Intra-party politics and minority coalition government in South Korea

Japanese Journal of Political Sciencehttp://journals.cambridge.org/JJP

Additional services for Japanese Journal of PoliticalScience:

Email alerts: Click hereSubscriptions: Click hereCommercial reprints: Click hereTerms of use : Click here

Intra-Party Politics and Minority Coalition Government inSouth Korea

YOUNGMI KIM

Japanese Journal of Political Science / Volume 9 / Issue 03 / December 2008, pp 367 - 389DOI: 10.1017/S1468109908003186, Published online: 29 October 2008

Link to this article: http://journals.cambridge.org/abstract_S1468109908003186

How to cite this article:YOUNGMI KIM (2008). Intra-Party Politics and Minority Coalition Government in South Korea.Japanese Journal of Political Science, 9, pp 367-389 doi:10.1017/S1468109908003186

Request Permissions : Click here

Downloaded from http://journals.cambridge.org/JJP, IP address: 193.225.200.93 on 25 Feb 2015

Japanese Journal of Political Science 9 (3) 367–389 Printed in the United Kingdom C© Cambridge University Pressdoi:10.1017/S1468109908003186

Intra-Party Politics and Minority CoalitionGovernment in South Korea

YOU NG M I KI M∗

University of [email protected]

Abstract

This paper examines the internal dynamics of Korean political parties tounderstand why the minority coalition government of Kim Dae-jung suffered frompolitical stalemate or deadlocks in the legislature. It shows that a focus on the sizeof the government in terms of a majority status in the legislature does not offera convincing explanation of why the Kim Dae-jung administration slid towardsungovernability. Instead better insights come from an analysis of party organization,an aspect of party politics rarely examined through in-depth analysis. The paper showsthat in terms of the key dimensions of organization (leadership type, factionalism,funding, linkage role) Korean political parties fail to connect citizens to the politicalsystem.

Since the start of democratization in 1987, coalition building has played a significant rolein South Korean politics. Building coalitions has been a crucial tool used by politicalparties towards electoral success. However, it did not ensure governability.1 After eachelection, in fact, the ‘fission and fusion’ of political parties in the legislature haveundermined the government’s capacity to legislate and eventually harmed governability.To avoid political deadlocks in the legislature, the ruling parties have relied on attractinglegislators from competing parties and on building coalitions with other parties in orderto enlarge the size of the government.

∗ Youngmi Kim is a Leverhulme Trust Early Career Fellow at the School of Social and PoliticalStudies, University of Edinburgh (UK). Her main interests lie in party politics, coalition building,and democratization, especially with reference to East Asia.

1 Governability is here understood as the ability of governments to make policies decisively − MichaelCoppedge, ‘Prospects for Democratic Governability in Venezuela’, Journal of Interamerican Studies andWorld Affairs, 36(2) (1994): 39–64; Michael Coppedge, ‘Party Systems, Governability and the Quality ofDemocracy in Latin America’, Paper presented at the Representation and Democratic Politics in LatinAmerica, Buenos Aires, 7–8 June 2001.

367

368 youngmi kim

The key question that this paper seeks to answer is whether ungovernability is linkedto the coalition government’s minority status in the legislature. Searching for answers,this paper speaks to different bodies of literature, which have thus far rarely talkedto each other: scholarship on coalition theories in parliamentary systems and regimestudies in presidential systems. Coalition research on parliamentary systems focuseson the formation, size, and termination of government.2 Research on presidentialsystems has concentrated on political deadlock as a result of divided (or minority)government.3 Recent scholarship questions the extent to which deadlock is structuralin the cases of minority governments or coalitions in presidential systems.4 Scholarsin this field contend that minority governments do not necessarily face deadlock andcoalition governments are not necessarily unstable.5

The case of the Kim Dae-jung administration in South Korea, with its high levelsof party fission and fusion, merger and break-ups, provides an interesting vantagepoint to examine the case of a minority coalition government in a semi-presidentialsystem where the ruling coalition faces a large opposition in the legislature, ultimately

2 Vernon Bogdanor, Coalition Government in Western Europe (London: Heinemann Education Books,1983); Lieven De Winter, Rudy B. Andeweg, and Patrick Dumon. ‘The State of Art in CoalitionResearch: Critical Appraisals and Alternative Avenues’, Paper presented at the Joint Sessions of EuropeanConsortium for Political Research, Torino, 22–28 March 2002; Michael Laver and Norman Schofield,Multi Government: The Politics of Coalition in Europe (Oxford: Oxford University Press, 1990); KaareStrøm and Woolfgang C. Muller, ‘Coalition Agreement and Governance’, Paper presented at the AnnualMeeting of the American Political Science Association, San Francisco 2001.

3 Josep M. Colomer and Gabriel L. Negretto, ‘Can Presidentialism Work Like Parliamentarism?’,Government and Opposition, 40(1) (2005): 60–89; Robert Elgie, Divided Government in ComparativePerspective (Oxford: Oxford University Press, 2001); Juan J. Linz, ‘The Perils of Presidentialism’, Journalof Democracy, 1(1) (1990): 51–69; Scott Mainwaring and Matthew Soberg Shugart, Presidentialism andDemocracy in Latin America (Cambridge: Cambridge University Press, 1997); Gabriel L. Negretto,‘Government Capacity and Policy Making by Decree in Latin America: The Cases of Brazil andArgentina’, Comparative Political Studies, 37(5) (2004): 531–62.

4 Jose Antonio Cheibub, ‘Minority Governments, Deadlock Situations, and the Survival of PresidentialDemocracies’, Comparative Political Studies, 35(3) (2002): 284–312; Jose Antonio Cheibub andFernando Limongi, ‘Democratic Institutions and Regime Survival: Parliamentary and PresidentialDemocracies Reconsidered’, Annual Review of Political Science, 5 (2002): 1551–79; Jose AntonioCheibub, Adam Przeworski, and Sebastian M. Saiegh, ‘Government Coalitions and LegislativeSuccess under Presidentialism and Paliamentarism’, British Journal of Political Science, 34 (2004):565–87; Scott Mainwaring and Matthew Soberg Shugart, Presidentialism and Democracy in LatinAmerica.

5 Jose Antonio Cheibub, ‘Minority Governments’; Jose Antonio Cheibub and Fernando Limongi,‘Democratic Institutions and Regime Survival’; Jose Antonio Cheibub, Adam Przeworski, and SebastianM. Saiegh, ‘Government Coalitions’; Scott Mainwaring and Matthew Soberg Shugart, Presidentialismand Democracy in Latin America; Gerald Peck, ‘Coalition Governments Versus Minority Governments:Bargaining Power, Cohesion and Budgeting Outcomes’, Public Choice, 121(1) (2004): 1–24; Kaare Strøm,Minority Government and Majority Rule (Cambridge: Cambridge University Press, 1990); Kaare Strømand Woolfgang C. Muller (eds), Coalition Governments in Western Europe (Oxford: Oxford UniversityPress, 2001).

intra-party politics and minority coalition government in south korea 369

leading to political deadlock.6 In particular, the paper pays close attention to the areaof intra-party politics and the mechanisms of negotiation and compromise.

By emphasizing the salience of party organization and showing how intra-partypolitics impacts on governability, the paper aims to contribute to the study of minoritycoalition governments under presidential and semi-presidential systems. The paperadvances two main claims. First, that ungovernability does not derive from governmentsize (the minority status not being a sufficient condition for the lack of governability);even when majority status was attained, instability persisted. Second, the paper proposesthat attention should be paid to the question of party organization. Understanding howthe party system and parties are run and function sheds significant light on the wholecoalition formation and instability process.

