We cannot verify your location
Browse Book and Journal Content on Project MUSE

Download PDF

David C. (David Chan-oong) Kang - Bad Loans to Good Friends: Money Politics and the Developmental State in Korea - International Organization 56:1 International Organization 56.1 (2002) 177-207

Bad Loans to Good Friends:
Money Politics and the Developmental State in South Korea

David C. Kang


Since gaining its independence in 1948, South Korea (hereinafter referred to as Korea) has seen a seemingly endless flow of corruption scandals bring down scores of elites. Among those who have served time in jail or been exiled are former presidents Chun Doo-hwan and Roh Tae-woo; members of many presidential staffs; and a slew of military officers, politicians, bureaucrats, bankers, businessmen, and tax collectors. 1 While numerous observers professed to be shocked--Shocked!--at the revelations, in reality such scandals are a recurrent theme in Korean political history, and the exchange of money for political influence has been not just an open secret but common knowledge.

For decades the literature on Asian development largely treated the prevalence of money politics as inconsequential or as peripheral to the "real story" of Korea: economic growth led by a developmental state composed of technocrats and austere military generals who emphasized export-oriented industrialization. Growth was so spectacular that the reality of corruption was concealed or was dismissed out of hand. And until late November 1997 and the stunning fall of the Korean won, observers argued that better government in Asia was a prime reason for that region's spectacular growth. 2

Has corruption historically been prevalent in Korea? If so, why? How can we reconcile the view of an efficient developmental state in Korea before 1997 with reports of massive corruption and inefficiency in that same country in 1998 and 1999?

Politics is central to the answer. In this study I make two arguments. First, money politics was extensive in Korea both during and after the high-growth era. Second, [End Page 177] political--not economic--considerations dominated policymaking. Focusing on the exchange of favors for bribes between state and business, I argue that politics drove policy choices even at the height of Park Chung-hee's rule, that bureaucrats were not independent of political interference in setting policy, and that business and political elites wrestled with each other over who would reap the rents to be had. Money politics was more extensive in Korea than the conventional wisdom allows. In fact, so rampant was corruption that we cannot dismiss it; rather, we need to explain it.

This study explains both the past and the present, and it compares the patterns of money politics in the early post-independence era with those that arose after the democratic transition in 1987. While during the Park era a balance of power between businessmen and politicians kept corruption from spinning out of control, the transition to democracy altered the basic business-state relationship, allowing business to exert a greater influence than usual over policy decisions. I use the term "money politics" because it is less normative than "corruption," and also because it highlights public-private interaction. Money politics subsumes both corrupt practices such as bribery and legal practices such as campaign finance. Both corruption and rent seeking are broader terms and can occur at the private-private level as well as vis-à-vis the state.

The political hypothesis advanced in this study suggests a new direction for our research about the developmental state. Situated at the intersection of international relations and comparative politics, and comprising a set of ideas about institutional arrangements and policy choices, the developmental state perspective held up Asia's seemingly neutral bureaucracies, effective politicians, and consistent trade policies as central factors in economic growth. 3

The Korean experience suggests broader implications for the study of government-business relations in developing countries. Most important, a model of politics is central to understanding the developmental state. We cannot assume benevolence on the part of the developmental state. A "hard" view of the developmental state--that the state is neutral, picks winners, and provides public goods because the civil service is insulated from social influences--is difficult to sustain empirically. However, even the "soft" view--that governments can have a beneficial effect however government action is attained--needs a political explanation. The Korean state was developmental--it provided public goods, fostered investment, and created infrastructure. But this study shows that this was not necessarily intentional. Corruption was rampant, and the Korean state intervened in the way it did because doing so was in the interests of a small group of business and political [End Page 178] elites. Producing public goods was often the fortunate by-product of actors competing to gain the private benefits of state resources.

Second, it is unwise to focus on individual policy choices (for example, export-oriented industrialization) or specific institutional arrangements (the bureaucracy) as isolated issues. Both institutions and policies are intervening variables, and the larger institutional environment, in this case the government-business relationship, affects any specific issue. 4 Institutions are more than just the organization of the state; they can be legal or corporate as well, and there are trade, regulatory, and financial policies. A distorted picture will emerge if we focus mainly on state institutions while ignoring industrial organizations, or if we focus on trade policy while ignoring lax regulatory and financial policies. 5 The Korean case shows that political and economic entrepreneurs are quite resourceful, and that institutional design or policy choices by themselves are subject to manipulation, evasion, and modification. 6

Third, because the institutional arrangements and policy choices in individual countries vary greatly, we must also avoid the temptation to look for a universal set of arrangements or variables that will lead to growth or stagnation. The argument I will present suggests that to understand the nature of policymaking in developing countries, we must first understand the particular political challenges that confront individual leaders and the manner in which business attempts to influence policies. The strategic allocation of economic policy and benefits is an important political resource.

Finally, the findings in this article raise questions about the role of domestic-international linkages in developing countries. Scholars have tended to focus on export-oriented trade policies when studying developing countries, but the rise of Korea, accompanied as it was by extensive corruption, leads to questions about how and why the Korean state had the opportunity to act as it did. International finance, the actions of international institutions, and the actions of the United States were all crucial for Korean development. In the study of development, as important as the study of domestic politics is the study of what international factors sustained the domestic situation, what international pressures existed, and what external actors participated in a domestic configuration or allowed it to exist.

I first introduce an informal model of the government-business relationship that focuses on describing and explaining the exchange of bribes for rents. I then show that policymaking followed a political--not an economic--rationale during the era of Park Chung-hee. In the third section I show how the democratic transition in 1987 increased the role of money politics. I conclude by revisiting the developmental state and offer a tentative explanation for how growth can occur despite corruption. [End Page 179]

Money Politics, Rent Seeking, and Corruption

I focus on the rent seeking and corruption that occur between public and private actors. At the heart of the model is the idea that those with excessive power will tend to abuse it. The dependent variable is the exchange between state and business of favors for bribes. The independent variable is the relationship between state and business.

James Buchanan defines rent seeking as "that part of the payment to an owner of resources over and above that which those resources could command in any alternative use." 7 Thus rents are created when an actor manipulates prices and causes them to diverge from competitive levels, and the existence of rents can lead to corruption by various actors attempting to gain access to the rents. By such manipulation, the actor itself, or some other actor on whose behalf the price manipulator is acting as an agent, is able to reap "excess profits." 8 Rents can be created in a number of ways, but a principal way is through state intervention. 9 The state uses its power to manipulate prices and markets and to generate rents. For example, import licenses confer rents by restricting the volume of imported goods that comes into a country, and thus those actors able to import the restricted goods will be able to sell those goods at a higher-than-market price, thus obtaining rents. 10 By intervening, the government creates incentives for business to attempt to influence policy decisions, and corruption occurs when businessmen use bribery, personal connections, or other means in an attempt to influence policy decisions. The distribution and volume of rents are thus a function of the relative strengths of the state and the business sector.

The Politics of Corruption

If there were no government distributing rents, there would be no corruption, and thus a key issue is how to model the government-business relationship. In examining both the supply and the demand for political corruption, this simplified model of the government-business relationship necessarily abstracts from a rich reality.

Following Andrei Shleifer and Robert Vishny, a state can range from coherent to fractured. 11 A state is coherent if it can formulate preferences independent of social [End Page 180] influences and if political leaders have internal control over their bureaucrats. 12 Although there are many possible configurations of the relationships among political leaders, bureaucrats, and political organizations (domestic politics: parties, associations, and so on), for the sake of simplicity I focus on two polar cases. The most coherent situation exists when political leaders have full control over their political organizations and their bureaucrats, and in this case leaders actively use domestic politics as a means of ensuring continued rule. At the other pole, the most fractured situation exists when leaders survive only tenuously, they engage in constant conflict with political organizations over the form and content of the state, and bureaucrats can play off "multiple principals" to their own advantage. 13 At the heart is the question of control.

However, it is the interaction of government and business that is of interest, and we thus need to understand business organization as well as government organization. My view of the business sector builds on the work of Michael Shafer. 14 He argues that the organizational characteristics of the predominant economic sector (for example, mining or agriculture) have different implications for its relationship to the state. In sectors with high asset specificity and high production inflexibility, companies will be less responsive to market signals and less able to adjust quickly to exogenous shocks, either political or economic. These types of firms will have more incentive to resist the state's attempts to intervene. Alternatively, in sectors with low asset specificity, low production inflexibility, and low factor inflexibility, firms will adjust more easily to exogenous forces.

