This article was downloaded by:[Chen, Ling] On: 6 March 2008 Access Details: [subscription number 791131633] Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK New Political Economy Publication details, including instructions for authors and subscription information: http://www.informaworld.com/smpp/title~content=t713439457 Preferences, Institutions and Politics: Re-Interrogating the Theoretical Lessons of Developmental Economies Ling Chen a a Department of Political Science, Johns Hopkins University, Baltimore, MD, USA Online Publication Date: 01 March 2008 To cite this Article: Chen, Ling (2008) 'Preferences, Institutions and Politics: Re-Interrogating the Theoretical Lessons of Developmental Economies', New Political Economy, 13:1, 89 - 102 To link to this article: DOI: 10.1080/13563460701859835 URL: http://dx.doi.org/10.1080/13563460701859835 PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf This article maybe used for research, teaching and private study purposes. Any substantial or systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material. Downloaded By: [Chen, Ling] At: 14:15 6 March 2008 New Political Economy, Vol. 13, No. 1, March 2008 REVIEW ESSAY Preferences, Institutions and Politics: Re-Interrogating the Theoretical Lessons of Developmental Economies LING CHEN David Kang, Crony Capitalism: Corruption and Development in South Korea and the Philippines (Cambridge University Press, 2002) Vivek Chibber, Locked In Place: State-Building and Late Industrialization in India (Princeton University Press, 2003) Yongping Wu, A Political Explanation of Economic Growth: State Survival, Bureaucratic Politics, and Private Enterprises in the Making of Taiwan’s Economy, 1950 – 1985 (Harvard University Press, 2005) Once upon a time, the emergence of Japan, South Korea and Taiwan as the stellar examples of industrial growth generated a spate of literature on the developmental state model. Then there came a time when the neoliberalists’ apocalyptic predictions of the death of the developmental state proliferated. In retrospect, ten years after the Asian financial crisis, what are the main theoretical lessons one can learn from these developmental economies? On the one hand, reality has obviously not met the neoliberal expectation. While no one would deny the various problems these economies face today, equally clear is the fact that these states have entered an era of adaptation and transformation rather than that of final demise.1 On the other hand, the original developmental state model has also been shown to be inadequate in understanding the development experience of these countries and both regional specialists and theorists have called for a reassessment of the theory. When emerging in the 1980s and early 1990s with the resurgence of statist approaches in social science, the developmental state model placed its emphasis on the autonomy of the state in transcending parochial social interests and the Ling Chen, Department of Political Science, Johns Hopkins University, 3400 N. Charles Street, Baltimore, MD 21218, USA. ISSN 1356-3467 print; ISSN 1469-9923 online/08/010089-14 # 2008 Taylor & Francis DOI: 10.1080/13563460701859835 Downloaded By: [Chen, Ling] At: 14:15 6 March 2008 Ling Chen capacity of the state in achieving development priorities through industrial policies.2 However, this state-centric perspective was subsequently criticised for inadequate attention to society. It is within this context that Peter Evans proposes the notion of ‘embedded autonomy’ to the developmental state theory. The notion combines Weberian bureaucracy’s insulation from manipulation of social groups with the dense social networks that bind the state and business together in achieving industrial transformation.3 In a similar vein, Linda Weiss employs the concept of ‘governed interdependence’ to explain a state – business relation in which public and private participants maintain their respective autonomy, and yet are governed by broader goals set and monitored by the state.4 In both formulations the developmental state is no longer an entity lording over society but a partner of the business sector under the joint project of industrial transformation. This delicate balance and the synergistic relationship between state and business is probably thus far the most fruitful contribution of the developmental state model and the most promising area for further enquiries. In a fastchanging world where particular development tools shift from one era to the next, the core meaning of the developmental state concept, as Chalmers Johnson notes, eventually boils down to a mutually beneficial state– business relationship.5 Furthermore, exploring such relationships also constitutes an important step in going beyond the tautological way of identifying developmental states as ‘those that induce development’.6 However, it is also precisely regarding the theorisation and explanation of state –business relations that the notions of embedded autonomy and interdependent governance are found wanting. How could state and business in developmental economies achieve a balance that is so difficult to arrive at and inherently paradoxical? What factors enable them to maintain such synergistic linkages without degenerating into malignant ties elsewhere?7 Three recent books by David Kang, Vivek Chibber and Yongping Wu shed new light upon state– business relations in these economies. Collectively their scholarship highlights the importance of ‘bringing business back in’. They argue that if business groups are after all the direct agent for achieving economic growth, the task of sustaining the state –business relationship should not be unilaterally shouldered by the state, but should depend on the preferences of businesses as well. A further call from this body of literature is the need to ‘bring politics back in’. Their works jointly demystify the romanticised scenario where the bureaucracy is ‘totally depoliticised, socially disembodied, and in rational pursuit of a self-evident national interest’.8 They add to the picture the real concerns of political survival as well as the impact of political struggles and bureaucratic conflicts on economic policies. More importantly, instead