Of course, the power to infer generalizable findings from a single case study islimited at best. Nevertheless, this in-depth analysis also contributes to broader debateson political actors’ behaviour and intra- or inter-party politics in recently democratizedcountries, in East Asia (Taiwan) and elsewhere (East Europe and Latin America). Bydoing so, it answers the requests in the literature that more attention be paid to singleor small-N comparative studies with in-depth qualitative methodology.7

The paper is divided into two main sections. The paper starts with a discussionof party politics and government–opposition conflict during the Kim Dae-jungadministration, which eventually led to a deadlock in the legislature. I show whythe deadlock occurred and especially focus on the behaviour of the main oppositionparty. The second section moves on to examine party organization as a crucial factoraffecting governability. Following the work by Panebianco, Lawson, Katz, and Mair onparty organization, I discuss, by looking at the level of factionalism within the party,the type of party leadership, the party funding system, and the extent to which theparty continues to play its linkage role between the political system and society atlarge.

Party politics and organization

As Katz and Mair point out, there has been remarkably little literature on ‘theorganizational development of parties’ and – more surprisingly – there are limited

6 Whether South Korea belongs to presidential system or semi-presidential system is a hotly debatedissue in the literature. See Robert Elgie, ‘Semi-Presidentialism: Concepts, Consequences and ContestingExplanations’, Political Studies Review, 2 (2004): 314–30.

7 Michael Laver, ‘Party Competition and Party System Change: The Interaction of Electoral Bargainingand Party Competition’, Journal of Theoretical Politics, I (1989): 301–25; Hanna Back, ‘Explainingand Predicting Coalition Outcomes: Conclusions from Studying Data on Local Coalitions’, EuropeanJournal of Political Research, 42(4) (2003): 441–72; Lieven De Winter, Rudy B. Andeweg, and PatrickDumon, ‘The State of Art in Coalition Research’; Charles Lees, The Red–Green Coalition in Germany:Politics, Personalities and Power (Manchester: Manchester University Press, 2000); Carol Mershon,‘Party Factions and Coalition Government: Portfolio Allocation in Italian Christian Democracy’,Electoral Studies, 20(4) (2001): 555–80; Youngmi Kim, ‘Explaining Minority Coalition Governmentand Governability in South Korea: A Review Essay’, Korea Observer, 39(1) (2008): 59–84.

370 youngmi kim

numbers of case studies of single parties or single countries.8 This relative neglectin the literature is particularly noteworthy because, as Lawson contends: it is ‘onlyby looking within can we understand internal play of party power and why partyfunctions’.9

Studies of presidential and semi-presidential systems have devoted little spaceto analysis of how the organization of political parties impacts on governability.Within regime type studies, a frequently made argument is that presidential systemsinstitutionally create divided or minority governments and they fatally lead to a ‘crisis ofgovernability’10 or political deadlock and political instability. Presidential regimes werewidely assumed to be less cooperative due to the tension in executive and legislativerelations in the 1980s, with many cases in Latin America where presidential regimes wereoften blamed for poor political performance and the failure of democratic governance.While arguing that ‘there is no universally best form of government’,11 Mainwaring andShugart point to ‘lower level of development and non democratic political culture’ asreasons for the poor performance of presidential systems.12 In this respect, Sartori’spoint,13 that undisciplined parties are problematic − not only in a presidential system,but they could be even worse in a parliamentary system − appears especially pertinent.In addition, Cheibub maintains that highly disciplined parties would mitigate tensionsin the legislature and avoid political crises such as deadlock by forming legislativecoalitions.14

Given the overall lack of in-depth studies of how party politics affects governabilityin presidential systems, in order to understand the dynamics of intra-party politicswe need to turn to the study of coalition governments in parliamentary systems,particularly in Western Europe.

Here an issue traditionally indicated as crucial for cabinet stability is the sizeof government. The size of government is a key variable commonly used to explaingovernability or the duration of the government. The small size of government, suchas a minority government or divided government, is widely believed to be related tothe poor performance of a government as it faces big opposition in the legislature.However, as will be shown in the following pages, there may be other factors other thansize affecting governability, which the paper identifies in the level of party organization.

8 Richard Kats and Peter Mair, ‘Changing Models of Party Organization and Party Democracy: TheEmergence of the Cartel Party’, Party Politics, 1(1) (1995): 2.

9 Kay Lawson, ‘When Linkage Fails’, in K. Lawson and P. H. Merkl (eds), When Parties Fail: EmergingAlternative Organizations (Princeton: Princeton University Press, 1988), p. x.

10 Juan J. Linz and Arturo Valenzuela, The Failure of Presidential Democracy: The Case of Latin America(Baltimore: Johns Hopkins University Press, 1994).

11 Scott Mainwaring and Matthew Soberg Shugart, Presidentialism and Democracy in Latin America, p. 3.12 Ibid.: 53.13

1994, cited in ibid.: 53

14 Jose Antonio Cheibub, ‘Minority Governments’.


This is important because, as some scholars have recently shown,15 this suggests thatthere might be a fundamental element of political stability beyond regime type and thesize of government, and this is where the two bodies of literature identified above cometogether.

A final point that has to be made is that the study of coalitions and coalitiongovernments is new in Korean scholarship.16 This can be partly ascribed to widespreadnegative perceptions of coalition building by many scholars such as Son17 and Park18

who view coalition building and the search for broader political alliances as key factorsin bringing about factionalism and conflicts among parties. A second reason stemsfrom the fact that South Korea does not seem to belong anywhere in the mainstream ofexisting coalition studies, mostly focused on Western cases and parliamentary systems.South Korea’s system is semi-presidential and its location in North-East Asia hassomehow isolated it from clusters of regional studies examining and comparing LatinAmerican and/or European countries. Also, despite an extensive literature on the size ofgovernment and formation of coalition governments19 scholars working on coalitionsseem to pay little attention to non-Western cases and non-parliamentary systems.Although a large-N research strategy has been traditionally applied to the study ofcoalitions, Korea has never fully been taken into account. Scholars working on Korea onthe other hand have mainly focused on democratic consolidation and political deadlock.

Coalition politics in South Korea

Since the start of democratization in 1987, two political leaders have won politicalelections through a strategy based on coalition building in Korea: Kim Young-sam(1993–98) and Kim Dae-jung (1998–2003). More recently, the outgoing president Roh

15 Jose Antonio Cheibub, Adam Przeworski, and Sebastian M. Saiegh, ‘Government Coalitions’;Cheibub, ‘Minority Governments’; Robert Elgie, Divided Government; Argelina Cheibub Figueiredoand Fernando Limongi, ‘Presidential Power, Legislative Organisation, and Party Behavior in Brazil’,Comparative Politics, 32(2) (2000): 151–70; Vicky Randall and Lars Svasand, ‘Party Institutionalizationin New Democracies’, Party Politics, 8(1) (2002): 5–29.

16 By Korean scholarship here I refer not only to Korean scholars, but in general to all those working onthe country.

17 Hyeokjae Son, ‘Gukmin Musihan Jeongchigwonui Saepanjjagi: Jeonggyegaepyeonui HyeonhwanggwaGwaje [Making New Political Frame Ignoring People: The Situtation and Assignments of PoliticalReorganization]’, Donghyanggwa Jeonmang [Tendency and Prospect], 41 (1999): 60–72

18 Chanpyo Park, ‘Hanguk Jeongdang Minjuhwaronui Banseongjeok Seongchal: JeongdangMinjuhwainga Taljeongdanginga? [Reflective Aspects on Party Democracy in South Korea: Is It PartyDemocracy or Defecting from Party?]’, Sahoe Gwahak Yeongu [Social Science Studies], 11 (2003): 137–64.