The approach used here examines business more broadly than Shafer did. In this model, a strong concentrated business sector is the diversified business group, comprising well-organized firms that cover many sectors of the economy. 15 As Ben Ross Schneider puts it, "big (and encompassing) is beautiful." 16 Under this definition of diversified firms, companies cover many sectors rather than one, may have import-competing subsidiaries as well as export-oriented subsidiaries, and may have both agricultural and urban firms. Given their cross-ownership of various subsidiaries and the range of their interests, these firms' interests cannot be neatly categorized. In addition, the larger that diversified business groups are relative to the economy as a whole, the more likely they are to attempt to influence government policy, and the more likely they are to wield political influence. These conglomerates can be differentiated from single-sector, smaller, and less-diversified firms. On a spectrum we might put individual craftspersons at one end and put Japanese [End Page 181] keiretsu, Korean chaebol, Philippine family conglomerates, and Mexican grupos at the other end. 17

We can now build the analogy with respect to politics and corruption, with a coherent/fractured state along one axis and a concentrated/dispersed business sector along the other (Figure 1). In this model I take as given the initial distribution of rights and the type of actors. These are exogenous to the model, and I remain agnostic as to why and how society came to look a certain way.

Types of Corruption:
Bottom-Up or Top-Down

Although the model is a simplified abstraction of the government-business relationship, it captures the underlying dynamics of how corruption occurs. There are two analytically distinct types of corruption: the top-down predation by a strong state on society, and the bottom-up rent seeking of powerful groups that overwhelm the state's ability to contain and channel their demands. Neither is analytically prior to the other, and they can occur simultaneously under the right circumstances.

Top-down corruption has best been explicated in the notion of a "predatory" state. 18 The predatory state takes advantage of a dispersed and weak business sector. Political elites pursue outright expropriation; they solicit "donations" from businessmen, who in turn either are "shaken down" by the regime or volunteer bribes in return for favors; and they employ other means as well. 19 In contrast, bottom-up corruption occurs when social actors have the power to overwhelm the state. When the concentration or strength of the business sector is enough to force concessions from the state, rent-seeking behavior results. Potential state influence over economic [End Page 182] life is vast, and those businessmen or groups privileged enough to receive low-interest loans or import quotas will benefit at the expense of others. 20 Indeed, a typical problem in developing countries is the state's ability to resist society's demands. 21 When rent-seeking demands become too onerous, growth is stifled because the state is incapable of implementing decisions.

The first two possibilities I consider are analogous to either a predatory state or a rent-seeking business sector. The typical case is that some group or segment of society has far more access to power than others, as in quadrant III of Figure 1. When a country has a coherent state and a dispersed business sector, the result is predatory behavior by the state (top-down behavior) where political elites can scrape off rents in a predatory manner. Political elites presiding over a coherent state will have the opportunity to take advantage of a fractured business sector.

Alternatively, when the business sector is concentrated and the state is fragmented, as in Cell II, rent seeking results (bottom-up behavior). Here rents created by the state flow to business, because the latter has colonized the former and transformed it into a sort of "executive committee." A business sector composed of strong interest groups may overwhelm the state with its various demands, leading to either policy incoherence or policy indecision. Many analyses of developing countries emphasize that the state is a relatively recent, and hence weak, addition to the political scene. Strong interest groups may be able to capture control of the state and use the power of the state for their own ends.

Two other possibilities exist. In quadrant IV there are numerous interest groups and diffuse power within the state. In this situation, no single group could have too much influence, and the "political market" would come close to clearing. This builds on Susan Rose-Ackerman's notion that "the role of competitive pressures in preventing corruption may be an important aspect of a strategy to deter bribery." 22 When both state and business are weak, rents are all but eliminated. Neither state nor business is powerful enough to take advantage of the other, and so exploitation is difficult. Many of the advanced industrial democracies--at least when compared with less-developed countries--may approximate this situation. As bureaucrats compete with one another to offer policy, thus driving the cost of a bribe toward zero, numerous capitalists also compete with one another for the policy, also driving the price toward zero. Corruption is lowest in quadrant IV.

The final and most interesting case is quadrant I, where government and business are equally strong: there is a relatively coherent state but also a small number of powerful interest groups. In this case, the level of rents is limited and the division is relatively equitable. The result is "mutual hostages," where the state and powerful groups may collude with one another but neither has the advantage. Quadrant I reflects the old saw, "If you owe the bank a little money, the bank owns you. If you [End Page 183] owe the bank a lot of money, you own the bank." This middle position exists when the state is relatively coherent but the number of powerful interest groups is small. In this mutual hostage situation both political and economic elites are powerful enough to harm each other, but the damage each can inflict on the other deters such actions. 23

In quadrant I, rents can be had and corruption can occur, but the level of rents is constrained by the power of the other group. Small-N (business concentration) reduces transaction costs, and hence rent-seeking, because a small-N eases monitoring and enforcement costs. In this situation, although both business and state can earn rents, the amount will be less than in the polar cases where one group dominates the other, and more than in the case where both groups are dispersed into a large number of small actors.

In this sense, strategic interaction between state and business corresponds to a Prisoner's Dilemma, and cooperation is not automatic. While in the short run either actor may be better off defecting and gaining all the rents for itself, the other actor retains the ability to punish defection over time, and thus grudging cooperation may ensue. 24 It is possible that quadrant I could lead to a mutual war of attrition, with the two sides slugging it out. However, even without active cooperation, the power of the other side will limit exploitation. In Korea, Park initially tried to take advantage of the business sector but was unable to do so. 25

Thus, the least corruption would occur in situations where both state and business are weak and disorganized, for neither group could take advantage of the other, and all the groups would compete against each other, driving the price of corruption close to zero. The most corruption would occur when one side, either state or business, is coherent. A middle position exists when both state and business are strong and can take partial but not total advantage of each other. 26


The theoretical concepts in this article are widely accepted and used in the social sciences, even though the difficulty in measuring them is also widely acknowledged. 27 Recognizing this difficulty, in this study I focus on the polar cases, in an attempt to lay out the ranges of the variables and test the model's plausibility.

Measuring the dependent variables of corruption and influence-peddling is difficult. By their very nature these are acts that actors wish to keep hidden. Although scholars have used polls of perceptions of corruption as evidence, have [End Page 184] traced a single pattern of corruption, or have relied on corruption scandals, there is no comprehensive indicator of corruption. 28 While conclusive measures do not exist, a variety of indicators give us a sense of the size and pattern of corruption. 29 Occasional scandals reveal the pattern of influence. Estimates of campaign spending, kickbacks, and secret funds are useful first approximations. Since precise measurement is difficult, in this study I assess only whether corruption is low, medium, or high.

Measuring the independent variables is only marginally easier. To measure the strength of the business sector, I focus on a series of indicators. Firms' value added as a proportion of gross domestic product (GDP) indicates their market and political power, and the composition and concentration of bank loans indicates their vulnerability to the state. To measure state coherence and low agency costs, I rely mainly on detailed case studies that follow the process of policymaking, and I do not attempt to provide a single quantifiable measure for the variables. Case studies can reveal whether leaders act on their parties' politics and domestic politics, or whether they respond to them. Process tracing of both policy decisions and institutional origins can reveal whether there is agency slack between leaders and bureaucrats.

Money Politics and the Developmental State

The series of institutional changes made under the Park Chung-hee regime (1961-79) is often used as the starting point of the high-growth era. Yet the Park regime was hardly de-politicized, and in fact money politics was pervasive. Not only was there extensive corruption, but political connections overrode economic criteria and allowed for overcapacity and bail-outs of indebted and poorly managed firms. The basic process was simple: Business and political elites exchanged bribes for political favors. Politicians used these political funds to buy votes and to serve basic greed. Businessmen used the rents from cheap capital to expand as rapidly as possible, thus ensuring their continued political and economic importance. Development and money politics proceeded hand in hand.