of simply putting forward a set of observations about state– business relations, their analyses bring to light the iterative interactions between state managers and business actors, and spell out the ways in which preferences, institutions and politics interplay to influence both the process and outcome of such interactions. I argue that compared with the state-cum-development paradigm, such a new institutionalist perspective is potentially more helpful in understanding the theoretical lessons of developmental economies in the long run.9 90 Review Essay Downloaded By: [Chen, Ling] At: 14:15 6 March 2008 Crony capitalism and transaction costs: Kang’s model of corruption and growth Despite its title Crony Capitalism, David Kang’s book departs from the purely neoliberal view of state intervention as the culprit for the 1997 financial crisis. It instead aims to reconcile economic growth with corruption through a new institutional economics approach, namely, a transaction costs explanation. The main puzzle Kang seeks to unravel is why cronyism coexisted with rapid economic growth in South Korea but impaired economic performance in the Philippines. In contrast with the statists’ depiction of an autonomous, meritocratic Weberian state in South Korea and a corrupt and predatory state in the Philippines, Kang shows that the internal organisation of the state does not lead to ex ante predictions about developmental efficiency. Under the regimes of both Park Chung-hee and Ferdinand Marcos, politicians have reigned and ruled and the bureaucracy has been distinctly subordinate to political interests. Both have cosy government– business relationships and dense patronage ties and both have seen endless reports of corruption scandals. The real difference, Kang argues, lies in the configuration of state– business relations. In South Korea, the balance of power and the mutual hostages between a small set of government and business elites enabled them to collude, but prevented either from gaining the upper hand. Money politics in this case served to reduce transaction costs and became the facilitator of growth. By contrast, government–business relations in the Philippines swung like a pendulum from bottom– up rent seeking by business oligarchies to top –down plundering of society by Marcos. Corruption thus incurred large deadweight losses and choked economic efficiency. Democratisation in the 1980s, however, fragmented the Korean state and strengthened the chaebols, thus breaking the state– business balance, whereas in the Philippines, democratisation constrained excessive state authority and balanced the relationship between the government and business. This finally led to the different experiences of the two countries in the 1997 crisis. Kang’s account brings to the forefront the importance of the ‘larger institutional environment’ of state– business interactions that the statist literature typically ignores. Regarding the puzzle of why efficacious state– business linkages in East Asia did not degenerate into malignant networks elsewhere, Kang goes beyond searching for answers in the characteristics of the state and draws attention to the ‘broad contours’ of government– business relationship, which produce different types of cronyism through different combinations of government and business strength. Despite this important contribution, the explanatory power of Kang’s model is nevertheless undermined by a number of problems regarding his basic assumptions and causal mechanisms. Rent seeking and the problems of the transaction costs explanation Kang derives his model from new institutional economics where actors face transaction costs in achieving long-term agreements. According to Kang, rent seeking can be useful in reducing such costs when there are mutual hostages among smallN providers (government) and consumers (business) of the rents: it can help to 91 Downloaded By: [Chen, Ling] At: 14:15 6 March 2008 Ling Chen overcome prisoner’s dilemmas, leading to better information, more effective monitoring and long-term iteration of the game among actors. On the contrary, when either side of the actors are large-N and fragmented, the power balance is tilted towards the small-N side, leading to cascades of supplicants to compete for rents and to bid up the expected price of rents – a situation which eventually increases the transaction costs.10 The problems for this explanation are two-fold: first, economists find that without resorting to other features of the political economy, the magnitude of transaction costs for rent seeking is indeterminate in terms of number of actors; second, even if such magnitude can be determined, the model only helps to explain the difficulty or ease in overcoming coordination problems, not its aggregate impact on the economy.11 In other words, cronyism can benefit overall economic performance only to the extent that the saved resources are put to productive use, which the model itself says nothing about. For example, why was cronyism used more productively as an informal tool for credit allocation under the Park regime than under the Marcos regime in which it was used for protecting oligarchies and assassinating political rivals?12 The model’s inability to capture this difference stems from its assumption that actors in both countries automatically maximise their self-interest and seek rents whenever they can. Variation in such outcomes is caused by the constraints of the ‘institutional environment’, which can be ultimately attributed to the number of actors on each side. Leaving no place for agency, the model paradoxically contradicts Kang’s own intention to ‘bring politics back in’, for although politics can incur different magnitudes of transactional costs, they are eventually homogenised and viewed as ‘qualitatively similar’ – that is, as tools that politicians employ at the expense of economic merits. However, both the preferences of the actors and the political ends they pursue differ across institutional settings. Institutions are not merely constraints derived from the number of players; they are also and prima facie the formal and informal rules that profoundly shape actors’ preferences. As such, the particular goals