[EAI Issue Briefing] Voter Ideological Composition and Presidential Candidate Preference in the 20th Presidential Election: Generational Divide, Class Betrayal, and Polarization

1. High Election Interest Amidst a Two-Party Competition

Contrary to concerns that voters might become indifferent to the election due to an unprecedented "dislike" contest, respondents showed relatively high average election interest. Among the 1,515 respondents, only 7.8% expressed no interest in the current election, while approximately 92.1% responded that they were "generally interested (34.1%)" or "very interested (58.0%)," confirming high interest in the March presidential election through this survey.

[Figure 1] Election Interest by Party Support

However, a notable observation by age group is the high rate of indifference (21.5%) among the 18-29 age group, unlike other age groups that show consistently high election interest. Furthermore, by ideological orientation, voters who identified as ideologically centrist also showed relatively high election interest (89.1%), similar to progressive and conservative voters. In contrast, by party support, not only those who are unaffiliated (none/don't know/no response) but also supporters of the Justice Party or other minor parties showed relatively low election interest.

This difference in election interest by party affiliation can also be analyzed as a consequence of the election being dominated by a two-party competition, centered around the ruling Democratic Party's "regime continuation" and the main opposition People Power Party's "regime change" narratives, which are rallying voters.

This trend is further confirmed in the survey on voting intention, where the proportion of respondents who answered "I will definitely vote" was relatively high among the progressive (89.1%) and conservative (88.8%) groups. However, it decreased to 79.5% among centrist voters. Similarly, by party support, the proportion was lower among supporters of the Justice Party (85.8%) and unaffiliated voters (64.3%) compared to supporters of the People's Party (90.4%), the Democratic Party (90.2%), and the People Power Party (89.4%).

2. Presidential Candidate Support by Voter Generation, Ideology, Region, and Class: Clear Generational Divide and "Class-Betraying Votes"?

[Figure 2] Presidential Candidate Supported (Candidate to Vote For)

Although referred to as an election of "dislikes," the survey results indicate that the presidential election is clearly unfolding as a two-way race between the Democratic Party's candidate Lee Jae-myung and the People Power Party's candidate Yoon Suk-yeol. The strong performance of the People's Party candidate Ahn Cheol-soo, who recorded a double-digit approval rating (12.7%), is noteworthy.

However, when categorized by party support, while the approval ratings for Lee Jae-myung among Democratic Party supporters and Yoon Suk-yeol among People Power Party supporters were high (83.5% and 82.2%, respectively), the fact that Ahn Cheol-soo's support among People's Party supporters was 64.8%, with a significant portion (18.1%) supporting the conservative candidate Yoon Suk-yeol, is worth noting. Meanwhile, those who had no candidate to support or selected "don't know/no response" accounted for only about 11%.

[Figure 3] Supported Presidential Candidate by Respondent's Generation

As is already known from various public opinion polls, distinct differences in candidate support are observed across age groups. While the Democratic Party's candidate Lee Jae-myung receives overwhelming support, particularly among those in their 40s and 50s, the People Power Party's candidate Yoon Suk-yeol shows high support among the 60+ age group.

Among the so-called MZ generation (those in their 20s and 30s), Yoon Suk-yeol recorded higher support than Lee Jae-myung. Specifically, in the 18-29 age group, Yoon Suk-yeol led Lee Jae-myung by more than 10 percentage points.

These differences in preferred presidential candidates by age group are also confirmed in the survey on "candidates I would absolutely not vote for if the election were tomorrow." In fact, respondents in their 40s and 50s showed high rates of not voting for Yoon Suk-yeol (54.3% for 40s, 45.6% for 50s), while respondents in their 60s showed a high rate of not voting for Lee Jae-myung (45.9%).

[Table 1] Supported Presidential Candidate by Gender within the MZ Generation (%)

The relatively high support for Yoon Suk-yeol among the MZ generation is presumed to be related to the conservatism of young men in their 20s, often referred to as "lee-dae-nam." Indeed, when candidate support is broken down by generation and gender, while women in their 20s show dispersed preferences (Lee Jae-myung 21.6%, Yoon Suk-yeol 13.7%, Sim Sang-jung 14.2%, Ahn Cheol-soo 18.2%), men in their 20s show very high support for the People Power Party's Yoon Suk-yeol (49.6%), followed by Lee Jae-myung (16.4%), Sim Sang-jung (1.5%), and Ahn Cheol-soo (14.2%). The combined support for Yoon Suk-yeol and Ahn Cheol-soo, candidates who can be classified as ideologically conservative or center-right, among men in their 20s, confirms the political conservatism of this demographic.