Politicians:BR> The Demand for Political Funds

Under Park the need for funds for electoral and political purposes was extensive. The election laws themselves, imposed by the military junta, restricted political activities so severely that, as Alexander Kim notes, "no party could be effective unless it had many wealthy members, or unless it could secure secret, illegal [End Page 185] donations--something the ruling party could do, but which an opposition party would find immensely difficult." 30 These donations were known as bi-chagum, or "secret funds." The DRP (Minju Gonghwa Dang, or Democratic Republican Party) not only required a large staff and budget but also bought votes through the traditional huwonhoi (personal vote) system--throwing parties, creating hiking clubs, and attending weddings and holiday festivities. Park and the DRP organized society through a number of means. The largest was the People's Movement for National Reconstruction (Kuk-To Gonsol Dan). Units were established down to the village level throughout Korea, and at one point the organization was reported to have 3.9 million members. 31 Park's coalitional base consisted of an extensive and expensive party apparatus that organized and controlled society and provided side-payments to Park's constituents. While the DRP won the southern and rural districts, the opposition tended to win in the cities, a phenomenon that came to be known as yochon-yado ("The ruling party is strong in the provinces, the opposition is strong in the cities."). As the Korean economy expanded, money began to slosh through the system at an unprecedented rate.

According to the financial report of the Central Election Management Committee, the annual costs of running the DRP were as high as $4.5 billion in the 1960s, or $35 million won (at 130 won to a U.S. dollar) in 1965. 32 The DRP secretariat alone was estimated to have 1,300 staff members, and the estimated cost of operating just the secretariat was $700,000 per month. 33 During the 1963 election, for example, the DRP was estimated to have spent 76.9 percent of total campaign spending, and during the 1967 elections an estimated $40 million was spent buying votes. 34

To fund their operations, political elites took massive donations from the chaebol in return for loans and sweetheart deals. Under Park this financial system of exchanging policy for bribes became quasi-institutionalized. Leading members of the DRP were in charge of political fundraising, the two most important persons being Kim Sung-Kon and Kim Jin-man. 35 These members of Park's inner circle had clear fundraising duties: One dealt with personal connections, another with the parties, and others with big business.

Businessmen often called "voluntary" donations jun joseh, or "quasi-taxes." 36 For example, the Saemaul Undong was a 1970s political organization whose aim was to improve the quality of rural life, but it was also a home for embezzlement, nepotism, and cronyism. Hyundai donated 7.4 billion won to the cause in 1971-75, Samsung [End Page 186] donated 7.8 billion, LG 6.7 billion, and Daewoo 4.85 billion (Table 1). Even if the chaebol donations were used for the intended humanitarian purpose, such donations were certainly part of the larger web of money politics. The fact remained that if businessmen did not provide politicians with sufficient funds when asked, the Bank of Korea called in their loans, or they suffered a tax audit, or their subsidy application was denied. The best example of this is Kukje's refusal to make "voluntary" donations to the Ilhae foundation, and Chun Doo-hwan's subsequent dismemberment of the company. 37 In 1985 the Kukje group, with 38,000 employees, was the seventh-largest chaebol in Korea and, like all chaebol, was highly leveraged with significant loans from the government. However, Kukje's president, [End Page 187] Yang Chung-mo, had seriously offended President Chun Doo-hwan by refusing to contribute significant sums to quasi-governmental organizations such as the Ilhae Foundation 38 and the Saemaul Undong ("New Village Movement"). As a result, the state refused to loan money to Kukje and refused to honor its checks. Within weeks Kukje could not service its debt and had to declare bankruptcy. 39

Distributive coalitional politics was one major reason for the prevalence of political funds. However, political funds had another major "use": to satisfy greed. After Park's death, half a million dollars were found in his personal safe. Kim Jong-pil reportedly amassed more than $50 million (at 1979 prices), including a 5,000-acre ranch, a tangerine orchard, a 2,100-head dairy farm in Chungchong-do, a newspaper company in Seoul, and over 3.7 billion won ($36 million) in real estate. Kim also reportedly had 4 billion won in thirty-four secret bank accounts, a 1.1 kilogram gold sword, a "huge collection" of antiques and jewelry, and $2.5 million in "cooperation fees" from companies. 40

Lee Hu-rak, who had been the presidential chief of staff, Blue House, accumulated $40 million. Oh Won-Chul, the architect of the Heavy and Chemical Industrialization Plan (HCIP), accumulated wealth of at least $4.5 million. Former Army Chief of Staff Lee Se Ho had $18.5 million, much of it from embezzling government funds. Former Deputy Speaker of the National Assembly Kim Jin-man had $17 million in assets; Kim Jong-pil's brother Kim Jong Nak had $15 million, presidential bodyguard "Pistol" Park Chong-kyu had $12.8 million, and former prime ministerial aide Lee Pyong-Hi had $4 million. 41 U.S. congressional staffer Edward Baker remarked after the 1977 investigation into Korean-American affairs, "One has to conclude that either Park Chung-hee was an idiot or he was condoning outrageous criminal behavior. Whether Park was putting money in his pocket or not, he knew he was heading a criminal enterprise." 42

The Supply of Funds in Return for Favors

Given the Korean state's total control over the financial sector in the 1960s and 1970s, businesses were naturally interested in gaining access to the enormous rents that accrued to a chaebol if it received a low-interest-rate loan. The state's inability to control firms and their growth led to endemic overcapacity. Firms rushed willy-nilly to expand at all costs, whether or not it was economically feasible. The result was that in most major sectors of the economy there was excess capacity and overlapping and duplication of efforts as each chaebol tried to be the biggest.

Rents in the form of U.S. aid, allocation of foreign and domestic bank loans, import licenses, and other policy decisions were based on a political funds system [End Page 188] that required donations from the capitalists. During the 1960s, the expected kickback became normalized at between 10 and 20 percent of the loan. 43 Park Byung-yoon points out that as early in Park's rule as 1964, 38 percent of total bank loans--43 percent of M1 money supply--was given to only nine chaebol, all of which had family members in powerful positions in the ruling party or in the bureaucracy (see Table 2). 44

Indeed, whether or not a market rationale existed for expansion of firms' activities, there certainly was a rationale for rent seeking when combined with the proper political connections. The Park era, far from limiting and controlling the expansion of chaebol, saw the opposite result. Firms rushed headlong into expansion, both to justify their continued receipt of cheap money and to make themselves so large that the government would have no choice but to keep supplying them with funds--a situation known as "moral hazard." Table 3 shows the year of incorporation or establishment of subsidiary companies for the ten largest chaebol in 1984. Sixty percent of chaebol expansion occurred during the 1970s, resulting in tremendous overcapacity.

The car industry serves as a good example. Despite numerous attempts to rationalize the automobile industry and force domestic firms to concentrate on core competencies, Korea throughout the 1970s had more capacity and more players than the Economic Planning Board (EPB) considered economically feasible. In 1969 Hyundai Motors operated at 49 percent of capacity, and in 1972 at only 25.8 percent [End Page 189] of capacity. The rest of the industry was no better. Finally, between August and October of 1980, on an EPB initiative, the Korean government ordered the automobile industry to "merge by decree." The plan was to merge Hyundai Motors with Daewoo's Saehan subsidiary and produce passenger cars, while forcing Kia Industries out of the passenger car market, but the plan failed. 45 Park Byung-yoon wrote that industry saw the government as ineffective, and "the government could not have done anything to make the business community comply with its program." 46

The incentive to become heavily indebted and to focus on expansion instead of efficiency had predictable results: Firms borrowed whether they needed to or not. Many firms expanded far too quickly and without adequate management expertise or planning. But because there were so few chaebol, it was politically dangerous to allow them to fail. As a result, the Korean state did bail out weak companies, and it rewarded political relationships, not necessarily economic success. Far from imposing performance standards, the Korean government was continually forced to bail out inefficient firms that had overextended themselves.