to which money politics are directed and, for lack of a better term, the particular ‘rents’ that are being exchanged cannot easily be measured in terms of quantity – they do make a qualitative difference. The differences between South Korea and the Philippines One of the most apparent differences between South Korea and the Philippines stems from the early post-colonial period land reform (or the lack thereof), which respectively produced the chaebols in South Korea and the landed oligarchies in the Philippines. Although both were large and family-based conglomerates and both sought favourable political policies, their impacts on the economy were decisively different: the chaebols harnessed political protection to compete on the international market, while the landed oligarchies used politics to oppose industrial transformation and dragged the whole economy down the industrial value chain.13 In addition, although political and economic elites in both countries were connected through varied ties, Park was able to orchestrate a bifurcation of political and economic bureaus that allowed him to build political patronage without sacrificing economic efficiency. Meanwhile, business seldom challenged the 92 Downloaded By: [Chen, Ling] At: 14:15 6 March 2008 Review Essay government politically.14 As a result, politicians and businesses often bargained as a whole with each other through peak associations. In stark contrast, large families in the Philippines were able to build their own political and economic empires. They turned to government not only for achieving economic monopoly, but also for pursuing their intense interests of getting into the political palace. Corruption was used simultaneously as a political and economic weapon that eventually turned the whole nation into a battlefield. More importantly, while political considerations played crucial roles in both countries’ economic policies, the type of political goals that the ruling class pursued grew out of domestic and international contexts. No one would deny that industrial policies in South Korea were sometimes used to satisfy personal needs. Yet under domestic unrest, possible US military withdrawal and an active North Korean threat, the competitiveness of the industries was associated with the life and death of the ruling elite, and thus was identified as convergent with their political interests. While the performance criteria that government imposed on chaebols might not be as stringent as Alice Amsden describes, evidence shows that Park indeed was able to exercise discipline over these business groups by rewarding good performers and penalising poor ones.15 In contrast, Philippine politicians never experienced any severe external threat under US protection and they kept alert to challenges from domestic rivalries. Thus Marcos did not hesitate to ally with sugar, coconuts and grain monopolies and spent his resources attacking the Lopez, Aquino and Osmena families. South Korea could not afford to do so, but the Philippines could. Not all these differences are neglected by the empirical part of Kang’s book. However, they are finally dropped out of Kang’s framework precisely because they contradict the model itself. In the end, this model attributes the Philippines’ lucky situation in the financial crisis to the balance of power between a fractured democratic state and a dispersed business group. Kang argues that such a ‘weak state plus weak society’ combination is the most favourable situation because it drives the price of corruption close to zero. However, this flies in the face of the Philippines’ unstoppable political instability and meagre economic performance in recent years. Therefore, although Kang’s work helps to draw our attention from the narrow focus of the state to the configuration of state– business relations, its oversimplified assumptions substantially undermine its ability to capture the real world difference. Bringing capital ‘back in’: Chibber and the ‘locking-in’ process Also comparing a case of development failure with that of success in South Korea, Vivek Chibber’s Locked in Place approaches the subject from a comparative historical perspective. While Kang focuses on the issue of money politics, which is only the middle of a chain of events that affects growth, Chibber directly tackles the whole process of building the developmental state. While Kang compares South Korea with the Philippines, a country that is far less likely to become a developmental state, its counterpart in Chibber’s book, India, is a nation that once had this option on the agenda but was never able to put the relevant institutions in place. Asking why this was the case, Chibber traces his answer to the very 93 Downloaded By: [Chen, Ling] At: 14:15 6 March 2008 Ling Chen moments when the state was installed, and situates his comparison within a much broader historical horizon. Like Kang, Chibber severely questions the state-centric paradigm and emphasises the alliances and conflicts between state managers and capitalists. In his account, though, business has gained even more importance. He finds that the preferences of the business class and its initial reaction to the idea of launching a state-led development programme were of paramount importance for subsequent paths. In India this programme was frustrated by a well-organised offensive from the capitalists, whereas in Korea the business accepted the state’s agenda. Chibber attributes the divergent orientation of the capitalists to the deeper incentive structure of the development model each regime chose to adopt. In the Indian case, the state undertook the import-substituting industrialisation (ISI) model that provided business groups with the incentive to embrace state’s protection and subsidies but oppose its right to regulate and monitor industrial performance. By contrast, Korea was able to switch from ISI to an export-led industrialisation (ELI) model where the rigours of international competition made it in the interests of business to submit to the state’s discipline in exchange for assistance. In other words, Chibber reminds us that building a developmental state depends on whether there is a ‘developmental bourgeoisie’ that the