19 William Riker, The Theory of Political Coalitions (New Haven, London: Yale University Press, 1962);Robert Axelrod, ‘Where the Votes Come From: An Analysis of Electoral Coalitions, 1952–1968’,The American Political Science Review, 66 (1972): 11–20; Michael Laver and Junko Kato, ‘DynamicsApproaches to Government Formation and the Genetic Instability of Decisive Structures in Japan’,Electoral Studies, 20 (2001): 509–27; Michael Laver and Norman Schofield, Multi Government; L.W.Martin and R.T. Stevenson, ‘Government Formation in Parliamentary Democracies’, American Journalof Political Science, 45 (2001): 33–50; Norman Schofield, ‘Genetic Instability of Majority Rule’, Review ofEconomic Studies, 50 (1983): 6966–705.

372 youngmi kim

Moo-hyun (formerly presidential candidate of the New Millennium Democratic Party,MDP, 2003–08) has also greatly benefited from entering a coalition with Chung Mong-joon, the leader of the People’s Power 21 (PP 21) Party in order to win the 2002

presidential election, even though Chung Mong-joon withdrew his support a fewhours before the presidential election.20

Coalition building had also played a role even in the transitional period fromauthoritarian rule to democracy, as the military regime was forced by the oppositioninto a process of bargaining and compromise. The ruling Democratic Justice Partymerged with two opposition parties in 1990: the Unification Democratic Party led byKim Young-sam and the New Democratic Republican Party led by Kim Jong-pil. Thethree parties merged and later renamed themselves as the Democratic Liberal Party andagreed to nominate Kim Young-sam as presidential candidate for the then approachingelections. As Figure 1 shows, most of the parties have their origins in the DemocraticRepublic Party and the Democratic Party going back to 1960.

A second moment where coalition-building seemed to be decisive in Korea’spolitical life was before the 1997 presidential elections, when the ruling party facedinternal divisions and the political environment was in turmoil following the outbreakof the ‘Asian crisis’ and the simultaneous condemnation of its perceived inefficiency bythe public. The opposition National Congress for New Politics (NCNP) party leader,Kim Dae-jung, formed an alliance with the United Liberal Democrats party leader, KimJong-pil, in order to fight the election. After a long history as an opposition leader, KimDae-jung finally became president.

When Kim Dae-jung became president in 1998, the NCNP (the president’s party)had 77 seats (26.3%) out of a total 293 seats, his coalition partner ULD had 43 seats(14.7%), and the opposition Grand National Party had 161 (54%). In combination, thetwo ruling parties of the coalition government had 120 seats (41%) out of 293, which isstill far below a majority (see Figure 2).

The Kim Dae-jung administration set out with an extensive agenda, and yet theruling party found it difficult to implement its policies from the early outset of thelegislature. Facing a political deadlock in the National Assembly, the ruling coalitionparties immediately started to enlarge the size of the ruling parties by attracting defectorsfrom the opposition. Enlarging the government’s size at all costs did not seem tobring any solution to the political deadlock, however. Quite the contrary, the conflictsdeepened not only with the opposition parties but also with the coalition partner(s) aswell as with factions within the ruling party.

As noted, Kim Dae-jung won office by building a coalition with the previouscoalition partner of the ruling party, Kim Jong-pil, a leading figure of the DemocraticRepublic Party under the Park Jeong-hee administration, who was determined tochange the constitution. Kim Jong-pil believed that a parliamentary system wouldwork better than the present presidential system.

20 Donga Ilbo, 20 December 2002.


Figure 1 Party system and coalition building (1960–2008)Notes: Party acronyms: DRP: Democratic Republican Party led by Kim Jong-pil (1963);DP: Democratic Party led by Park Sun-cheon 1963 (Later New Democratic Party in 1964and its former party of the New Korea Democratic Party (NKDP) in the 1980s led by KimYoung-sam and Kim Dae-jung); DJP: Democratic Justice Party – ruling party led by ChunDoo-hwan (1980) and later by Roh Tae-Woo (1987); NKDP: New Korea Democratic Partyled by Kim Young-sam and Kim Dae-jung (1985); UDP: Unification Democratic Party ledby Kim Young-Sam (1987); NDRP: New Democratic Republican Party led by Kim Jong-Pil (1987); PDP: Peace Democratic Party led by Kim Dae-Jung (1987); DLP: DemocraticLiberal Party after the parties’ merger (1990); DP: Democratic Party led by Lee Ki-taek(1990); UPP: United People’s Party led by Chung Ju-young (1992); NKP: New KoreaParty led by Kim Young-sam (1996); PNP: People’s New Party led by Lee In-je (1997);GNP: Grand National Party led by Lee Hoe-chang (1997) and Lee Myung-bak (2008) ;ULD: United Liberal Democrats led by Kim Jong-pil (1995); KNP: Korea’s New Party ledby Kim Yong-hwan (2000); NCNP: National Congress for New Politics led by Kim Dae-jung (1996); MDP: New Millennium Democratic Party led by Kim Dae-jung and later RohMoo-hyun (2000∗); PP 21: People’s Power 21 led by Chung Mong-joon (2002); UP: UriParty led by Roh Moo-hyun (2003∗); DLP: Democratic Labour Party led by Kwon Young-kil(2004); UNDP21: United New Democratic Party (2007) led by Chung Dong-young andlater led by Son Hak-kyu; LFP: Liberty Forward Party (2008) led by Lee Hoe-chang22;CKP: Creative Korea Party led by Mun Kook-hyun (2007).Source: Author.

The 1990 coalition was created around the agreement between the leaders to changethe system from presidential to parliamentary (the objective was never achieved). The

21 The United New Democratic Party (UNDP) was founded by a group of people that defected from theUri Party before the presidential elections in 2007. The UNDP later merged with the New MillenniumDemocratic Party (MDP) and renamed the merged party as the Democratic Party.

22 Lee Hoe-chang defected from the Grand National Party and became a non-party affiliated candidatefor the presidential election in 2007 and gained 15% of total votes ranking the third in the presidentialvotes. The first was Lee Myung-bak gaining 48% and the second was Chung Dong-young with 26%.

374 youngmi kim

Figure 2 Party seat numbers after coalition building in February 1998Notes: NCNP − The National Congress for New Politics (later the New MillenniumDemocratic Party) led by Kim Dae-jung. ULD − The United Liberal Democrats led byKim Jong-pil. GNP − The Grand National Party led by Lee Hoi-chang.Source: Author.

second coalition with Kim Dae-jung and his party, the National Congress for the NewPolitics, set out again to reach this goal. If the coalition were to win the elections,Kim Dae-jung would become president and Kim Jong-pil prime minister. The primeminister would thus be able to carve out a role in making decisions on domestic affairs.23

Deadlocks: Does government size matter?The approval vote. The case of the vote of approval for the prime minister

illustrates the nature of Korean politics well. As a part of the coalition agreement, thefirst action of the president was to appoint Kim Jong-pil as Prime Minister, a nominationthat would then need to be ratified by the members of the National Assembly. This wasa necessary move to consolidate the ruling coalition, as Kim Dae-jung’s party couldcount on only 77 lawmakers and the coalition partner’s 43 lawmakers. A majority (50%plus 1) of the National Assembly members’ attendance for the vote was needed to allowKim Jong-pil to become Prime Minister. It also required two-thirds of the votes fromthose who took up their seats in the chamber when the vote was called.

In theory, this condition was not insurmountable. If the NCNP and the ULDsucceeded in negotiation with non-party affiliated legislators, then the ruling coalitionparties would need the support of just 13 out of a 163 opposition members belonging

23 Chosun Ilbo, 19 January 1999.


to the GNP. Kim Jong-pil could rely on personal connections with many memberswithin the Grand National Party ranks. Many of them were in fact from his nativeChungcheong province, including some from the same high school, while others werealumni from the same university. Some of them also used to work with Kim Jong-pilin previous parties, the Democratic Republic Party in the Park Jeong-hee era and theUnited Liberal Democrats under the Kim Young-sam administration. There was morethan enough margin to reach a compromise. So why did they not reach one?