Indeed, as far back as 1971 the United States was alerting Korea to the problems it faced because of the personalistic manner in which it disbursed the foreign aid and [End Page 190] loans it received. At a 30 March 1971 meeting between officials from the Economic Planning and the U.S. Treasury and State Department the U.S. officials stressed that "insolvent industries, caused by political favoritism in earlier years, represented a major problem. They stressed that foreign banks looked closely at the prevalence of favoritism towards basically unsound companies." 47 However, Korea took no action until 3 August 1972, when, instead of rationalizing poorly performing and heavily indebted firms, Park Chung-hee engaged in a massive bailout of those companies. Designed to alleviate the curb market, the bailout disproportionately helped more heavily indebted firms. The decree essentially placed a moratorium on new loans, and old loans were rescheduled to be paid back over five years, after a three-year grace period, with a substantially lower interest rate. 48

Extensive political connections overrode economic considerations. In the context of an authoritarian regime that handed out favors selectively, size was an advantage. A capitalist who had already acquired a license to invest in a project was at a great advantage in acquiring further credit allocation from the government. Since only a small number of capitalists had shown their ability to produce in the past, they stood out. 49

Mutual Hostages

Money politics remained constrained because Korean elites existed in a mutual hostage situation where neither political elites nor economic elites could take excessive advantage of the other. This balance of power allowed them to pursue corrupt activities, but it also limited the chances for excessive advantage. Government intervention was subject to political influence in a number of ways that reduced both rent-seeking by entrepreneurs and transaction costs for the politicians and bureaucrats involved in monitoring the policy process.

While the "strong state" has been the focus of much of the literature, the surprising strength of the business sector has received less attention. State control over the financial sector had enormous consequences for the organization and conduct of business in Korea. 50 Since Korean companies were highly leveraged, they were vulnerable to state control. Paradoxically, this weakness became a source of strength in relation to the state. The Park regime actively encouraged the centralization and enhancement of economic power in the chaebol. From Park's initial decision in 1961 to pardon the "illicit wealth accumulators" to the bailout of highly leveraged firms on 3 August 1972, to the 1976 decision to promote General Trading Companies, 51 state policy has continuously encouraged the rise of the chaebol. [End Page 191]

If the state has control over various policy instruments, can control or manipulate the judiciary and the legislature, and can redesign at will the terms of any agreement it makes, the problem will be that no action it takes will be credible. The farsighted ruler who realizes that he is at risk will take action to infuse greater confidence. One way the state can make a believable commitment to various policy initiatives is by fostering domestic sources of power in specific areas--power centers that would later prove to be tremendously costly to overturn.

By encouraging the formation of large conglomerates that accounted for large percentages of the Korean economy, the state and the chaebol in effect became mutual hostages (quadrant I of Figure 1). Tables 4 and 5 show that while each needed the other, neither was able to gain the upper hand, and thus state and chaebol were forced to work together. 52

Korea under Park approximated quadrant I of Figure 1 quite closely. The government's control over capital made any threat to cut off credit to the chaebol lack credibility, because all actors knew that it would hurt the regime as much as it would hurt the chaebol. A flow of political payoffs to politicians strengthened this pipeline of easy money. For politicians, the need for political funds to run elections and other political activities, as well as a natural proclivity toward greed, gave the government no incentive to sanction companies unless the business managers were [End Page 192] utterly incompetent. Bad loans to good friends were supported by external aid and loans. What kept the process from spinning out of control was the balance of power between political and economic elites. While both benefited from the close ties, neither was able to dominate the other. This is analogous to a Prisoner's Dilemma: Both sides would prefer to defect, but they are restrained from doing so by the other side's ability to respond in kind, and hence a form of stability ensues. In Korea the exchange of such hostages led to an uneasy balance of constrained collusion that existed throughout the period of high growth.

Democracy Increased the Role of Money Politics After 1987

The model proposed here is dynamic, and the initial configuration of state and business sets out the development path. Over time this path itself leads to changes in both the composition and the power of the actors involved. The chaebol moved quickly up the technology chain, diversifying and becoming larger and more complex. An increasingly wealthy middle class, long chafing under the collusive ties between state and chaebol, now came to the fore as a politically influential group. However, the most significant change was the 1987 democratic transition. 53 A country's shift from authoritarian institutions to democratic ones will have different results depending on the relationship between state and business (Figure 2). In Korea, where both the state and business were strong, a shift to democratic institutions benefited business more than the state--the state was weakened by the imposition of democratic processes. Democratization does not change the business sector's generally high demand for rents, but it does affect the supply. With more politicians competing on the supply side, fewer limits were placed on the behavior of the business sector.

Korea's position moved from one of government and business as mutual hostages (quadrant I) to one of predominance by rent-seeking business (quadrant II). Democratization led to increased demands for political payoffs, as politicians began to genuinely compete for electoral support, and to a decrease in the state's ability to resist or contain the demands of the business sector. The small number of massive Korean firms, unrestrained by domestic market forces because of their size, made increasingly risky decisions.

The democratic transition had two major implications for the pattern of money politics in Korea. First, an increase in demand for political payoffs shifted the advantage to business. Second, Korea's legal and corporate institutions remained underdeveloped even in the 1990s and allowed a continued business focus on debt-led investment in overcapacity and overdiversification. [End Page 193]

Increased Demand for Political Payoffs

The scandals of the late Kim administration showed that the transition to democratic rule by no means reduced the ability of rent-seeking groups to exercise political influence; indeed, the demand for campaign funds and secret funds (songo-chagum and bi-chagum) has probably increased. One reason that Kim Dae-jung and Kim Young-sam are no different from previous political elites in their manner of political fundraising and their appetite for it is that the costs of winning elections and running a party are immense.

Table 6 gives estimates of spending on elections from 1981 through 1997. Most striking is the vast expansion in spending after 1987. During the 11th National Assembly election (1981) under Chun, a popular phrase was "ship-dang, gurak," which roughly translated means "Expenditures of 1 billion won [$1.3 million] win the seat, expenditures of 900 million won [$1.2 million] will lose." 54 By the 14th National Assembly election (1992), the phrase had become "iship-dang, shimnak," or "Expenditures of 2 billion won [$2.6 million] win the seat, expenditures of 1 billion won [$1.3 million] will lose." 55

The total cost of the National Assembly and presidential campaigns of 1992 was estimated at 5 trillion won ($5.1 billion) or the equivalent of 16 percent of the government's annual budget. 56 However, in twenty days of legal campaigning, National Assembly campaign expenditures were capped at 83 million won ($105,000). For the 1996 National Assembly campaign it was estimated that each candidate spent 1.2 billion won ($1.5 million) during that period, not including expenditures before the legal campaign. In addition, the cost of running an office [End Page 194] over four years is estimated to be 4.8 billion won ($6 million)--so a winner needs 6 billion won ($7-8 million) just to win and operate a normal National Assembly seat. 57

Normal party expenditures are also enormous. The Korean phrase oribbal ("duck feet") describes the situation: On the surface decorum is maintained, while below the surface the parties scramble like mad to raise funds. In the 1970s Park's Blue [End Page 195] House disbursed approximately 10 billion won ($20 million) each year. 58 After democratization, each local district had expenses of approximately 10 million won ($13,000) per month, meaning that annual party expenditures in the early 1990s were approximately $39 million. 59 Over five years, Roh reportedly gave at least 180 billion won ($240 million) to his political party for normal operating expenses and disbursed 30 billion won ($40 million) to local constituencies four times a year, while personally raising at least $800 million in secret funds. 60 In 1996 political parties officially reported raising 315 billion won ($420 million). 61 As one observer noted, "if they admit to raising 315 billion won, we know the actual total must be far greater." 62 In comparison, in the United States, a country with six times the population and an economy fifteen times larger, the Republican party's fundraising goal in 2000 was $179 million.

Political funds in Korea came from business. Tables 7 and 8 compare estimates for quasi-taxes paid by business for the periods 1980-87 and 1994-98. Although quasi-taxes are largely accountable, they are still imposed in a coercive manner and are part of the overall government-business relationship. To not make "voluntary" donations is to risk payback in the form of tax audits or rejected loan applications, for example. Total estimated quasi-taxes for 1980-87 were 3.9 billion won, and the estimated total for 1994-98 was 4.2 trillion won. Part of this increase is due to the increased size of both the economy and the largest chaebol, but neither the economy nor the chaebol expanded a thousandfold over the decade. Additionally, the figures for quasi-taxes include neither entertainment expenses nor individual donations. According to the National Tax Administration, between 1988 and 1992 the top [End Page 196] thirty chaebol spent approximately $1.4 billion for banquets and entertainment and only $215 million for research and development. 63