state can ally with. This takes us further than the standard explanation of India’s failure and Korea’s success: that the former failed to take advantage of bourgeois enthusiasm while the latter had the unique power to coerce chaebols into the developmental agenda. Instead, Chibber illustrates that the state was not the only player in the drama and that the interests of capitalists made a great difference for development outcomes. The fact that ISI strategies have almost uniformly met with failure in country after country, while ELI has been viewed as the common ‘secret’ for North-east Asia’s success, suggests that the orientation of the capitalist class is not entirely a product of particular histories and local culture but is also impacted by certain identifiable mechanisms embedded in economic models themselves. Chibber fleshes out his argument not by ex-post reasoning but through solid reassessment of the historiography. Drawing on newly unearthed archives, he presents convincing evidence that Indian capitalists’ putative advocation of a developmental state was a myth at the very start, and the genuine purpose behind the Bombay Plan was to forestall socialist attacks. He also demonstrates that it was not that Park Chung-hee first built a strong developmental state and then unilaterally imposed the ELI strategy on chaebols. Instead, the onset of ELI was in congruence with the requests from domestic business, who wanted to gain entrance to the lucrative US market provided by the Japanese multi-national corporations (MNCs). Development models and institutional reproduction The core contribution of the book, however, does not lie in uncovering the difference between ISI and ELI as development models per se. More than enough literature has already shown how ISI breeds uncompetitiveness and how ELI helps to move the industrial products up the value chain.16 Rather, a far more useful way to read the two models in Chibber’s work is to understand them 94 Downloaded By: [Chen, Ling] At: 14:15 6 March 2008 Review Essay from a historical institutionalist perspective, that is, in the light of how these models constituted ‘critical junctures’ for two nations and ‘locked in’ subsequent paths. Both ELI and ISI rallied relevant actors around certain sets of institutional arrangements under which actors adjusted their strategies to accommodate the prevailing behaviour patterns. These development models kicked off the processes in which institutions reinforce themselves through what historical institutionalists call the mechanisms of reproduction and positive feedback, so that it became increasingly difficult to reverse the trajectories.17 What emerged from the onset of ELI in Korea in the 1960s was a tight state – business alliance cemented around a clearly defined economic strategy. Private firms’ dependence on the state for assistance in finance, technology and market coordination provided the latter with leverage to demand performance from the former. By helping business to succeed and gain profits in the international market, the state managers in turn gained legitimacy for continued intervention. Meanwhile, state structures were also realigned to meet the needs of industrial monitoring. Steady information flows between various economic ministries and the Economic Planning Board were ensured through project evaluation and monthly reports, and effective channels of communication between the state and firms were also established. These institutional arrangements both facilitated industrial success and further reinforced the model. By relying on an ISI strategy, however, India stepped on a starkly different path. Chibber’s story does not stop at the years of critical juncture (1947– 51), when installing a developmental state was strongly opposed by the capitalists. He also shows how the consequences were amplified afterwards. The massive offensive by business and demobilised labour opened an opportunity for forces within the state ministries to revolt against the planning agenda. As a result, the Planning Commission (PC) and the Industrial Bill (IB), key elements of economic planning, were radically compromised to the advantage of the business group. Although the PC was never able to function as a genuine coordinating agency, and the IB was never able to extend discipline to the industrialists, Chibber finds that ironically they were the first to be accused and paralysed when the economy hit crises. This shut the door for reform in the mid 1960s and provided capital with further justification and freedom to divert investment from priorities, creating bottlenecks for planning. As both the power of the planning authorities and the legitimacy of the planning process were repeatedly eroded, it strengthened the force that rolled back state intervention in favour of liberalisation. The potential problem of determinism Nevertheless, the picture Chibber portrays pushes him to the brink of determinism. Such determinism is first of all historical. While the Japanese offered Korea a ready entrée into the US market and enabled them to step on a successful path, India did not have that fortune and thus had to remain in the old development groove for decades. The determinism is meanwhile structural. Once the ISI model was put in place, both business and the state followed their structurally driven incentives so that the former became increasingly offensive and the latter 95 Downloaded By: [Chen, Ling] At: 14:15 6 March 2008 Ling Chen had to make one concession after another. Contingency seems to lie only at the beginning – that is, when critical juncture occurs. As scholars have noted, such a ‘contingency followed by lock-in’ formulation is quite limited in capturing the real dynamics of institutional development.18 By emphasising the ‘immutability’ of the institutions, Chibber fails to give enough attention to inherent tensions that are initially built into or endogenously generated by the institutions. These tensions can produce self-undermining mechanisms along with those self-reinforcing ones, which ultimately leads to change.19 For India, the seeds of ISI’s self-destruction process were created at the very beginning, and the continuation of ISI debilitated the