A similar pattern is observed among voters in their 30s when categorized by gender. While the support gap between Lee Jae-myung and Yoon Suk-yeol among women in their 30s is not large (Lee Jae-myung 29.7% vs. Yoon Suk-yeol 28.5%), it widens among men (Lee Jae-myung 31.8% vs. Yoon Suk-yeol 42.4%).

[Figure 4] Favorability of Politicians by Generation

The generational differences in candidate preference are further confirmed by the politician favorability survey. On an 11-point scale from 0 (strongly dislike) to 10 (strongly like), President Moon Jae-in and the main candidates in the current presidential election received favorability scores. Notably, high favorability (5 points or above) for President Moon Jae-in was clearly observed among those in their 40s and 50s. Conversely, favorability for President Moon Jae-in was lower among the 18-29 age group and those aged 60 and above.

The generational favorability pattern for the Democratic Party's candidate Lee Jae-myung is similar to that of President Moon Jae-in, who belongs to the same party. However, except for those in their 50s, where President Moon and Lee Jae-myung received the same score (5.1 points), Lee Jae-myung's favorability was lower than President Moon's across all age groups.

Furthermore, high favorability for Yoon Suk-yeol was observed among the elderly (5.5 points) and subsequently among the 18-29 and 30s age groups.

Meanwhile, the People's Party's candidate Ahn Cheol-soo received relatively positive favorability ratings across all age groups, particularly among the 18-29 age group, where he recorded the highest favorability. He also received high favorability among the elderly (60+), second only to Yoon Suk-yeol. Sim Sang-jung, on the other hand, showed low favorability (around 3.5-3.9) across all age groups.

[Figure 5] Supported Presidential Candidate by Respondent's Ideological Orientation

Presidential candidate preference is also divided by voter's ideological orientation. As expected, progressive voters show a high preference for the Democratic Party's Lee Jae-myung, while conservative voters show high support for Yoon Suk-yeol.

The People's Party's candidate Ahn Cheol-soo, who advocates for ideological centrism, shows relatively high support among centrist voters. However, the candidate most preferred by centrists is Lee Jae-myung (35.3%), followed by Yoon Suk-yeol with 29.0% support.

However, when the question format is changed to "candidates I will absolutely not vote for" (negative framing) instead of "candidate to vote for," the gap in support (dislike) between Lee Jae-myung and Yoon Suk-yeol among centrist voters narrows slightly (34.4% and 33.9%, respectively).

[Figure 6] Politician Favorability by Ideological Orientation

When categorized by ideological orientation, high favorability for Lee Jae-myung is particularly evident among voters who identify as ideologically progressive. However, this high favorability (6.5 points) is still lower than that of President Moon Jae-in among progressives (7.3 points). Furthermore, Lee Jae-myung's favorability falls short of President Moon's across all ideological groups.

Yoon Suk-yeol received positive favorability ratings primarily from the conservative group. Ahn Cheol-soo received the most favorable assessment from the centrist group and also garnered relatively positive evaluations from conservative voters.

[Figure 7] Supported Presidential Candidate by Respondent's Residential Region

Meanwhile, regionalism is confirmed in this presidential election, with differences in candidate support based on the respondent's region of residence. While the support gap between Lee Jae-myung and Yoon Suk-yeol is not significant in the Seoul Metropolitan Area (Seoul, Incheon, Gyeonggi) and the Chungcheong region, the gap between the top two candidates is primarily observed in the Honam and Daegu/Gyeongbuk regions. Notably, Lee Jae-myung's high support (61.0%) in the Honam region, with relatively low support for Yoon Suk-yeol and Ahn Cheol-soo (12.2% and 11.9%, respectively), is impressive.

In the PK region (Busan, Ulsan, Gyeongnam), Yoon Suk-yeol leads Lee Jae-myung in support, but the gap is not as large as in the TK region. This indicates a weakening of regionalism in the Busan, Ulsan, and Gyeongnam areas, similar to trends observed in elections since 2016.

[Figure 8] Supported Presidential Candidate by Respondent's Socioeconomic Class

Furthermore, when voters are categorized by household income to examine presidential candidate preferences, the phenomenon of "class-betraying votes," observed in various past studies, appears to be replicated in this election.

