Changing cleavage structure in new democracies: An empirical analysis of political cleavages in Korea

Abstract

Cleavage structure cuts across the members of a state and affects the electorate's choice of parties and candidates. Social cleavage structure can be stable or go through changes depending on on-going social change as well as political parties' electoral strategies. With the collapse of communism and the advent of the third wave of democracy, many countries in Eastern Europe, Latin America, and Asia have experienced rapid social change. New political parties have formed, and democratic elections have been introduced in most of these countries. With rapid social change and the resulting shifts in party strategies, do we also expect rapid change in social cleavage structure in these countries? Or does cleavage structure change only gradually and remain stable over a short-term period? In this paper, we attempt to answer these questions by examining the relative importance of three cleavages commonly claimed to have relevance to politics in the 2000s—regionalism, ideology, and generational differences—in voters' choices in Korea. We analyze the voting behavior of the electorate in the 16th and 17th National Assembly elections, in 2000 and 2004, respectively, by employing multinomial logistic analyses of voter survey data. Our analysis indicates that the relative importance of various cleavages changed during the early 2000s: the influence of regionalism remained strong, but its intensity declined; the importance of ideology grew; and a new cleavage of generational differences appeared. Our study shows that the cleavage structure is fluid even in a short-term period, especially in new democracies, as citizens and elites adjust their behaviors in response to changing social situations. Political parties may also introduce a particular electoral strategy to induce change in the cleavage structure.

Introduction

Cleavage structure cuts across the members of a state, affecting the electorate's choice, and thus, the resulting party system (Schattschneider, 1960, Lipset and Rokkan, 1967, Rae and Taylor, 1970). Examples of such cleavages include class, religion, ethnicity, and language, among others. Party systems are the institutions designed to link citizens and leaders, aggregating the electoral resources of the citizens (McDonald and Budge, 2005; Kim et al., 2006). Party system configurations open up some possibilities for representation and constrain others. On one hand, they tend to have great continuity across elections (Lipset and Rokkan, 1967) because of the weight of voter images and organizational structure, yet they are also capable of undergoing sharp changes as new issues emerge or new electoral strategies are chosen (or eschewed; see Kitschelt, 1994). In the end, social cleavage structure can be stable or go through changes depending on on-going social change and parties' electoral strategies.

With the collapse of communism and the advent of the third wave of democracy, many countries in Eastern Europe, Latin America, and Asia have experienced rapid social change. New political parties have formed, and democratic elections have been introduced in most of these countries. With rapid social change and the resulting shifts in party strategies, do we also expect rapid change in social cleavage structure? Or does cleavage structure change only gradually and remain stable over a short-term period? These are intriguing questions for political scientists, but until now, few studies have looked at changing or stable cleavage structures in new democracies. In this paper, we attempt to analyze cleavage structures utilizing data from one of those new democracies, South Korea (Korea hereafter).

Korea has a relatively homogeneous population, sharing ethnicity, language, and historical traditions. The authoritarian rule imposed on the Korean people, first by General Park Chung-hee (1961–1979), then by General Chun Doo-hwan (1979–1987), suppressed the emergence of truly social cleavages; the single most dominant issue in Korean politics was the authoritarianism-democracy divide until 1987, when the transition to democracy began.

With democratic opening and free elections came regionalism as the dominant cleavage in Korean politics. This was especially the case in the rivalry between the southeastern region of Youngnam and the southwestern region of Honam, due to the electoral strategies of political parties. Political leaders who emerged as alternatives to authoritarian rule were perceived as representing certain regions of Korea without nationwide appeal and with their party support firmly entrenched in those regions. It is widely agreed among observers of Korean politics that since the democratic opening, electoral competition has been used as an expression of regional frustration and animosity, with regionalism playing the decisive role in the choices the Korean electorate made (Chung, 1993, Park, 1993, Choi, 1995; Lee, 1998, Lee, 1999). Table 1, which shows the percentage of National Assembly seats obtained by each of the major parties since the 13th National Assembly elections in 1988, exemplifies the importance of regionalism for the party support of the electorate.

However, for various reasons, it was largely expected that the impact of regionalism would decline in the 17th National Assembly elections scheduled for 2004: first, three key charismatic leaders who single-handedly controlled their own regionally-based parties had retired from politics by this time (two of them after having served as President). This led to adjustments in each party's electoral strategy to maintain the same level of political support under different circumstances; second, candidate Roh Moo-hyun of the New Millenium Democratic Party (NMDP) won the presidential election in 2002. The NMDP was a regional party representing the Honam region, but Roh himself came from the rival Youngnam region, thus diluting the effect of regionalism in politics; third, it was widely viewed that Roh Moo-hyun was elected with support primarily from the younger generations and the more progressive elements of Korean society, thus paving the way for ideology and generational issues to play greater roles in Korean politics.

Since the National Assembly elections in 2004, many students of Korean politics have debated the impacts of newfound cleavages of ideology and generational differences, as well as the potentially declining role of regionalism on Korean politics. However, there is little agreement among scholars, and the existing studies seem to have methodological deficiencies (Kang, 2004, Lee and Leem, 2004). There were more than two major parties competing in the National Assembly elections in Korea from which voters could choose. Therefore, in order to properly delve into what determined the electorate's choice among several candidates, it is crucial to employ a statistical method that can deal with multiple choices. In this paper, we attempt to do so by employing multinomial logistic analysis, a statistical method designed to accommodate nominal dependent variables with more than two categories. We will specifically state and test our hypotheses using appropriate data. By doing so, we will examine the relative importance of regionalism, ideology, and generational differences in voters' choices and investigate the possibility of changing cleavage structure in Korea.

In the following section, we identify the factors impacting the electorate's decisions in Korea and develop hypotheses based on our identifications. Then, we present statistical models to test our hypotheses with the description of the data, methodology, and the measures we use. We analyze the voters' choices in the 16th and 17th National Assembly elections in 2000 and 2004, respectively, using multinomial logistic models. We conclude the paper by summarizing our findings about Korea and by discussing their implications for the stability or fluidity/volatility of cleavage structure in new democracies.