planning agenda and eventually led to the demise of the model. For South Korea, however, the self-undermining effects were endogenously generated later in the development process, when chaebols became increasingly powerful and independent of the state.20 Such determinism also downplays the role of politics and political entrepreneurs. While the power of the state elite in lording over society has certainly been exaggerated in much of the statist literature, it does not mean that government and political entrepreneurs can only passively respond to structural constraints imposed by the development strategies. Strategies themselves are created and monitored by state managers in the first place. While Japan’s cooperation was crucial for Korea’s switch to ELI, it was after all the Park government which made sure the foreign partnership was favourable to domestic producers and pushed firms toward greater efficiency. When necessary, Park did not hesitate to bypass bureaucracy and negotiate directly with chaebol leaders to ensure compliance. Moreover, what is missing from Chibber’s account is that, even after launching the ELI, Park never suspended ISI, but continued secondstage ISI for capital-intensive products, using ELI as the source of foreign exchange after the phasing out of US aid. The use of both ELI and ISI reflected the economic, political and security considerations of state managers. By contrast, although the Nehru regime also demonstrated a certain degree of developmental ambition, the tentativeness of the planners limited the extent to which India could be compared with Korea in terms of political entrepreneurship. The space and degree of influence that such political leadership could carry was in turn conditioned by historical context and institutional settings, aside from the different leverage that each development model bestows upon the state. The political basis of industrial development: Wu’s study of Taiwan’s private sector Sharing Kang and Chibber’s concern with state– business relationships and their critiques of the statist approach, Wu’s re-examination of Taiwan’s route to growth sheds new light on another developmental economy. Central to Wu’s explanation of Taiwan’s economic success is its unique industrial structure, which was first characterised by a ‘strict division of labor’ under which the state owned enterprises (SOEs) dominated the upstream industries, the large enterprises (LEs) were entrenched in the intermediate-stream industries, and the small and medium-sized enterprises (SMEs) focused on downstream activities. Meanwhile, the economy was also characterised by a ‘dual market structure’ in 96 Downloaded By: [Chen, Ling] At: 14:15 6 March 2008 Review Essay which the SOEs and LEs jointly monopolised the domestic market while the SMEs dominated export production.21 Similarly to Chibber, Wu traces the origins of such an industrial structure back to the moment of state formation, when the Kuomintang (KMT) adopted particular strategies for different enterprises. As an émigré regime that faced a hostile local majority, the KMT not only consolidated its rule directly through SOEs, but also sought political support by selectively nurturing LEs through an ISI strategy. Yet to prevent the rise of political rivalries, the state meanwhile kept these Taiwanesedominated LEs from upward expansion. In contrast, the state initially ignored SMEs, which had little political might and were neither promoted nor restricted by the state. However, such a structure took shape not solely because of the state’s intention but also through the strategic interactions between the state and private actors. An important example for Wu was the state’s promotion of an export-oriented strategy in the 1960s. While the state originally expected LEs to be the main exporters, it was the SMEs that took the role. Much like Chibber’s account of Indian business, Wu finds that forms of protection and subsidies under ISI drove the LEs to opt for monopolistic profits in the domestic markets and discouraged them from taking risks in international markets. Precisely because of entrenchment of SOEs and LEs in the domestic markets, SMEs were able to step into the export vacuum and became the major exporters. Moreover, SMEs also became the engine of the entire economy by virtue of their backward linkage to the upstream SOEs and intermediate-stream LEs.22 If the formation of the unique economic configuration is largely a by-product of the strategic interactions between state and business actors, Wu’s analysis of subsequent state action clearly demonstrates that once the state found such configuration politically and economically useful, it deliberately safeguarded and reinforced it. On the one hand, the state continued to strengthen its ties with large businesses and kept their positions in domestic market. On the other hand, recognising the SMEs’ economic significance in the mid 1970s, the state began to develop special measures such as credit allocation to promote SMEs’ exports. Furthermore, Wu’s study of the hi-tech industries shows that the state not only reinforced but also gradually modified the industrial structure in late 1970s and early 1980s. In response to the legitimacy crisis, the pressure of liberalisation and the growing strength of the LEs, the state began to encourage market integration against previous market division and devoted increasing resources to nurturing the semi-conductor and computer industries in SMEs. The importance of politics and the irrelevance of institutions? Wu’s study contributes both empirically and theoretically to the study of developmental economies. Empirically, it fills the lacuna in the previous statist accounts of Taiwan’s industrial success, and provides explanations for the SMEs’ success both despite and because of the state’s industrial policies. Theoretically, it draws attention to the complexity of state– business relations. Instead of portraying a relationship in which the state simply either colludes or conflicts with business, he examines how government interacted differently with each kind of business and 97 Downloaded By: [Chen, Ling] At: 14:15 6 March 2008 Ling Chen how such interactions affected the incentives