Indeed, Hyundai founder Chung Jo-yong said, "I personally handed to the ruler about 1 billion won yearly during the 3rd Republic [Park], about 5 billion won yearly during the 5th Republic [Chun], and 10 billion won yearly in the 6th Republic [Roh]." 64 With the coming of democracy, Roh Tae-woo's regime had by 1988 instituted a system in which over 200 dojang ("stamps," or bureaucratic approvals) had to be obtained in order to undertake any project in South Korea. Thus, even the smallest bureaucrat now had veto power, and the scale of the economy was much larger. As one businessman lamented to me, "By the late 1980s Roh and later YS [Kim Young-sam] had established so much 'democracy' that I needed over 100 envelopes [bribes] in order to build a factory last year. That never occurred under Park or Chun--they eliminated the middleman, and while you had to pay for access, you could do it at the top levels, and not worry so much about the bureaucracy." 65

Under Kim Young-sam the trend continued. By the time of the collapse of the chaebol Hanbo in 1997, between 1.5 and 2 trillion won (over $1 billion) had disappeared from Hanbo's accounts, most likely ending up in the pockets of political or business elites. 66 In 1999 the market research firm Taylor Nelson Sofres PLC polled 1,250 middle-class, upper-income Korean decision makers, mostly white-collar or self-employed urban men. Of the five categories considered--entertainment, [End Page 197] law enforcement, government officials, business, and education--respondents said that only the education sector was less corrupt than it was five years earlier. For business generally, 42 percent of respondents said the level of corruption was the same as before, and 32 percent said it was worse. 67

This exchange allowed greater business influence in policymaking, and the chaebol continued to expand at the expense of small and medium-sized industries. Although many assumed that globalization and liberalization would reduce rent seeking and the power of the chaebol, the opposite might very well be the case. Table 9 shows that although in 1986 the four largest chaebol added 5.7 percent to Korea's GNP, by 1995 their share had grown to 9.3 percent of value added to GNP.

Unless liberalization is matched by stringent regulatory oversight that limits collusive practices and the exercise of market power, it can provide new opportunities for large firms to buy favorable policy. While measures to rein in the chaebol are popular politically, because of government-business ties such policies were unsustainable even after 1997.

Weak Legal and Corporate Institutions

Not only did the number of opportunities for business to influence government increase after democratization; the importance of personal relationships (inmaek) in legal and corporate institutions also increased. A historically weak legal environment--and the corresponding importance of personal ties--creates an environment where the founder/chairman can control a vast array of subsidiaries while having little or no formal title to them and can evade or influence government policy. 68

Within this institutional environment, the rule of law in Korea has been vague and has seldom been enforced. Until 1997, chaebol regulations consisted of a credit [End Page 198] control system, the Monopoly Regulation and Fair Trade Act (MRFTA), and other policies. Direct cross-shareholdings between any two subsidiaries of each of the thirty largest conglomerates, as well as cross-loan guarantees, were prohibited. However, legislation by itself is no guarantee of solid institutions, nor does it guarantee equal treatment under the law. Kim Jong-Seok notes that

Korean regulations are not just pervasive and large in number, but also highly judgmental and vague, so that most of the decisions and interpretations of the regulations are left with the regulators themselves. . . . The subsequent opaque procedures, ambiguous rules, and unpredictable results create total instability for businesses in Korea. Under such regulatory circumstances, both regulators and regulatees know that the regulation cannot be enforced as written, and thus the regulatees have a strong incentive to lobby the regulators to circumvent enforcement. 69

As an example of how loose Korean corporate governance has been, it was only in 1993 (for bank accounts) and 1995 (for real estate) that the government outlawed the system of "false names." Since the 1960s, it had been possible to register a false name (a grandmother, "Mr. Kim," and so on) and use it to buy stock, engage in land speculation, and transfer money, thus avoiding taxes and circumventing regulations that prohibited land speculation and imposed corporate shareholder limits. Indeed, when this system was outlawed by then-president Kim Young-sam, there was a horrific outcry from the business community. The Ministry of Finance estimated that in 1993 politicians and businessmen used over one million false names, accounting for over 30 trillion won ($40 billion) in assets.

In this fluid institutional environment, personal ties between chaebol and politicians--always important--have become even more critical to business success. The transition to democracy did not change this need. Rather, the 1990s saw expanded opportunities for personal connections, influence peddling, and a "bigger is better" mentality. Business concentration continued to increase, while cross-holding ownership remained a standard Korean business practice.

Instead of being limited to their "core competencies," the chaebol increasingly overlapped in their efforts in the automotive, electronic, and financial sectors. Why did this happen? Because neither Kim Young-sam nor Roh before him coupled their executive orders with realistic incentives for the chaebol to comply. Banks would not invest in any firms aside from the chaebol, which were the only internationally competitive domestic firms. In addition, the rest of the government bureaucracy--which had been knee-deep in big business-government relations--was also not convinced that a policy of limiting the chaebol would truly be successful. Thus, because there has been no credible shift in government policy, traditional government-business practices have continued.

This has served the political purposes of growth and protection. Much of the pressure for "structural adjustment," therefore, stems from this practice rooted in the [End Page 199] chaebol system. Table 10 shows the levels of indebtedness in the Korean economy in 1996.

Thus, the democratic transition led politicians to increase their demand for political funds and hence led business to exert greater influence. Rapid expansion was politically useful. Bigger has always been better in the Korean context, both as a means of ensuring a continuous flow of funds to the companies and as a way for entrepreneurs to protect themselves from the vicissitudes of Korea's fluid political sphere.

Corruption and the Developmental State

Money politics has always been a central aspect of the pattern of Korean politics, and this study has shown that we cannot assume benevolence on the part of the state. This study has also cast doubt on the "hard" version of the developmental state. Evidence of massive corruption undermines the argument that the state is neutral, picks winners, and provides public goods because the civil service is insulated from social influences. However, a more limited version of the developmental state argues that governments can have a beneficial effect regardless of how the politics of government action is attained. Korea did develop, and numerous studies have shown how the Korean state acted in a number of developmental ways to provide public goods and nurture investment. So how do we reconcile the politics and the economics?

Money politics did not swamp development for three reasons. First, the mutual-hostage situation in Korea was a key factor in keeping corruption from spinning out of control. The bargain that the elites struck was collusive, not cooperative, and each group took as much advantage of the other as possible. Yet the balance of power [End Page 200] meant that neither the political elite nor the bureaucratic elite was able to gain a decisive edge over the other. The process never spun out of control, because the elites were vulnerable with respect to each other, and they were thus limited in their ability to force their wishes on the other side. In addition, the smaller the number of actors, and the less competitive the rent-seeking process, the lower the total costs arising from the process. 70 Competition for rents is not so great in situations involving few actors, and thus resources are not necessarily expended in obtaining rents. Much rent seeking entails competing with other actors to win rents and then building entry barriers and other mechanisms to protect those rents. A group that has exclusive access to the rent markets has lower information costs and can collude over time with other rent seekers to lower costs. 71 The implication from this is straightforward: Having smaller numbers of rent seekers reduces the total social cost because property rights over the rent are more secure.

Second, bribes are transfers. As such, corruption does not necessarily imply any deadweight loss, and the political story I have told here does not necessarily affect the overall provision of public goods. 72 Corruption may indeed consist of struggles over the distribution of state policy and goods rather than struggles over the absolute level. Rent-seeking models assume a perfect "state of nature," where the government does not interfere with economic activity. Yet in reality, every society already has a distribution of rights that benefits some actors more than others, and thus bribery related to rent seeking is itself a mechanism that could lead to a more efficient distribution of rights than before. The relevant question is whether the resources would be put to more productive use by the seeker or by the one being sought, and theoretically only the lowest-cost firm could afford the highest bribe. 73 None of this necessarily entails rent seeking that would distort economic efficiency: If the transfer of wealth is from businessmen to politicians and it results in productive investments, a nation may benefit even from machine-style politics. 74

Finally, although the Korean state may have provided public goods and supported investment, that may not be why those goods were provided. The Korean state intervened in the way it did because to do so was in the interests of a small group of business and political elites. Building roads, apartment complexes, and power stations provides some public goods, but it also provides private goods. Access to the private benefits of state resources was often contingent upon production of public goods. 75 Although Samsung and Daewoo accrued enormous private benefits from having privileged access to state capital and policies, society benefited as well from improved infrastructure, employment, and opportunities. My argument is complementary to Alice Amsden's argument that the state exchanged subsidies for [End Page 201] performance. 76 The difference is that I provide a political story that explains the patterns of exchange, and I also show that the process was based on money politics, was heavily biased, and was not nearly as efficient as Amsden argues.

