Korea, and Taiwan—entered a new phase in their growth in which other political economy topics—from financial crises to inequality and social policy to social media—have received more attention. At the regional level, a focus on international institutions and trade agreements has, with some exceptions (for example, Manger 2014; Ravenhill 2017) been almost entirely divorced from a focus on the interplay between state and business.

But DNR goes beyond this earlier sectoral analysis by virtue of its distinction among stages of industrialization, its emphasis on “goodness of fit” between institutions and development challenges at the sectoral level, and its emphasis on the socio-political underpinnings of institutional capacities. These foci resemble in important ways the varieties of capitalism literature and go well beyond the state autonomy assumptions and what Robert Wade labeled the “thin politics” of much of that earlier literature (Wade 1992, 307).

Quantitative research on East Asian growth

Another possible reason for the decline of such work may be the availability of new statistical data, including on the nature and impact of “governance” (see McGuire 2020; Bernhard et ak, 2015). The value of this work, especially those studies exploring the impact of political institutions on various aspects of development) is indisputable. However, by its very nature, this work typically neglects several issues highlighted in DNR. To date, it has not been able to capture the impact, evolution, and political origins of domestic institutions dedicated to productivity enhancement of local firms. It assumes that evidence as to the utility of certain measures, such as targeted FDI incentives, will be adopted and implemented by agencies such as investment boards rather than exploring the underlying politics. And the logic of inquiry in most quantitative research on governance follows what Abbott (1988) calls “general linear reality,” in which variables are additively related to one another (rendering it impossible to see the interactive relationships across institutional domains) and the focus is on average treatment effects (thus obscuring pathbreaking cases and follow-the-leader dynamics).

Relatedly, it assumes the relevance of “good governance” indicators (Kaufman et al 2010), despite evidence from case studies that such indicators miss key elements of governance on the ground at the sectoral level (Kurtz and Schrank 2007). DNR demonstrate how aggregated indicators such as “good governance” are unable to explain variance in performance across industrial sectors—Malaysia is an obvious case, a country that for decades has scored well on good governance and indicators of macroeconomic management but whose policies in the automotive sector were a colossal failure. Further, DNR's cross- and within-case analysis is exemplary of mid-range and a mixed-methods approach that stands in contrast with an obsession on external validity to the neglect of context (Haggard 2018, 93–94).

Links to the literature on global value chains

In the last two decades, much of the analysis of industrial development has been conducted through the lens of global value chains. Much of the literature is concerned with how value chains are organized and, in particular, on the forms of governance of inter-firm relationships within the chains (Gereffi, Humphrey, and Sturgeon 2005).

In moving beyond an exclusive focus on the state and domestic institutions, the literature on global value chains, including work on the automotive industry (e.g. Sturgeon et al., 2016), provides a useful supplement, pointing to how inter-firm networks can assist local firms to upgrade their capabilities and gain access to international markets. Much of the literature, however, is characterized by atheoretical case studies and/or typologies that have failed to generate any significant cumulation of knowledge. By focusing on within-value chain opportunities, that is, primarily on relations among firms, many studies within the GVC framework provide only a partial sense of the factors influencing firm upgrading. Sophisticated analysts have always noted that the institutional context in which value chains operate has a profound impact on their effects (e.g., Ponte 2002). This nuance was frequently lost when the multilateral economic institutions became enthusiastic advocates of developing economies making themselves as attractive candidates as possible to the lead firms in value chains primarily through the application of policies associated with the Washington Consensus.

More recent work on GVCs is more consistent with DNR by virtue of its emphasis on “bringing the state back in” (Eckhardt and Poletti 2018) and on various kinds of non-market, often local institutions (Corredoira and McDermott (2014). Reinforcing this move toward institutions, one of the few “meta-analyses” of the burgeoning literature on GVCs found that a crucial element in firms’ ability to upgrade within GVCs is the local institutional environment and, more specifically, “the capacity of the state” and “public–private collaborative engagement” (Pipkin and Fuentes 2017, 547, 549). These factors were critical in the (relatively rare) success of some firms in upgrading while the majority, in their terms, underwent a process of “treadmilling” in which attempts to upgrade were followed by “backsliding, decay and obsolescence.” The implication of this more recent extension of the GVC and related literature is consistent with a core DNR contention—namely, that the absence of effective local institutions designed to promote spillovers from foreign producers will result in weak technological capabilities on the part of local firms and, as a result, fragile linkages between them and foreign lead firms (see also Ravenhill 2014).¹

Yet, with some exceptions (e.g. Corredoira and McDermott 2014), the GVC and related literature pay little if any attention to the institutions that contribute more directly to improvements in sector and firm upgrading. Nor do they address one of the core elements of DNR—namely, the origins of these institutions, i.e. the factors leading political leaders to expend scarce resources on their creation.²

Upgrading, technological catch-up and the new industrial policy

Finally, DNR's concern with the productivity and technological capacities of local firms so central to intensive growth makes it highly relevant to a growing literature on technological catch-up, inspired by earlier work of Sanjaya Lall and Linsu Kim (e.g. Lee et al 2021, Yeung 2016). Drawing on both large-N and in-depth case study analysis, the scholarship on technological catch-up recognizes the challenges of technology absorption by local firms, sheds light on specific challenges and stages of such absorption, and recognizes the importance of institutional ecologies and innovation systems for countries facing middle-income traps and the challenge of taking advantage of participation in global value chains.

This literature's appreciation of institutions is informed by the “new industrial policy,” which places less emphasis on top-down, technocratic directives than on a recursive process of discovery, learning and diagnostic monitoring to ensure ongoing error correction. Such functions require effective institutions.

Because of its own sociology of knowledge, this literature has tended to shy away from the political economy question of where these industrial policies and corresponding institutions came from in the first place, or the question of what DNR call “origins.” There are scattered mentions of factors such as the power of vested interests, the tendency of political leaders to have short time horizons, and the potentially utility of crises (e.g. Lee et al 2021). But these are not part of the core arguments and tend to occupy the same explanatory space as “political will” in the work of the early neo-classical economists who turned their attention to the region over the course of the 1970s. Rather than addressing the question of why political elites would expend scarce resources to develop institutions with effective monitoring capabilities, this literature emphasizes rather their functionality for growth.

Two cases, one positive and one negative, are illustrative. Fundación Chile, the “business incubator” key to Chile's successful salmon industry, is frequently and rightly cited as the kind of public–private institution to which other middle-income countries should aspire. Not mentioned is the critical question of why Pinochet, a leader best known for his turn to neoclassical market orthodoxy, would support such an institution-intense approach to sectoral development. A second case involves Malaysia's PEMANDU (Performance Management and Delivery Unit), established by the Malaysian government in 2009 with the expressed purpose of coordinating diverse public agencies and private sector interests in a recursive, “new industrial policy”-type process of learning and monitoring (Sabel and Jordan 2015). In the event, PEMANDU has not lived up to expectations owing to intra-bureaucratic infighting and broader political resistance.

Our purpose in this roundtable is not to generate an altogether new research paradigm, but simply to reflect on the ways in which the early industry focus of research on the region's rapid growth might be revived, updated and appropriately amended. The structure of the roundtable is as follows. The journal invited four respondents to reflect on the strengths and weaknesses of DRT: Natasha Hamilton-Hart (University of Auckland), Eun Mee Kim (Ehwa University), Roselyn Hsueh (Temple University), and Robert Wade (London School of Economics). The authors, as always, have the last word.