and strategies of all actors involved. This interplay also suggests another way for developing countries to switch from ISI to ELI strategies: instead of exercising discipline over those once heavily protected firms and forcing them to export, the KMT simply got around this problem by turning to the group of capitalists who were previously ignored. This makes Wu’s account much less deterministic than Chibber’s both in terms of the flexibility of the state– business configuration and the role played by political and economic agents. Industrial structure emerged as an outcome of the state– business interplay rather than as a well thought-out plan, and it could be subjected to modification when necessary. State managers were able to both formulate and change their strategies to meet domestic and international challenges. Businesses, especially the SMEs, also developed their own strategies and explored their own advantages both because of and in spite of state policies. In short, the configuration of the state– business relation was not fixed once for all. Another important contribution that Wu makes is that, like Kang, he brings attention to the political dimensions of Taiwan’s economic growth and challenges the view of the state as a unitary actor. He shows that, similar to its Korean counterpart, Taiwan’s economic bureaucracy was neither coherent nor autonomous but full of bureaucratic struggles, conflicts of interest and ideological divisions. Yet for Wu the role of politics goes far beyond corruption and rent seeking. He contends that the whole story of Taiwan’s industrial development should be understood as a ‘politically inspired industrial success’ instead of simple stateled development. He also reminds us that both industrial strategies and industrial structure had their political origins, and that the KMT state formulated or changed its industrial policies not out of purely economic considerations, but predominantly out of political motivations. However, Wu’s analysis shares an important shortcoming with Kang’s in that despite the full attention he gives to politics, he fails to differentiate between the different impacts that political motivations have on economic development. By pointing out that ‘economic goals are compromised for the sake of the political goals’, he tends to assume that political motivations are detrimental to development.23 This is true in some cases, as illustrated in his study of the automobile industry, where industrial policy failed because of bureaucratic conflicts. However, the impact of most other political goals highlighted in his work points in the opposite direction. The strong concern in establishing political legitimacy, the perception of threat from mainland China, and the growing pressure from the cessation of US aid all facilitated or even forced the state to make industrial development as a priority. There was nothing inherently contradictory, to use Wu’s own phrase, in the state picking both the ‘economic winners’ and the ‘political supporters’.24 Another major problem of Wu’s argument is the mirror image of Chibber’s. Throughout the book, Wu emphasises the importance of political leadership and the role of the strongman. He argues that since formal institutions are typically fragile in developing countries, it was individuals who played a leading role in economic policy formulation and implementation. Drawing on the insights of Graham Allison’s classic model of bureaucratic politics, he contends that the steering capacity of economic bureaucracy was highly questionable. He further 98 Downloaded By: [Chen, Ling] At: 14:15 6 March 2008 Review Essay demonstrates through case studies that it was politicians and strongmen who finally settled the coordination problems among different layers of the bureaucracy and between the state and the business sector. Although these findings are useful in challenging us to pay more attention to the role of individuals, they do not lead to his conclusion that institutions are not relevant to Taiwan’s economic success. It is highly problematic to judge the role of institutions merely through formal state organisations. As Wu himself realised, in Taiwan’s case informal rules and norms played more important roles and the state followed them in its interaction with different types of business sectors. While institutions can be and should be ‘humanised’ and ‘politicised’, as Wu proposes, it is misleading to dichotomise political individuals and the institutional environment in which they are embedded. Conclusion: preferences, institutions and politics in developmental economies Have the works discussed here led to the wholesale discrediting of developmental state theory? Hardly. The role of states in steering economic development and the effective use of industrial policies is still shown to be crucial. What distinguishes these analyses from previous explanations of the developmental state is that, instead of starting with the state’s own character and working towards the appropriate way to tame businesses, they spell out the patterns in which states and businesses interact with each other under the common project of industrial transformation. The notion of ‘interaction’ emphasised here goes beyond state– business connectedness and networks. As Kang’s work reminds us, there exists only a thin line between cooperation and collusion, and between embeddedness and capture.25 While all over the world political and economic actors are somewhat closely connected with each other, only a few of them could further use such connections for joint developmental projects. The success of such projects, as these three works have shown, depends on the interplay of interests, institutions and politics. This scenario differs from the view of both rational choice institutionalism and historical institutionalism regarding the role of institutions in economic development. According to the former, institutions as constraints of behaviours affect economic performance by setting a set of pay-offs that either encourage or inhibit productive activities. Yet, this view fails to take seriously the variation of actors’ interests across historical and institutional settings and devotes overwhelming