Thus, this study suggests that understanding even a developmental state requires knowing the political story. The Korean state made many decisions that fostered investment and growth over the past forty years, but politics played a major role. Economic policy choice is only one of many issues with which political elites must concern themselves. A political-economy explanation must focus closely on how preferences emerge and on the institutions that direct the implementation of those preferences. Such an explanation must be sensitive to the nature of the political coalition that supported the elites, to resource constraints, and to the vulnerability that elites faced. Given that elites must constantly be concerned about retaining power, corruption and policy access can be powerful political tools. 77

There are four main implications that arise from this analysis. First, the evidence presented here shows that knowing the political story is essential to understanding how the developmental state functions. I have shown how the needs of the elites drove policymaking, created enormous graft, and patterned the political and economic systems. It is this larger institutional environment, more than any specific institution or policy, that previous scholarship has neglected.

Second, microanalytical explorations of political economy must increasingly be balanced with an understanding of the larger institutional environment within which actors operate. Focusing too closely on the details of party organization or institutional configurations of the Economic Planning Board may cause scholars to lose sight of the larger forces at play. Both approaches are valid, and they complement each other.

Third, because the institutional arrangements and policy choices in individual countries vary greatly, we must also avoid the temptation to look for a universal set of arrangements or variables that will lead to growth or stagnation. This is not to argue that development is a random process. We have a good understanding of some of the variables, such as savings and education. But because Korea is one of the few successful late developers, scholars need to carefully explore the role of less analytically tractable variables, such as corruption. The evidence presented here suggests that in order to understand the nature of policymaking in developing countries, we must first understand the particular political challenges that individual leaders face and the manner in which business attempts to influence government policy. The strategic allocation of economic policy and benefits is an important political resource.

Finally, my argument raises a number of new questions regarding the developmental state and international relations. State-led industrialization transforms the international division of labor, and state autonomy is often understood as necessary [End Page 202] for this transformation. My argument, if it is correct, modifies our notion of state autonomy and makes the rise of Korea even more puzzling. The domestic focus of this article leads logically to the need to understand domestic-international linkages that may have sustained the Korean state or may have allowed it to behave as it did while still promoting growth. The current literature has emphasized an export-oriented trade regime as being important for the state and for development. However, I have shown that because Korea existed in a geopolitical situation that made its survival and growth important for larger U.S. interests, understanding the flow of international finance, the role of international institutions, and the role of dominant actors in the international system is also consequential for understanding Korea's development. A logical subsequent area for study is thus how and why these larger, and somewhat neglected, international forces interacted with domestic Korean politics. Growth with corruption is an issue that needs to be explained, and until scholars directly address the issue of politics, our understanding of the political economy of development will remain incomplete.


David C. Kang is Associate Professor in the Government Department and the Tuck School of Business at Dartmouth College, Hanover, New Hampshire. He can be reached at dave.kang@dartmouth.edu.


For valuable comments on an earlier draft, thanks to Bill Bernhard, David Lake, Michael Mastanduno, Jongryn Mo, Chung-in Moon, Sang-Young Rhyu, Kaja Sehrt, Dave Stuligross, and Dave Waldner. I especially thank Richard Doner and Bob Bullock.

1. For good overviews of the 1995 scandals, see Ahn 1995; and Kim 1995.

2. For representative views, see Schneider 1998; Evans 1995; Amsden 1989; and Johnson 1987.

3. The focus on state institutions includes Weberian bureaucracies that are autonomous from political and social interference. Among the many who hold this view, Peter Evans has argued that "highly selective meritocratic recruitment and long-term career rewards create commitment and a sense of corporate coherence." Evans 1995, 12. Alice Amsden also writes that "economic success in Korea challenges the assumption . . . that government intervention degenerates into 'rent-seeking.'" Amsden 1989, 327. For other specific instances, see Fields 1997, 126; Johnson 1987, 152; and Önis 1991, 114. The major policy focus is on export-oriented industrialization, where "in direct exchange for subsidies, the state exacts performance standards from firms." Amsden 1989, 146.

4. See North 1986; and Williamson 1991.

5. The statist approach implies that if an actor is capable enough, it can design an institutional structure that will resolve the tendency to shirk. Gary Miller (1992, 3) points out that this view is mechanistic and seriously underestimates the role of political leadership.

6. Williamson 1985.

7. Buchanan 1980, 3.

8. See Bardhan 1997; Braguinsky 1996; Nti 1997; and Barzel 1982.

9. See Chang 1994; Rose-Ackerman 1978; Shleifer and Vishny 1993; Levi 1988; Milgrom and Roberts 1990; Bag 1997; and Khan 1996a,b.

10. Corruption is thus a subset of rent seeking. Rents may be allocated purely on the basis of merit, or they may be allocated toward bribes.

11. Shleifer and Vishny (1993) model different types of government structures, but they avoid studying how business organization may affect corruption. Susan Rose-Ackerman (1978) discusses types of state organization similar to those presented in my work, although she does not use the terminology I employ here.

12. On autonomy and state strength, see Katzenstein 1978. On internal control (agency costs), see McCubbins and Schwartz 1984; and Ramseyer and Rosenbluth 1993, chaps. 6 and 7.

13. Spiller 1990.

14. Shafer 1994.

15. Amsden 1989, 8.

16. Schneider 1998, 109.

17. On Latin American conglomerates, see Weyland 1997.

18. Levi 1988.

19. Tilly 1985.

20. For an interesting discussion along these lines, see Milgrom and Roberts 1990; and Chang and Kozul-Wright 1994.

21. Migdal 1988.

22. Rose-Ackerman 1988, 278. See also Bliss and Di Tella 1997; and Shleifer and Vishny 1993.

23. This is based on the idea of bilateral monopoly. See Kreps 1990, 551-73. See also Blair, Kaserman, and Romano 1989.

24. Axelrod 1984.

25. The 1961 "Illicit Wealth Accumulation Act," or bujong chukje-an, is an example of this. For more, see Kang forthcoming.

26. The evidence in this article fills in quadrants I and II. For comparative work that uses the Philippines to examine all four quadrants in the matrix, see Kang 2002.

27. See Krasner 1978, especially chap. 1.

28. Daniel Treisman (1997) uses the Transparency International index of perceived corruption as his measure of actual corruption; Robert Wade (1985) traces the sale of office in India; and Chalmers Johnson (1986) follows the corruption scandals in Japan. Other theoretically informed empirical work includes Morris 1991; Doner 1997; Kiser and Tong 1992; Levy 1995; and Wedeman 1999.

29. See Hutchcroft 2000; and Sandholtz and Koetzle 1998.

30. Kim 1975, 245.

31. Kim 1971, 158.

32. Throughout this article I have intermixed both Korean won and the U.S. equivalents. I use U.S. dollars to provide some sense of the scale and Korean won to give the specific amount.

33. See Park 1967; and Kim 1975.

34. Kim and Park 1968, 88.

35. Lee 1987.

36. Although much of these quasi-taxes was accountable and did not derive directly from corruption, the taxes were part of the government-business relationship and hence are important to understanding the political process.

37. Kang 1995, 560-61.

38. The name given to this research foundation, created by Chun Doo Hwan, was Chun's pseudonym.

39. See Clifford 1994, 208-12; Christian Science Monitor, 28 March 1985, 23; and Wall Street Journal, 2 February 1985, 32(W).

40. See Martial Law Enforcement Headquarters 1980; Chosun Ilbo, 19 June 1980; and Oh 1986, 42.

41. Martial Law Enforcement Headquarters 1980.

42. Clifford 1994, 89.

43. Woo 1991, 108.

44. Park 1982, 210.

45. Lew 1992, chap. 5. See also Choi 1987, 129-30.

46. See Park 1979; and Lew 1992, 176.

47. U.S. House of Representatives 1978, 121.

48. See Kyunghyang Ilbo, 31 August 1991; and Economic Planning Board 1983, 18.

49. Park 1967, 207.

50. Jones and Sakong 1980.

51. Cho 1987.

52. Lew 1992.

53. Haggard and Kaufman 1995.

54. Park 1989, 290.

55. Ku 1999, 57.

56. Yang 1997, 31.

57. Park 1992, 271.

58. Han and Ahn 1994, 205.

59. Yoon 1995, 138.

60. See Yoon 1995, 138; and Ha 1998, 230.

61. Donga Ilbo, 12 April 1998, 4.

62. Ibid.

63. Coalition for Social Justice/Citizen's Coalition for Economic Justice 1993, 272.

64. Chosun Ilbo, 9 January 1992, 3

65. In Korea bribes and honoraria are traditionally passed to their recipients in white envelopes. See also Ha 1998, 237-39.