If Kim Jong-pil had become Prime Minister, it was thought it would damage theGNP in the following local election in June 1998 and the general election, especiallyin Chungcheong province, where Kim Jong-pil had loyal electoral support.24 If theGNP were to support Kim Jong-pil’s bid as Prime Minister, this would appear as if theGNP accepted the legitimacy of the ruling coalition parties. This would also help KimJong-pil to gain stronger support from Chungcheong province. The National Congressfor the New Politics had strong support in Seoul, Gyeonggi, and Jeolla provinces, so ifthe GNP lost in the following local election, this would give the Grand National Partya chance to become the party of Gyeongsang province only.25 Therefore, the GrandNational Party had to fight for its own survival in the legislature. The main objective ofthe GNP was to break the coalition between the NCNP and the ULD and gain poweragain.

At the approval vote for the prime minister, all the legislators of the GNP were‘advised’ to attend the vote but to submit empty ballot sheets. In doing so, the oppositionparty would avoid popular blame for not attending an important National Assemblymeeting. At the same time, the empty ballot sheets would force the vote to be annulled.Within 20 minutes of the start of voting on 2 March 1998, the legislators of the UnitedLiberal Democrats became aware that opposition party members were handing in blankballots. The National Assembly descended into chaos, with the ruling party legislatorsasking to have the ballot declared null and void and the opposition instead demandingthe voting be completed.

The vote was cancelled at the last moment and the president appointed Kim Jong-pil as a deputy (or an acting) Prime Minister on 3 March. It took nearly six months toarrange another vote, which in fact took place on 17 August. Kim Jong-pil became theprime minister gaining 171 votes of support in the National Assembly. However, thiswas only after political reorganization (Jeonggye-Gaepyeon) of the ruling parties, whichattracted high numbers of defectors from the opposition parties. By then the NationalCongress for New Politics had 88 legislators and the United Liberal Democrats had 49

legislators. Except for one member belonging to the ULD, the members of the ruling

24 Since democratization regionalism has emerged as a crucial political cleavage in South Korea’s politics.Kim Dae-jung had strong support from Jeolla Province and Kim Jong-pil from Chungcheong Provinceand the Lee Hoe-chang from Gyeongsnag province.

25 The National Congress for the New Politics (NCNP) was the ruling party led by Kim Dae-jung. InJanuary 2000, it changed its name to the New Millennium Democratic Party (MDP).

376 youngmi kim

party all voted to approve Kim Jong-pil as prime minister and by then the ruling partyhad managed to gain around 30 votes from the Grand National Party.26

The previous vote in February had shown that independent voting for legislatorswas not an option given the prevailing Korean political culture. The question arisinghere is of course how parties manage to control secret ballots. Considering that eachparty is home to different factions (e.g. regional basis, school ties), that party leadersdiscipline party members to coordinate party decisions so strictly is surprising. ChosunIlbo reported that the committee of the Grand National Party announced that if anymembers voted to approve Kim as prime minister, the member(s) would be requiredto leave the party (as it happened in a handful of cases).27

When Cho Sun and Lee Ki-taek, formerly leaders of the ‘little’ Democratic Party,were excluded from standing as candidates in the 2000 elections, this confirmedLee Hoe-chang’s tendency to marginalize legislators holding dissenting views.28 Thismarginalization of dissent points to the lack of democracy in Korean party politics,where fear of not being allowed to stand again in elections acts as a powerfulconstraint on the deputy’s behaviour. There seemed to be no margin for compromise ornegotiation among party members belonging to different parties unless they defectedfrom one party and joined another. Park No-cheol, a chief secretary of Jeong Hee-su,the Seventeenth National Assembly member for the GNP, mentioned that:

There are four different levels of punishment when the individual lawmakerdoes not follow the party’s directives. The legislator acting against the party’sinstructions will be sanctioned by the party committee and may lose hisposition in the party. Some may even have to leave the party. Especially whenthe legislators are not elected through the general elections but by proportionalrepresentation, they would not have any choice but to leave the party.29

Too many bills, not enough votes. After Kim Dae-jung took office, the legislaturewas supposed to deal with an overwhelming 300 law proposals. The numerous conflictsand political crises in the legislature provided the Kim Dae-jung administration withthe opportunity to legitimize its attempt to reorganize the whole party system. Theruling party managed to attract large numbers of defectors from the opposition partiesand managed, in September 1999, to gain 160 seats out of 293 total seats, with 133 seatsgoing to the opposition Grand National Party. Because of the opposition party’s boycottof the National Assembly, this was occupied by members of the ruling parties, while the

26 Sekye Ilbo, 18 August 1998.27 Chosun Ilbo, 9 February 1998.28 When Kim Young-sam merged his party, the New Democratic Party, with two other parties, the

Democratic Liberty Party and the United Liberal Democrats in 1990 some party members, includingLee Ki-taek and Roh Moo-hyun, left the party and founded the Democratic Party but after failingin the general election they all scattered; some such as Roh Moo-hyun joined the New MillenniumDemocratic Party and some the Grand National Party. Since that time, the Democratic Party is oftencalled the ‘little’ Democratic Party.

29 Interview held in Seoul on 2 March 2006.


opposition party spent most of the time outside it. Until August 1999, meetings of theNational Assembly were held 19 times from the 189th National Assembly ExtraordinarySession to the 207th National Assembly Session; about half of the meetings were heldeither by the ruling parties or the opposition party who otherwise took to the streetsprotesting against the government.

In the eyes of the public, the government looked inefficient and unable to deliver itspolicies. Legislators only appeared interested in particularistic infighting and blockedeach other’s bills. While the budget bill and related bills were waiting to be approved,the Assembly Session was paralyzed for ten days, from 3 to 13 October 1999, due tothe conflict over the media reform report. There were 112 bills still awaiting to beapproved, which were all related to reform policies, including public subsidy bills,national security, human rights, etc.30 In this section I have shown that the enlargedruling parties did not secure governability, as the opposition went on to protest outsideof the National Assembly rather than attending the National Assembly meetings.31

I have revisited what happened in the legislature and why deadlock occurred fromthe very beginning of the administration. I have shown that enlarging the size of thegovernment was the main strategy adopted by the government to ensure governability.The strategy nonetheless failed, suggesting that other factors may have been hinderingcompromise and negotiation, in particular a highly disciplined, or more rigid partyorganization, as the next section illustrates.

Party organization in South Korea

Korean political parties hardly fit in any existing party typology, presenting featuresof mass parties, as well as those of cadre parties.32 Most parties in Korea do not haveenough due-paying members to be regarded as mass parties. However, in terms oforganization, Korean parties show a mass-party structure. Parties are structured intocentral (core) and peripheral levels. Parties are mass bureaucratic organizations withleaders; however, they do not have members. This would be, as Duverger notes, likehaving a school with teachers but without pupils.33

Most party members are mobilized by the staff recruited to run the electioncampaigns. As will be shown later in the paper, they are not party members but

30 Kukmin Ilbo, 4 November 1999.31 When the MDP was the opposition party (Kim Young-sam administration), the way it protested against

the ruling party was very similar. For example, the members protested outside the National Assemblyand refused to attend National Assembly meetings.

32 A mass party is ‘a party open to all and/or followed by masses of people’, see Giovanni Sartori, Partiesand Party Systems: A Framework for Analysis, Vol. I (Cambridge: Cambridge University Press, 1976).It is highly bureaucratized and the elites control the party organization. On the other hand, cadreparty is built around a select group of influential persons. These support candidates, conduct electoralcampaigns, and handle the finances of the party.