attention to the rules of the game.26 By contrast, historical institutionalists often consider outcomes of economic development as results of particular paths determined by history at certain critical junctures, and tend to underestimate the potential sources for institutional evolution and change. Finally, both perspectives tend to neglect the important influence of politics on economic failure or success. Understanding the interplay of interests, institutions and politics first requires us to examine the concrete preferences of state and business actors. As these works convincingly demonstrate, it is not very helpful to either assume such preferences beforehand, as in Kang’s model, or induce them backwards from the economic outcome, as some statist accounts tend to do.27 Instead, these preferences partly 99 Downloaded By: [Chen, Ling] At: 14:15 6 March 2008 Ling Chen emerged from historical contexts and institutional environments that condition the state– business interactions. For example, what sets the Philippines apart not only from North-east Asia but also from India is the former’s lack of external threat and land reform, as well as the intermingling of political and economic powers. This inhibited any serious interest in establishing any joint developmental project between the state and the business. For countries that did succeed in launching such projects, albeit with different degrees of success, the preferences of the state and businesses are further shaped by the type of development model and industrial structure that binds the state and business together, as suggested in Chibber’s comparison of India and Korea and Wu’s study of Taiwan’s experience. The institutions underpinning such models or structures, which include formal and informal rules, pattern the actors’ interaction not only by influencing the direction of their preferences, but also by granting them different institutional positions and bargaining leverage. Thus, even if it is the case that in North-east Asia the preference of the state has a larger influence over the development outcome, as the statists claim, it is something that needs to be explained rather than assumed. The state– business interaction should be construed as an iterative process in the longue durée rather than a once-and-for-all equilibrium. Both Chibber and Wu situate their analysis in a thick historical background and trace such relations to the ‘critical juncture’ of state formation, when institutions started to take shape. Moreover, the configuration of state– industry relations was not imposed by the state against business preferences, but came into being either in congruence with the interests of the latter (as in Chibber’s book) or as a by-product of interaction between the two (as in Wu’s book). While Chibber’s work illustrates how mechanisms of endogenous reproduction emerging out of state– business interaction could firmly lock in subsequent development paths, Wu’s work suggests the possibility of both institutional reinforcement and institutional evolution. Between preferences and institutions lies the role of politics – another theme emphasised but inadequately theorised by this body of literature. In fact, more than two decades ago, in the very book in which the concept of developmental state was coined, Johnson pointed out that ‘in the developmental state economic interests are explicitly subordinated to political objectives’.28 What these three works further suggest is the fact that there was no separation of reign and rule in Korea’s and Taiwan’s government, and that the bureaucracy had never been autonomous from politics. However, by putting forward a vague notion of ‘political motivations’, these authors fail to distinguish between the long-term, overarching political motivations, such as concern for legitimacy, security and national survival, from the parochial bureaucratic struggles that are more or less common among all countries. The former tend to prioritise industrial development and view it as a pre-condition for reaching political goals, whereas the latter can often make economic competitiveness a sacrifice. The extent to which each influences industrial transformation in these countries is not only directly linked to the preferences of political actors but is also heavily influenced by the national and institutional environment in which these actors are situated. In turn, politics and political entrepreneurs can not only affect the effectiveness of institutions, but 100 Downloaded By: [Chen, Ling] At: 14:15 6 March 2008 Review Essay also create or modify institutional arrangements to meet their political and economic needs. Understanding the lessons of developmental economies in terms of the repeated interplay among preferences, institutions and politics not only could capture the dynamics of state –business relations more accurately than the previous statecum-development paradigm, but is also more useful for generating a new research agenda. Several recent studies on the evolution of political economy in East Asia have shown that a seemingly stable configuration between state and business may lead to self-destruction later on, and that an effective development strategy during period of catch-up may stifle further industrial upgrading in the era of economic liberalisation.29 This suggests that flexibility and adaptability in a changing economic environment are more essential for economic survival in the long run. As Johnson observed, the history of Japan’s economic bureaucracy is one of a constant search for appropriate public– private relations and effective methods of intervention.30 Yet, in the light of the literature reviewed here, we should probably add that precisely because of the uncertainty of the future, the state needs to work closely with business during this process instead of searching for the answer alone. In this regard, the interplay of preferences, institutions and politics emphasised in these three works constitutes a more dynamic approach for understanding the changing state– business relations and the evolving developmental path. Notes 1. Linda Weiss, ‘Developmental States in Transition: Adapting, Dismantling, Innovating, not “Normalizing”’, Pacific Review, Vol. 13, No. 1 (2000), pp. 21–55; Linda Weiss, ‘Guiding Globalization in East Asia: New Roles for Old Developmental State’, in Linda Weiss (ed.), States in the Global Economy: Bringing Domestic Institutions Back in (Cambridge University Press, 2003), pp. 245–70; Joseph Wong, ‘From Learning to Creating: Biotechnology and the Postindustrial Developmental State in Korea’, Journal of East Asian Studies, Vol. 4, No. 3 (2004), pp. 491–517; Steven Vogel, Japan Remodeled: How Government and Industry Are Reforming Japanese Capitalism (Cornell University Press, 2006). 