66. Yoon 1997, 201.

67. Wall Street Journal, 31 December 1999.

68. Donga Ilbo, 27 March 1998, 2.

69. Kim 1996, 11.

70. Tullock 1988, 228.

71. Chang 1994, 120.

72. Khan 1997, 13.

73. Bardhan 1997, 1322.

74. See Khan 1996a,b; Doner and Ramsey 1997; Shleifer and Vishny 1993; Theobald 1990; and Klitgaard 1988.

75. See Olson 1982; and Broz 1997, 2-7.

76. Amsden 1989.

77. See Jomo and Gomez 1998; Weingast 1993, 287; and Root 1994.


Ahn, Byoung-yong. 1995. Pichagum kwa taekwon yokui chuakhan janch'i (The Disgusting Feast of Illicit Funds and Presidential Hunger for Power). Shindonga (December):112-21.

Amsden, Alice H. 1989. Asia's Next Giant: South Korea and Late Industrializations. New York: Oxford University Press.

Axelrod, Robert. 1984. The Evolution of Cooperation. New York: Basic Books.

Bag, Parimal Kanti. 1997. Controlling Corruption in Hierarchies. Journal of Comparative Economics 25 (3):322-44.

Bank of Korea, Statistics Division (Chosabu). 1998. Uri nara kiop ui puch'ae kujo punsok (An Analysis of the Structure of Korea's Indebted Companies). 21 October.

Bardhan, Pranab. 1997. Corruption and Development: A Review of Issues. Journal of Economic Literature 35 (3):1320-46.

Barzel, Yoram. 1982. Measurement Cost and the Organization of Markets. Journal of Law and Economics 25:27-48.

Blair, Roger D., David L. Kaserman, and Richard Romano. 1989. A Pedagogical Treatment of Bilateral Monopoly. Southern Economic Journal 55 (4):831-41.

Bliss, Christopher, and Rafael Di Tella. 1997. Does Competition Kill Corruption? Journal of Political Economy 105 (5):1001-23.

Braguinsky, Serguey. 1996. Corruption and Schumpeterian Growth in Different Economic Environments. Contemporary Economic Policy 14 (3):14-25.

Broz, J. Lawrence. 1997. The International Origins of the Federal Reserve System. Ithaca, N.Y.: Cornell University Press.

Buchanan, James. 1980. Rent Seeking and Profit Seeking. In Toward a Theory of the Rent-Seeking Society, edited by James M. Buchanan, Robert D. Tollison, and Gordon Tullock, 3-15. College Station: Texas A and M University Press.

Chang, Ha-Joon. 1994. The Political Economy of Industrial Policy. New York: St. Martin's Press.

Chang, Ha-Joon, and Richard Kozul-Wright. 1994. Organising Development: Comparing the National Systems of Entrepreneurship in Sweden and South Korea. Journal of Development Studies 30 (4):859-91.

Cho, Dong-song. 1987. The General Trading Company: Concept and Strategy. Lexington, Mass.: Lexington Books.

Choi, Byung-sun. 1987. Institutionalizing a Liberal Economic Order in Korea. Ph.D. diss., Harvard University.

Choi, Dong-gyu. 1986. Kiopui chun jose pudame kwanhan silchongchok yonku (Concerning the Burden of Quasi-taxes on Companies). Seoul: Korea Economic Research Institute.

Choi, Sung-no. 1997. 1996 nyon 30 dae kiopchipdan (An Analysis of the 30 Korean Conglomerates in 1996). Seoul: Korea Economic Research Institute.

Clifford, Mark L. 1994. Troubled Tiger: Businessmen, Bureaucrats, and Generals in South Korea. Armonk, N.Y.: M.E. Sharpe.

Coalition for Social Justice/Citizen's Coalition for Economic Justice. 1993. Uridului pukkuron chahwasang (Our Shameful Self-Portrait). Seoul: Umjiginun Chaek.

Doner, Richard, and Ansil Ramsey. 1997. Competitive Clientelism and Economic Governance: The Case of Thailand. In Business and the State in Developing Countries, edited by Sylvia Maxfield and Ben Ross Schneider, 237-76. Ithaca, N.Y.: Cornell University Press.

Economic Planning Board. 1983. Kyonje kihoekwon isipnyonsa (Twenty-Year History of the EPB). Seoul: EPB.

Evans, Peter. 1995. Embedded Autonomy: States and Industrial Transformation. Princeton, N.J.: Princeton University Press.

Fields, Karl. 1997. Strong States and Business Organization in Korea and Taiwan. In Business and the State in Developing Countries, edited by Sylvia Maxfield and Ben Ross Schneider, 122-51. Ithaca, N.Y.: Cornell University Press.

Greenspan, Alan. 1998. Remarks before the 34th Annual Conference on Bank Structure and Competition. 7 May. Federal Reserve of Chicago.

Ha, Jong-dae. 1998. No-ssi pichagum 8 ch'on ok nomnunda (Roh's Secret Funds Exceeded $800 Million). Shindonga (March):230-51.

Haggard, Stephan, and Robert R. Kaufman. 1995. The Political Economy of Democratic Transitions. Princeton, N.J.: Princeton University Press.

Han, Heung Soo, and Byung Hoon Ahn. 1994. Hankuk ui pihap popchok chungch'i chagum ui yuhyong kwa siltae (Patterns and Realities of Illegal Political Funds in Korea). Tong Suh Yonku (East and West Studies) 7:198-215.

Hutchcroft, Paul. 2000. Obstructive Corruption: The Politics of Privilege in the Philippines. In Rents, Rent-Seeking, and Economic Development: Theory and Evidence in Asia, edited by Mushtaq H. Khan and K. S. Jomo, 207-47. New York: Cambridge University Press.

Johnson, Chalmers. 1986. Tanaka Kakuei, Structural Corruption, and the Advent of Machine Politics in Japan. Journal of Japanese Studies 12 (1):1-28.

___. 1987. Political Institutions and Economic Performance: The Government-Business Relationship in Japan, South Korea, and Taiwan. In The Political Economy of the New Asian Industrialism, edited by Frederic Deyo, 136-64. Ithaca, N.Y.: Cornell University Press.

Jomo, K. S., and Edmund Terence Gomez. 1998. Rent Generation and Distribution, and their Consequences in Malaysia. Unpublished manuscript, University of Malaya.

Jones, Leroy P., and Il Sakong. 1980. Government, Business, and Entrepreneurship in Economic Development: The Korean Case. Cambridge, Mass.: Harvard University Press.

Kang, David C. 1995. South Korean and Taiwanese Development and the New Institutional Economics. International Organization 49 (3):555-87.

___. 2002. Crony Capitalism: Corruption and Development in South Korea and the Philippines. Cambridge: Cambridge University Press.

Katzenstein, Peter J., ed. 1978. Between Power and Plenty: Foreign Economic Policies of Advanced Industrial States. Madison: University of Wisconsin Press.

Khan, Mushtaq. 1996a. A Typology of Corrupt Transactions in Developing Countries. IDS Bulletin 27 (2):12-21.

___. 1996b. The Efficiency Implications of Corruption. Journal of International Development 8 (5):683-96.

___. 1997. The Input-Output Function for Rent-seeking: A Comparative Analysis of Differential Effects. Unpublished manuscript, University of London.

Kim, Chun-gun. 1988. Hankuk kiop ui chunjose ui kwanhan yonku (Research on Quasi-taxes on Korean Companies). Seoul: Korea Economic Research Institute.

Kim, Jin-bae, and Chang-rae Park. 1968. Ch'agwan (Foreign Loans). Sindonga (December):80-102.

Kim, Jong-Seok. 1996. Korea's Regulatory Reform: A Critical Review. In Korea's Economy 1996, edited by Robert Rich, 10-14. Washington, D.C.: Korea Economic Institute of America.

Kim, Joungwon Alexander. 1975. Divided Korea: The Politics of Development. 1945-1972. Cambridge, Mass.: Harvard University Press.

Kim, Kwan Bong. 1971. The Korea-Japan Treaty Crisis and the Instability of the Korean Political System. New York: Praeger.