33 Maurice Duverger, ‘Caucus and Branch, Cadre Parties and Mass Parties’, in Peter Mair, The WestEuropean Party System (Oxford: Oxford University Press, 1990), p. 41.

378 youngmi kim

temporary electoral campaigners.34 Kim Young-ho notes that 6,146,187 party memberswere registered in the political parties in 1999, but 6,122,300 members were recruitedto carry out electoral campaigning and were paid for their effort.35 In fact, regularmembers paying dues mounted to only 23,889, that is only 0.07% of the total. That isvery low compared to that of Austria with 17.66% and even the UK with 1.92.36

Many scholars note that parties in Korea closely resemble cadre parties.37 In Korea,the candidate in each electoral district bears most of the financial burden on his orher own, with some subsidy from the central party. In terms of the party, it is morelikely that all candidates rely on the party leaders to raise funds. At both central andperiphery levels, the party revolves around the leader. The organization is often calleda one-man party or boss party. In addition, the financial burden is on the shouldersof the party leaders and so is the right to nominate candidates. Considering that thestakeholders in a mass party can be regarded as all of the members who share thefinancial burden for the functioning of the party by paying party dues, the ownershipof Korean parties de facto belong to the leaders. Herein lies the reason why, when theparty leader leaves his/her own party, the party often collapses and re-emerges underanother name (with the same leader). This is another reason why leaders are the onlyactors in the coalition-building process, whereas ordinary party members are virtuallyexcluded.

Others refer to Korean political parties as a type of ‘cartel party’, a concept developedby Katz and Mair.38 Similarly, to other cartel parties, political parties in Korea do notfinancially rely on party dues, but mainly on the state’s subvention. As a result, asJang Hoon notes, since democratization parties in Korea have functioned as cartelparties between the state and society.39 Finally, Gunther and Diamond argue thatclassical party typology does not accommodate real-world parties, especially not West-European parties.40 According to their typology, parties in South Korea, especially

34 Youngguk Jeong, ‘Hangukjeondangeu Jigudang Jojiggwa Gineung [Organizational System andFunctions of the District Party Chapters in Korean Politics: An Alternative argument]’, HangukgwaGukjejeongchi [Korea and International Relations], 6 (2000): 239.

35 Youngho Kim, Hanguk Jeongdang Jeongchiui Ihae [Understanding Party Politics in Korea] (Seoul: NanamPublisher, 2001), p. 299.

36 Yongho Kim (ibid.) notes that the Korean data are based on the Central Electoral Committee report,2000, 99 Party Activities and Revenue Report, p. 16, and pp. 33–4, whereas those EU-related are basedon Mair and Van Biezen, 2001, Party Membership in Twenty European Democracies, 1980–2000, p. 9.

37 Soonghoom Kil, ‘Hanguke Isseoseo Jeongdangjeongchiwa Jeongchimunhwa [Party Politics and PoliticalCulture]’, Hanguk Nondan, 4 (1990): 46–53; Youngho Kim, Hanguk Jeongdang Jeongchiui Ihae; ChanpyoPark, ‘Hanguk Jeongdang Minjuhwaronui Banseongjeok Seongchal’.

38 Richard Kats and Peter Mair, ‘Changing Models of Party Organization and Party Democracy’ 8−19.39 Hoon Jang, ‘Jeongdang Jeongchiwa Jibangjachi; Hanguk Minjuhwa 10 Nyeonui Jeongdang Jeongchi;

Yeonsokseonggwa Byeonhwa [Party Politics and Local Politics]’, Paper presented at the Hankuk JeongchiHakhoe Conference, Seoul 1997; Also see Hoon Jang, ‘Cartel Jeongdangchejeui HyeongseonggwaBaljeon: Minjuhwa Ihuui Hangukui Gyeongu [the Rise of the Cartel Party System in DemocraticKorea]’, Hangukgwa Gukjejeongchi [Korea and International Relations], 19(4) (2003): 31–60.

40 Richard Gunther and Larry Diamond, ‘Species of Political Parties: A New Typology’, Party Politics, 9(2)(2003): 170.


during the Kim Dae-jung administration, can be characterized as a hybrid type of elite-based and clientelistic party. They are organizationally thin and similar to Panebianco’s‘electoral professional party’. For electoralist parties, personal charisma or attractivenessis more important than organizational functions or ideology. Gunther and Diamondadd that parties in South Korea are catch-all parties but, given the strong regional basefor elections, parties in South Korea can also be said to be similar to ‘ethnic parties’.

Having provided a short overview of Korean political parties as organizationsblurring the divides between different party types, the section now turns to examininghow these parties are actually organized by looking at the following dimensions:leadership, factionalism, funding, and party–citizens linkage.

LeadershipParties in Korea did not create the leaders, rather leaders created the party. Party

leadership is fundamentally based on perceptions of legitimacy and support fromthe majority of the followers. Panebianco argues that the legitimacy of leadershiprelies in the control of public goods (collective incentives) and private goods (selectiveincentives).41 When the benefit is not distributed fairly enough, the legitimacy of theleadership is undermined. Looking into the party system in Korea, where the linkagebetween the state and civil society is limited, parties are unlikely to achieve an equaldistribution of either public goods or private goods.

At the local branch level, the legislators fail to come up with policies reflecting theviews of the majority of the people and tend to focus on some voters (generally relativesof other members) and not others. Thus, there are a large number of alienated citizensor voters who do not enjoy equal distribution of public goods. It is no different at levelof the central party. The party is more interested in gaining power and, in order to doso, the leaders are busy collecting funds from a few conglomerates or interest groupsthan in chasing after the small dues paid by the majority of the party’s members.

Kim Dae-jung won office with only 40.3% of the total votes, a mere 1.6% more thanhis competitor, Lee Hoe-chang. The problem is that his supporters were concentratedin the Honam (or Jeolla) region. Kim Dae-jung gained 90% of the total votes inJeolla province 44.2% in Seoul, 43% in Chungcheong province, but only 13% in theYeongnam (Gyeongsang) regions and 20% in Gangwon province.42 He gained relativelyhigh votes compared to the previous presidential election in Chungcheong provincebut this was due to the coalition partner. As scholars like Linz, Skach and Stepan, andMainwaring and Shugart note, the dual legitimacy of the executive and the legislativebranches of government structurally induces conflicts in the legislature.43 Kim

41 Angello Panebianco, Political Parties: Organization and Power, Translated by Marc Silver (Cambridgeand New York: Cambridge University Press, 1988), p. 40.

42 Donga Ilbo, 19 December 1997

43 Juan J. Linz, ‘The Perils of Presidentialism’; Alfred Stepan and Cindy Skach, ‘Constitutional Frameworksand Democratic Consolidation: Parliamentarism Versus Presidentialism’, World Politics, 46(1) (1993):1–22; Scott Mainwaring and Matthew Soberg Shugart, Presidentialism and Democracy in Latin America.

380 youngmi kim

Dae-jung gained legitimacy with a higher vote rate from the presidential election;however, the opposition GNP hardly sees itself as a loser, as it received much highervotes in the general election and therefore became the party with the largest numberof seats in the legislature. This dual legitimacy created strains in the legislature for theentire period of the Kim Dae-jung administration. With small numbers of supporters,it started to govern, although the lack of legitimacy hindered the implementation ofreform policies. The government was driven off track by the big opposition party nearlyevery time it tried to pass any new bills.