2. Chalmers Johnson, MITI and the Japanese Miracle: The Growth of Industrial Policy, 1925–1975 (Stanford University Press, 1982); Alice Amsden, Asia’s Next Giant: South Korea and Late Industrialization (Oxford University Press, 1989); Robert Wade, Governing the Market: Economic Theory and the Role of Government in East Asian Industrialization (Princeton University Press, 1990); Ziya Onis, ‘The Logic of the Developmental State’, Comparative Politics, Vol. 24, No. 1 (1991), pp. 109 –26. 3. Peter Evans, Embedded Autonomy: States and Industrial Transformation (Princeton University Press, 1995). 4. Linda Weiss, The Myth of the Powerless State (Cornell University Press, 1998); Linda Weiss & John Hobson, States and Economic Development: A Comparative Historical Analysis (Polity Press, 1995). 5. Chalmers Johnson, ‘The Developmental State: Odyssey of a Concept’, in Meredith Woo-Cumings (ed.), The Developmental State (Cornell University Press, 1999), p. 60. 6. Evans, Embedded Autonomy, p. 30. 7. On these questions, both Evans and Weiss eventually slip into a statist or a state-cum-development approach that finds answers in the character of state organisation and the policies associated with that type of the state. 8. T. J. Pempel, ‘The Developmental Regime in a Changing World Economy’, in Meredith Woo-Cumings (ed.), The Developmental State (Cornell University Press, 1999), p. 144. 9. The notion of ‘institutions’ emphasised in this article differs from the conventional concept used in the statist literature. Drawing on the new institutionalist perspective developed by Douglass North, Peter Hall and many others, this article refers to institutions not only as formal state organisations, but also and mainly as the formal and informal rules that structure the behaviours of both states and business. 10. For a detailed economic explanation of Kang’s model, also see David Kang, ‘Transaction costs and Crony Capitalism in East Asia’, Comparative Politics, Vol. 35, No. 2 (2003), pp. 439–58. 101 Downloaded By: [Chen, Ling] At: 14:15 6 March 2008 Ling Chen 11. Mushtaq Khan, ‘The Efficiency Implications of Corruption’, Journal of International Development, Vol. 8, No. 5 (1996), pp. 683–96. 12. Trying to slip over this problem, Kang hovers between two different dependent variables – money politics and economic development. Meanwhile his work is caught between tensions of two contradictory logics: one compares the beneficial with the detrimental effects of corruption, and the other emphasises the different levels and magnitude of corruption in two countries. 13. By comparing industrial structure Kang argues that the Philippines has been as industrialised as Korea since 1960s. However, he simply equates manufacturing with industrialisation, ignoring the nature of products each country produced and exported. 14. Amsden, Asia’s Next Giant, p. 146. 15. See the five works mentioned in Chibber, Locked in Place, fn. 53, p. 271. 16. As Chibber notes, the ELI model ought not to be confused with its forerunner that was based on the export of agricultural commodities and raw materials. For comparisons of ISI and ELI strategies, see Gary Gereffi & Donald Wyman (eds), Manufacturing Miracles: Paths of Industrialization in Latin America and East Asia (Princeton University Press, 1990); Stephan Haggard, Pathways from the Periphery: The Politics of Growth in the Newly Industrializing Countries (Cornell University Press, 1990). 17. Kathleen Thelen, ‘Historical Institutionalism in Comparative Politics’, Annual Review of Political Science, Vol. 2, (1999), pp. 369 –404. 18. Paul Pierson, Politics in Time: History, Institutions, and Social Analysis (Princeton University Press, 2004); Kathleen Thelen, How Institutions Evolve: The Political Economy of Skills in Germany, Britain, and the United States, and Japan (Cambridge University Press, 2004). 19. Thelen, How Institutions Evolve; Avner Greif, Institutions and the Path to the Modern Economy: Lessons from Medieval Trade (Cambridge University Press, 2006). 20. Although Chibber acknowledges such dynamics in the epilogue, he fails to integrate them in his own theoretical framework, particularly in the Indian case, where such dynamics occurs before the rule of Indira Gandhi. 21. Wu, A Political Explanation of Economic Growth, pp. 2, 183. 22. Meanwhile the SMEs were able to mobilise a series of ‘social goods’ such as flexible production and cheap labour, and to curb financial markets to facilitate their export production. 23. Ibid., p. 325. 24. Ibid., p. 104. 25. While it remains an empirical question whether the bureaucracy in South Korea and Taiwan was indeed as incoherent as these works suggest, particularly compared with other countries, it is at least clear that one cannot rely purely on state organisations themselves to tell the difference. 26. For a representative work, see Douglass North, Institutions, Institutional Change and Economic Performance (Cambridge University Press, 1990). 27. In other words, there is neither a natural ‘developmental state’ nor a natural ‘developmental bourgeoisie’. 28. Johnson, MITI and the Japanese Miracle, p. 24. 29. Marie Anchordoguy, Reprogramming Japan: The High Tech Crisis under Communitarian Capitalism (Cornell University Press, 2005); Steven Vogel, Japan Remodeled; Stephan Haggard, Wonhyuk Lim & Euysung Kim (eds), Economic Crisis and Corporate Restructuring in Korea: Reforming the Chaebol (Cambridge University Press, 2003); Sea-Jin Chang, Financial Crisis and Transformation of Korean Business Groups (Cambridge University Press, 2003). 30. Johnson, ‘The Developmental State’, p. 57; Johnson, MITI and the Japanese Miracle, p. 29. 102