Kim, Yong-suh. 1995. No Tae-u kusokgwa YS ui sontaek (The Detention of Roh Tae-woo and Kim Young-sam's Choices). Sisa Wolgan (December):56-65.

Kiser, Edgar, and Xiaoxi Tong. 1992. Determinants of the Amount and Type of Corruption in State Fiscal Bureaucracies: An Analysis of Late Imperial China. Comparative Political Studies 25 (3):300-31.

Klitgaard, Robert. 1988. Controlling Corruption. Berkeley: University of California Press.

Krasner, Stephen D. 1978. Defending the National Interest: Raw Materials Investments and U.S. Foreign Policy. Princeton, N.J.:. Princeton University Press.

Kreps, David M. 1990. A Course in Microeconomic Theory. Princeton, N.J.: Princeton University Press.

Ku, Bon-hong. 1999. Hanguk sahoe eso pup'ae wa chagume kwanhan yonku: 6 kong hwakuk ul chungshimuro (Research on the Sociology of Korean Corruption and Funds with a Focus on the 6th Republic). Ph.D. diss., Yonsei University.

Kuk, Minho. 1995. The Governmental Role in the Making of Chaebol in the Industrial Development of South Korea. Asian Perspective 19 (1):108-31.

Lee, Jong-won. 1997. 97 Taeson: DJ pichagum (The 1997 Election: DJ's Secret Funds). Jugan Chosun (23 October):26-34.

Lee, Young-suk. 1987. Songo chagum kwa chaebol (Election Funds and Chaebols). Wolgan Chosun (November):222-39.

Levi, Margaret. 1988. Of Rule and Revenue. Berkeley: University of California Press.

Levy, Richard. 1995. Corruption, Economic Crime, and Social Transformation Since the Reforms: The Debate in China. Australian Journal of Chinese Affairs 95 (33):1-25.

Lew, Seok-jin. 1992. Bringing Capital Back In: A Case Study of South Korean Automobile Industrialization. Ph.D. diss., Yale University.

Martial Law Enforcement Headquarters. 1980. Kwollyokhyong ch'ukchaesa kyolgwa (The Investigation and Disposition of Corrupt Officials). Government press release.

McCubbins, Matthew D., and Thomas Schwartz. 1984. Congressional Oversight Overlooked: Police Patrols Versus Fire Alarms. American Journal of Political Science 28 (1):165-79.

Migdal, Joel S. 1988. Strong Societies and Weak States: State-Society Relations and State Capabilities in the Third World. Princeton, N.J.: Princeton University Press.

Milgrom, Paul, and John Roberts. 1990. Bargaining Costs, Influence Costs, and the Organization of Economic Activity. In Perspectives on Positive Political Economy, edited by James E. Alt and Kenneth A. Shepsle, 57-89. Cambridge: Cambridge University Press.

Miller, Gary. 1992. Managerial Dilemmas: The Political Economy of Hierarchy. Cambridge: Cambridge University Press.

Morris, Stephen D. 1991. Corruption and Politics in Contemporary Mexico. Tuscaloosa: University of Alabama Press.

North, Douglass C. 1986. The New Institutional Economics. Journal of Theoretical and Institutional Economics 142 (1):230-37.

Nti, Kofi O. 1997. Comparative Statics of Contests and Rent-Seeking Games. International Economic Review 38 (1):43-59.

Oh, Hyo-jin. 1986. JP, Taumun imcha ch'aryeya: Chonghwadae mildam (JP, Your Turn Is Next: Behind the Scenes at the Blue House). Wolgan Chosun (December):38-49.

Olson, Mancur. 1982. The Rise and Decline of Nations: Economic Growth, Stagflation, and Social Rigidities. New Haven, Conn.: Yale University Press.

Önis, Ziya. 1991. The Logic of the Developmental State. Comparative Politics 24 (1):109-26.

Park, Byung-yoon. 1982. Chaebol kwa chongch'i (Chaebol and Politics). Seoul: Hankuk Yongoso.

___. 1979. Chadongcha Sanob naemak (The Inside Story of the Automobile Industry). Shindonga (November):152-66.

Park, In-Hwan. 1992. 1 chowon kongsa: Kim Young-sam, Kim Dae-Jung, Chung Juyoung songo chakum (The Billion Dollar Deal: Campaign Funds and Kim Young-sam, Kim Dae-jung, and Chung Ju-young). Shingonda (December):268-77.

Park, Jong-yol. 1989. 5 kong hwakuk chongch'i chagum (Political Funds in the 5th Republic). Shindonga (January):280-300.

Park, Kyung-suk. 1967. Taet'ongnyong kukhoe wiwon songo chagum (Campaign Funds for Presidential and Assembly Elections). Sindonga (May):204-17.

Ramseyer, Mark, and Frances McCall Rosenbluth. 1993. Japan's Political Marketplace. Cambridge, Mass.: Harvard University Press.

Root, Hilton. 1994. The Fountain of Privilege: Political Foundations of Markets in Old Regime France and England. Berkeley: University of California Press.

Rose-Ackerman, Susan. 1978. Corruption: A Study in Political Economy. New York: Academic Press.

___. 1988. Bribery. In The New Palgrave Dictionary of Economic Thought, edited by John Eatwell, Murray Milgate, and Peter Newman, 277-78. London: Macmillan.

Sandholtz, Wayne, and William Koetzle. 1998. Accounting for Corruption: Economic Structure, Democratic Norms, and Trade. Unpublished manuscript, University of California, Irvine.

Schneider, Ben Ross. 1998. Elusive Synergy: Business-Government Relations and Development. Comparative Politics 31 (1):101-22.

Shafer, D. Michael. 1994. Winners and Losers: How Sectors Shape the Developmental Prospects of States. Ithaca, N.Y.: Cornell University Press.

Shleifer, Andrei, and Robert W. Vishny. 1993. Corruption. Quarterly Journal of Economics 108 (3):599-617.

Spiller, Pablo J. 1990. Politicians, Interest Groups, and Regulators: A Multiple-Principals Agency Theory of Regulation, or 'Let Them Be Bribed.' The Journal of Law and Economics 33 (1):65-101.

Theobald, Robin. 1990. Corruption, Development, and Underdevelopment. Durham, N.C.: Duke University Press.

Tilly, Charles. 1985. War Making and State Making as Organized Crime. In Bringing the State Back In, edited by Peter B. Evans, Dietrich Rueschemeyer, and Theda Skocpol, 169-91. Cambridge: Cambridge University Press.

Treisman, Daniel. 1997. The Causes of Corruption: a Cross-national Study. Unpublished manuscript, University of California, Los Angeles.

Tullock, Gordon. 1988. Rents and Rent-seeking. In The Political Economy of Rent-Seeking, edited by Charles K. Rowley, Robert D. Tollison, and Gordon Tullock, 51-62. Boston: Kluwer.

U.S. House of Representatives. 1978. Investigation of Korean-American Relations (Hearing Before the Subcommittee on International Organizations of the Committee on International Relations. House of Representatives. Ninety-Fifth Congress). Part II.

Wade, Robert. 1985. The Market for Public Office: Why the Indian State Is Not Better at Development. World Development 13 (4):467-97.

Wedeman, Andrew. 1999. Systemic Change and Corruption in China. Paper presented at the 95th Annual Meeting of the American Political Science Association, Atlanta, Georgia.

Weingast, Barry R. 1993. Constitutions as Governance Structures: The Political Foundations of Secure Markets. Journal of Institutional and Theoretical Economics 149 (1):286-311.

Weyland, Kurt. 1997. 'Growth with Equity' in Chile's New Democracy? Latin American Research Review 32:37-67.

Williamson, Oliver E. 1985. The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting. New York: Free Press.

___. 1991. Comparative Economic Organization: The Analysis of Discrete Structural Alternatives. Administrative Science Quarterly 36:269-96.

Woo, Jung-en. 1991. Race to the Swift: State and Finance in Korean Industrialization. New York: Columbia University Press.

Yang, Gi-Dae. 1997. Toduk konghwaguk (Republic of Thieves). Seoul: Donga Ilbo.

Yoon, Sung-yong. 1995. Pichagum (Secret Funds). Seoul: Jiyangsa.

Yoon, Young-ho. 1997. Chong Tae-su wa komun ton (Chung Tae-soo and Black Money). Shindonga (March):188-204.