Another significant problem for political leadership in Korea is that presidentialpower and the ruling party’s elite are not equal; the party elite are subjugated by thepresident. When the parties are not separated from the power of the president, this easilycan lead to delegated democracy with leadership highly concentrated on the presidentand the president simply relying on his charismatic popular support to govern. This alsoleads to lack of legitimacy in the legislature, especially when facing an opposition partyor parties with more seats in the chamber. Decision-making procedures as well as theright to select the candidates are highly concentrated in the party leaders. If the leadersfail to gain agreement in committee meetings in the legislature, the ordinary legislatorsbarely have the right to vote according to their preferences but are required to follow theleaders’ indication. In the case of ruling parties, the party chairmen (Chongjae) do nothave much power but they are more like delegates of the original charismatic leaders:Kim Dae-jung and Kim Jong-pil respectively. In the case of the MDP, the ultimateleader of the ruling party is the president. Therefore, above the party representative(mostly chairman, Chongjae) of the floor negotiation group, the president’s decisionis also crucial in the process of compromise and negotiation. Lee Yang-hee, a formerlegislator of the ULD, pointed out that:

The power of the president is bigger than that of a King under the JoseonDynasty. Our constitutional laws allow too much power to the president. Thisis a disadvantage of the presidential system in South Korea. That is why wethink the parliamentary system would work better in Korea. In the case ofthe US, if a republican or democratic president suggests a new policy, anindividual legislator of the respective party is not required to follow the partywhip in the legislature, as in the ‘Westminster’ system. In the Korean system,however, if you do not follow the president’s line a range of disadvantages willarise. This is particular the case if a legislator harbours ambitions to be thechairman, a position appointed by the president. If you are the chairman, youessentially have to listen to the president who appointed you. This is not doingpolitics but being controlled.44

This also hinders party leaders from compromising with the opposition party. Thelegislature cannot play its balancing role against the executive. The president occupiesa pivotal place in the government and often dominates his own party. In the legislature,

44 Interview held in Seoul on 9 March 2006.


the ruling party is not free from the president’s decisions and this hinders negotiationsand compromise among the parties. The opposition party, of course, can aim to restrainthe power of the executive but the ruling parties will not or cannot reach an agreementwith the opposition on numerous pieces of legislation. This strict top–down hierarchyalso matters in the process of negotiations and compromise. Once the decision hasbeen made by the party leaders, the followers − that is, the ordinary legislators − willnot vote against the party’s decision. The reason is straightforward: to do so wouldmean no possibility to stand as a candidate for the following general election, no morepromotion within the party and no more financial support from the party leader. Inshort, when legislators do not obey the party decision, they are punished in these andmany other ways. Some are even forced to leave the party.

FactionalismAccording to Panebianco, an analysis of dominant coalitions can be carried

out in terms of the degree of internal cohesion, degree of stability, and the party’sorganizational power map.45 When the level of internal cohesiveness is low, the coalitionis likely to be unstable. In the case of the Kim Dae-jung administration, this was not adominant coalition in a Western parliamentary system, but a minority coalition facinga big opposition party under a semi-presidential system. So the government was notgoing to collapse as a result of investiture or a no-confidence vote and, no matter howmuch difficulty the government faced, it was highly unlikely to collapse during the termof the presidency. In other words, the period of political rule is institutionally secured.However, regardless of this stability, the Kim Dae-jung administration seemed to rolltowards ungovernability from the start, as a result of the level of factionalism withinthe coalition and even within Kim’s own party.

The high level of factionalism within Korean parties also hindered the consolidationof party organization in the country. The case of the NCNP, and of its oldest faction(the so-called Donggyo-dong faction), is illustrative in this respect.46 The membersof Donggyo-dong have all been secretaries or colleagues of Kim Dae-jung since thebeginning of his political life. When the government was launched, four out of eightexecutive members of the NCNP were from the Donggyo-dong faction, two were formerlawyers, one was a businessman, and the other was a journalist. The leaders of the NCNPwere mainly from the Donggyo-dong faction and played dominant roles within the party.

The NCNP leaders were highly dependent on President Kim Dae-jung and thisrelationship acted as a barrier in reaching an intra-party consensus. The dependencyof the Donggyo faction on the president means the party itself is dependent on thepresident. It is also a fundamental reason for the personalized characteristics of the

45 Angello Panebianco, Political Parties.46 Donggyo-dong is a name of area in Seoul where Kim Dae-jung’s former house was located. As his former

secretaries or colleagues, the members of his faction, spent their political life with Kim Dae-jung, literallydecades since the Park Jeong-hee administration.

382 youngmi kim

party. If the leader disappears, so does the party. In other words, loyalty to the presidentwould undermine not only the internal cohesiveness of the party, but also the balanceof power between the legislature and the administration. This also hinders creatingmechanisms of compromise in order to address conflicts with the opposition party andcoalition partner.

The factions within the NCNP were also divided according to high school alumninetworks. With Kim Dae-jung (from Jeolla province) in power, some among hishigh schools’ alumni from Jeolla province were appointed to senior positions in theadministration. High school ties are very important in Korean politics (and arguablyin economy as well) as they provide one of the key networks in terms of employmentand career development. Having school connections to a ‘senior’ who now holds animportant office is a key ‘door-opener’ in Korean politics. For example, during theKim Young-sam presidency, Gyeonggi high school (based in Seoul) and Gyeong-namhigh school (based in Busan, Gyeongsang province) alumni enjoyed high status andpositions in the administration as well as judiciary and legislature. Under Kim Dae-jung, the Gwangju high school, Gwangju Il high school, and Jeonju high school, allbased in Jeolla province, gained pre-eminency. For the sixteenth general election all ofthe Jeonju high school alumni of the MDP members were selected as candidates by theMDP leader. Within the ruling party, the high school factions were divided into Jeonju,Gwangju Il, and Gwangju high schools alumni factions. These factions competed overleading positions in the administration, judiciary, and legislature, as candidates for theelections, and even in the army.47

The unfair personnel policy of the Kim Dae-jung administration undermined theefficiency of the organization and came under criticism by the opposition party. KimPan-seok argues that official appointments were concentrated in the president’s handsand there was no professional human resource department in the party able to suggesthighly competent personnel.48

During the Kim Dae-jung administration, each minister kept his/her post for anaverage period of 10.54 months – that is the lowest average compared to 11.45 months forthe Kim Young-sam administration, 12.88 months for the Roh Tae-woo administration,17.11 months for the Chun Doo-hwan administration, and 24.54 months for the Park

47 This trend of sharing official positions by the ruling parties did not change even during the RohMoo-hyun administration. The president-elect Lee Myung-bak hinted at widespread corruption andthe need reform in state-owned public enterprises such as the Korea Expressway Corporation, KoreaElectric Power Corporation, Korea Development Bank, and Korea Telecom. The posts of president ofthe state-owned companies were often used to reward those who had helped the presidential candidatein his political career. The practice was often referred to as ‘parachute personnel’ as the personnelare appointed from the top, and not following the company’s internal appointment procedures (SeoulFinance, 6 January 2008).

48 Panseok Kim, ‘Daetongryeonggwa Insa: Jeongmu Gowijik Insahyeoksin [Presidential Personnel:Personnel Reform for Presidential Appointees]’, Hankuk Jeongchi Hakhoe [Journal of Korean PoliticalAssociation], 37(2) (2004): 392.


Jeong-hee administration.49 This also illustrates that the Kim Dae-jung administrationdid not hold hegemonic power. The government could not back up its ministers againstthe opposition party or public opinion in the case of political disputes. It is obviousthat, if the government cannot secure the position of a minister for even a year, theminister can hardly implement new policies. It was not only the short period of theminister’s tenure that illustrates this point, but also the chief executives’ positions couldnot be secured for more than 14 months.50

The process of personnel appointment was not institutionalized, let alone openand fair. Instead, it was carried out through either factional connection to the presidentor party leaders.

LinkageUnlike other types of parties, the cartel party does not link the state with civil

society. When the party does not link civil society with the state, and fails to representcivil society it is more likely to represent a small number of elite interests and thiseventually will lead to the erosion of legitimacy. Lawson argues that:

Participation, leadership recruitment, allocation of resources, the creation,and propagation of values, the control of behaviour (through the control offorce and/or through educative communication) all require the creation ofconnections between different levels of aggregation. Creating linkages is itselfan extremely important function of politics.51

In other words, to secure a party’s legitimacy and power in the legislature the partymust provide linkages between the state and the citizens. When the citizens are alienatedfrom participation, resource allocation, and developing policies, it is hardly likely toconsolidate internal power with a wide range of support. Even if the lawmakers achievea majority status through building coalitions, as long as elections are held for the partyto be judged by the citizens, the ruling party will face the citizens’ judgement. However,apart from the election, citizens and sometimes even ordinary legislators are excludedfrom the functions of party organization. The highly centralized power among the elitesfails to provide the essential linkage between the state and the citizens (see Table 1).Lawson argues that linkage matters when the party fails.52 For Key, linkage means‘interconnections between mass opinion and public decision’.53 Lawson emphasizes thatcreating linkages is essential in order to gain legitimacy and consolidate internal power.The party should remain in the core position between the state and the grassroots.Linkages between the two means that neither of them should be isolated from theother and the two should remain balanced. When the party wins office, the function

49 Ibid.50 Ibid.51 Kay Lawson, ‘When Linkage Fails’, p. 15.52 Ibid.53 Valdimer Orlando Key, Public Opinion and Public Policy Models of Political Linkage (New Haven: Yale

University Press, 1964), p. 14.

384 youngmi kim

Table 1. South Korean political parties between the state and civil society

Civil society

Linked Less linked Unlinked

State One man One man party(MDP, ULD, GNP)

Small number of elites Cadre Party Cartel partyMass Mass Party

Source: Author.

of creating linkages between the government and civil society or citizens is crucial togaining legitimacy and wide support. If the party fails to link with the state and thecitizens, the ruling party would hardly find wide and strong support from the people,no matter how good the reform policies. The Kim Dae-jung administration failed tolink itself with civil society. Leadership without nationwide legitimacy underminedgovernability during his administration.

FundingAccording to the National Audit Report, the rate of dues paid by each party’s

membership is 0.5%. In late 1997, just before Kim Dae-jung took office, the New KoreaParty and then ruling party officially had 3,723,138 members; however, only 22,793

members paid their dues − that is, 0.6%. The NCNP (by then led by Kim Dae-jung)had a 0.5% rate among the party members, the Liberal Democratic Party had 0.03%,and the People’s Nation had a rate of 0.1%.54

Most of the electoral campaign in Korea is funded by the individual party candidateand the candidate mainly relies on self-generated funds, often from his own businessor business supporters. Therefore, the number of members who pay party dues islimited. At the local level, no members pay their dues; rather, they are mobilized toattend electoral rallies and campaigns either through the use of gifts. Such gifts includewatches, towels, meals, and sometimes tours to tourist spots. ‘The more money or gifts,the more votes’ is a popular saying in Korea. Citizens have no chance to play a more activerole in politics because they have no opportunity to receive education and thus rely onthe favours by local leaders, such as the redistribution of gifts and funds. Accordingto an investigation carried out by the Chosun Ilbo, a candidate distributed envelopswith 100,000 won in cash to 30 potential voters in a village meeting in Busan. Anothercandidate confessed that he had to give a minimum of 30,000 won in each envelop to tryto woo potential voters; this increases to 70,000 won as the election day approaches. Atthe central level, parties receive annually 800 won per voter as an ordinary subvention.In the year of the presidential election and general election, an additional 800 won

54 Donga Ilbo, 10 December 1998.


Table 2. Central party income in 2000 (Unit: Million won)

Party Dues donation trust money subvention etc Total (%)

GNP 3,310 (9.6) 4,505 (13.1) 0.4 (0.001) 21,046 (61.5) 5,338.6 (15.6) 34,200NCNP 2,163 (2.8) 40,000 (53.1) 0.4 (0.0005) 18,417 (24.4) 14,665,6 (19.4) 75,246ULD 477 (1.8) 7,815 (30.7) 0.2 (0.0007) 9,565 (37.6) 7,577.8 (29.7) 25,435Total 5,950 (4.4) 52,320 (38.8) 1 (0.0007) 49,028 (36.3) 27,582 (20.4) 134,881

Source: Adapted from Jiyun Sim and Minjeon Kim, ‘Hanguk Jeondangui Suipgujo Yeongu –1997–2002 Nyeondo Jeongdang Hoegyebogo Jaryoreul Jungsimeuro [Revenue Structure ofKorean Political Parties]’, Hankukkwa Kukjejeongchi, 38(3) (2002): 163.

subvention is granted per voter. When it is the year with local elections that are heldon three levels (Si/Do Council Members, Autonomous Gu/Si/Gun Council Members,Education Board Members) the subvention is given three times so it amounts to 1,800

won per voter.55 Half of the subventions are given to the floor negotiation groups andthe other half is given to the parties on the basis of the number of seats in the legislature,the percentage of the national votes each party gained at the general election, and thepercentage of votes each party gained at the local election. The subvention is given tothe parties at central level.

Political donations can be made by an individual or organizations. An individualcan donate from 10,000 won to a hundred million won or 5% of the previous year’sincome. A company can donate five million won or 2% of the previous year’s incomeand organizations can donate 5 million won.56 The local party branches can raise dues,donations, or can carry out some other activities to raise income; however, they can onlybe subsidized by the central party and cannot receive subventions from the state directly.

The problem is that most of the local branches do not rely on membership dues buton the legislator of the area who is the leader of the local branch: in most cases, he is theonly source of finance. In terms of dues (membership fees), only a very small numberof members pay. In 2000, only 9.6% of the GNP’s central income was from members’dues, 13.1% from donations, and 61.5% from subventions out of a total income of about34 billion won. The NCNP received 2.8% from members’ dues, 53.1% from donations,and 24.4% from subventions out of a total income of about 75 billion won. For theULD, the ratio was 1.8% from members’ dues, 30.7% from donations, and 37.6% fromsubventions out of a total income of about 25 billion won.57 Overall, parties reliedon members’ dues on average 4.4%. As the ruling party, the NCNP, had the biggestproportion of donations and the ULD, as its coalition partner, also seemed to enjoymore donations as well (see Table 2).

55 Jiyun Sim and Minjeon Kim, ‘Hanguk Jeondangui Suipgujo Yeongu – 1997–2002 Nyeondo JeongdangHoegyebogo Jaryoreul Jungsimeuro [Revenue Structure of Korean Political Parties]’, HankukkwaKukjejeongchi, 38(3) (2002): 157.

56 Ibid.57 Ibid., 163.

386 youngmi kim

In terms of party income, the difference between the ruling party and theopposition party is significant. The GNP had the largest amount of subvention (about21 billion won) as a result of the large number of legislature seats compared to that of theNCNP (about 18 billion won). However, the total income was nearly half of the NCNP’sincome. This shows that political donors expect some benefit from their donations.

There is a higher rate of membership dues paid at the local branch of the partycompared to the central party. But this results from donations by the members ratherthan from membership dues. In 2000, the GNP’s 85,000 members paid 12.8 billion wonin dues.58 This means that an individual member paid 1.5 million won on average. TheNCNP’s 6,300 members paid 10 billion won, that is 1.6 million won per head. The ULD’s3, 700 members paid 4.7 billion won, that is 1.3 million won per head on average. In thelocal branch, the ratio of the subvention received from the central party is relativelylow at 23.5% for the GNP, 13.4% for the NCNP, and 4.5% for the ULD.59

As these figures show, the membership dues are not the same as the dues paid toso-called mass political parties in Western democracies. In Korea, membership duesare paid by very small numbers of members and parties are more likely to be dependenton legislators’ personal source of finance, donations, and subventions. Overall most ofthe party income is based on the small numbers of party leaders able to attract moredonations at the party’s central level and at the local level as well as the subvention andtheir own personal financial resources. To raise donations, candidates at the generalelection hold supporters’ meetings (Huwonhoe). The supporters’ meetings can be heldin various forms, as illustrated by inviting people to concerts, a party for a book launch,or exhibition of the candidate’s own calligraphy.60 Jang Seong-min, a former legislator,contends that:

Although the situation has changed slightly, even now but still supportersdonate funds under the table. Many supporters donate funds to prevent achange in the laws. The recent incident with the Private School Law is a case inpoint.61 Many involved in the private schools administration donated fundsto politicians in order to protect the existing laws.62

58 If a party member makes the donation, it seems to be classified as membership dues. However, theaverage amount of donation is far too large to call it party dues.

59 Ibid, 166.60 Some candidates for the general election write autobiographies before the election.61 The Roh Moo-hyun administration faced fierce opposition over the new reform policies on

Private School Laws, which aimed at introducing an open process of selecting board members.According to the new laws, a few members of the school board would have to be recommendedby parents and teachers associations of the school in order to prevent corruption by the schoolowners or founders. The GNP led by Park Geun-hye deserted the National Assembly meetingsopting for protesting outside the legislature (Oh My News, 9 December 2005). Available from:http://www.ohmynews.com/articleview/article_view.asp?no=263007&rel_no=1 [Accessed 5 October2006].

62 Interview held in Seoul on 8 March 2006


Taking care of local voters’ needs is a very important role played by the legislator whois the leader of the local branch. Each local branch spends an average of 10−30 millionwon a month on these sorts of activities.63 The main expenses for the local branch liein taking care of the family events of local voters.64 Attending and sending money to afuneral or wedding is a part of traditional culture in Korea to help each other on a happyday or sad day; however, most of the legislators who have local supporters in their areasspend more than two-thirds of their monthly expenses in taking care of the potentialvoters’ family events.65 Lee Yang-hee, a former legislator, has stated that ‘only a fewwho are in high positions have enough funds, especially in the ruling party. Ordinarylegislators mainly rely on their very close relatives as supporters. Being a legislator doesnot have advantages economically.’66

However, this comment actually contrasts with other legislators I interviewed. Aformer legislator of the ruling party stated that:

During the supporters meeting many supporters come. Imagine one supporterdonates 100 million won (about 100,000 US dollars) each. How much wouldit be? Yes, it costs 200 million won to 300 million won to run the local officewhere you are elected, but if you manage to have more than 10 supportersdonating 100 million won each, you still have 700 million won in your pocket.67

This shows how unequally the funding is allocated. Pye describes such leadership asfollows:

Korean rulers, like Korean fathers, are expected to be embattled, needingto prove themselves in adversary contacts; but they are also expected to bemasterful at all times, for like the Chinese leader-father, the Korean is supposedto be an aloof, lonely authority figure, able to cope single-handedly with all ofhis wishes. Yet again, he is expected to be sympathetic, nurturing, and sensitiveto the wishes of his followers’ family, though at the same time vicious andaggressive in fighting external foes.68

Similarly, the party leader takes care of the ordinary legislators with subsidies to sustainthe local branch and to carry out electoral campaigns or appointment of the candidatein the next election or executive position in the government in the case of the rulingparties. What happens if legislators do not conform to this pattern of behaviour? JangSeong-min, a former legislator, points out that:

63 Kukmin Ilbo, 11 May 1999; Seoul Sinmun, 10 May 1999.64 The political parties’ local branches were abolished in March 2004 during the Roh Moo-hyun

administration.65 People in South Korea gather for key family events but these not only involve the family but also friends

and acquaintances. On such occasions, guests typically bring money gifts. This kind of gathering showsthe person’s social status: that is, the higher position or wealthier he or she is the more money is given,the more flowers on display at a particular event and so forth.

66 Interview held in Seoul on 9 March 2006.67 Interview was held in Seoul on 10 March 2006.68 Geir Helgesen, Democracy and Authority in Korea: The Cultural Dimension in Korean Politics (Surrey:

Curzon Press, 1998).

388 youngmi kim

There are three kinds of punishment if ordinary legislators do not follow theleaders of the party. First, the allocation of the expenses for the local branchdepends on the legislators’ loyalty toward the leaders or elite members of theparty. When a legislator does anything against the leaders’ opinion the supportfunding for the local branch can be decreased right away. Second, when thereis an important meeting among legislators in the party the less loyal legislatorsare not contacted. They are simply marginalized in the legislators meeting inthe party. Third, there is no possibility of gaining a position within the partywhen he or she is not loyal to the leaders.69

Conclusion

Through an analysis of the case of the Kim Dae-jung administration in SouthKorea, the paper has shown that size (majority status most notably) is not a sufficientcondition for a government to secure governability. Instead, paying closer attentionto the dimensions of party organization offers better insights into the determinantsof governability (and ungovernability). A study of party organization in South Koreareveals that parties were created from above (by the elites during the American militaryoccupation), not from below (by the grassroots). When the party leadership failedto provide mechanisms for compromise with the opposition party, and the rulingparty faced deadlock in the legislature, governability was undermined. Failure by theruling party to consolidate the cohesiveness of the party organization contributedto factionalism within the party as well as the coalition and increased conflict withthe coalition partner. Without party consolidation within the ruling parties, thegovernment was unable to implement reform policies. As a result, the administrationrapidly entered the lame-duck period.

The findings of the paper have important implications for scholars of Koreanpolitics. Research on South Korea has thus far concentrated on the presidency andon the broader issue of democratic consolidation.70 Overlooking the contribution ofcoalition theories has meant understanding how political actors behave, and why andhow their behaviour actually influences the process of compromise and negotiation,has been neglected. Understanding the intra-party politics and the mechanisms ofnegotiation and compromise within the party and the party system in South Koreasignificantly enhanced our understanding of why coalitions form and especially breakdown in Korea and how that affects governability.

In addition, explaining coalition building in Korea, therefore, has implicationsthat go beyond the specificity of the case study. Coalition building is not just an issuethat concerns South Korea (or Western Europe, for that matter), but it is frequent in

69 Interview was held in Seoul on 8 March 2006.70 Youngho Kim, Hankuk Jeongdang Jeongchieu Ihae; Larry Diamond and Byung-Kook Kim, Consolidating

Democracy in South Korea (Colorado, London: Lynne Rienner Publishers, 2000); Chanpyo Park,‘Hanguk Jeongdang Minjuhwaronui Banseongjeok Seongchal.


East Asia (Taiwan, Japan), Central-Eastern Europe (Poland, Slovakia, Czech Republic,Romania), and Africa (Zambia). Therefore, the paper contributes to enhance ourunderstanding of factors affecting governability with coalition building across regimetypes.71 By emphasizing the importance of party politics (and of party organization mostnotably), the paper not only bridges two streams of scholarship (on coalition studiesin the parliamentary systems with West European countries data sets and democraticgovernability in the presidential regimes with Latin American data sets) but it showshow party organization, and in particular a low level of institutionalization, affectsgovernability.

In conclusion, the study of coalitions in South Korea adds a new non-Western caseto the expanding field of coalition studies as well as to the debates on regime typesand democratic governability allowing these two fields of scholarship to broaden theirgeographic scope, opening the way to new research across regions as well as acrossregime types in order to shed light on the determinants of ungovernability.

71 (Cheibub, Przeworski and Saiech, ‘Government Coalitions’; Cheibub, ‘Minority Governments’; Elgie,Divided Government; Figueredo and Limongi, ‘Presidential Power’; Randall and Svasand, ‘PartyInstitutionalization’.

Intra-party politics and minority coalition government in South Korea

Documents

Transcript of Intra-party politics and minority coalition government in South Korea