As is well-known, the People Power Party's Yoon Suk-yeol receives relatively high support from the group with household incomes of 3 million won or less. In contrast, the Democratic Party's Lee Jae-myung consistently garners high support from the group with household incomes of 4 million won or more. Notably, Lee Jae-myung's relatively high support among the highest income bracket (households earning 6 million won or more) is impressive.

This tendency for class-betraying votes is further confirmed in the "candidates I will absolutely not vote for" data. Among the income group below 2 million won, the proportion of respondents who stated they would absolutely not vote for Lee Jae-myung was highest at 43.7% (Yoon Suk-yeol recorded 23.5% in the same group). Conversely, Yoon Suk-yeol showed high levels of dislike (exceeding 40%) among the high-income bracket, specifically in the 5-6 million won income group (43.6%), 6-7 million won group (49.7%), and 7 million won or more group (40.1%).

3. Voter Ideological Landscape and Partisan Polarization: Partisan Sorting, Generational Gender Gap, and Emotional Polarization

[Figure 9] Distribution of Party Support and Ideological Orientation

Based on data where respondents were asked to rate their ideological orientation on a 10-point scale from 0 (very progressive) to 10 (very conservative), as of January 2022, centrist voters (5 points) still constitute the largest proportion (40.4%) in South Korea, surpassing progressives (25% for points 1-4) and conservatives (33% for points 6-10). As is well-known, the proportion of conservative voters slightly exceeds that of progressive voters.

Furthermore, when voters are grouped by party support, a thick distribution of centrist voters is observed across all party groups. Nevertheless, Democratic Party and Justice Party supporters tend to have a more concentrated progressive base, while supporters of the People Power Party and the People's Party have a more concentrated conservative base. In other words, a relatively clear partisan sorting is evident among South Korean voters.

Meanwhile, when respondents were asked to rate the ideological orientation of the main presidential candidates on a scale of 0-10, the candidates were ranked from most progressive to most conservative as follows: Lee Jae-myung (3.1), Sim Sang-jung (3.2), Ahn Cheol-soo (5.5), and Yoon Suk-yeol (7.3). Although the gap is not large, it is interesting that the general public perceives the Democratic Party's Lee Jae-myung as a progressive candidate comparable to the Justice Party's Sim Sang-jung.

However, the perception of candidates' ideological orientation by voters varies depending on their own political orientation (e.g., party support and self-identified ideology). For instance, People Power Party supporters perceive Lee Jae-myung as a considerably radical candidate (rated 2.0, lower than Sim Sang-jung's 3.2), while Democratic Party supporters perceive Lee Jae-myung as a relatively moderate progressive (3.8). Furthermore, Justice Party supporters tend to perceive Yoon Suk-yeol as closer to the extreme right than other party support groups.

[Table 2] Evaluation of Main Candidates' Ideological Orientation by Party Support

Interestingly, the People's Party's candidate Ahn Cheol-soo consistently receives an evaluation of being ideologically close to the center from all party support groups, which is noteworthy.

Similarly, the perception of the main presidential candidates' ideological orientation differs based on the respondent's own ideological orientation. For example, conservative voters perceive Lee Jae-myung as more ideologically radical compared to other ideological groups. Progressive voters, on the other hand, perceive Yoon Suk-yeol as closer to the extreme right.

Furthermore, progressive voters perceive the Justice Party's Sim Sang-jung as more moderately progressive (3.6 vs. 3.3 & 2.7) than other ideological groups. Conversely, they perceive Ahn Cheol-soo, who received a centrist evaluation (5.3 points) from other ideological groups (centrist or conservative), as more conservative (6.0).

The conservatism of men in their 20s and the gender gap in ideological orientation within this age group, confirmed by recent opinion polls, are also evident in this survey. As shown in [Figure 11], which presents the average ideological orientation of male and female groups by generation, the 18-29 age group exhibits characteristics of significantly conservative men, unlike the generally progressive women in the same age group (Average for men in their 20s: 5.9, Average for women in their 20s: 4.6).

The average ideological orientation for women (4.9) and men (5.5) in their 30s also appears relatively high compared to other age groups. In contrast, the ideological average for women and men in their 40s was 4.8 and 4.7, respectively, showing the smallest gap among the age groups.

[Figure 10] Generational Gender Gap in Ideological Orientation

[Figure 11] Ideological Orientation and Candidate Favorability Gap[1]

Meanwhile, examining the gap in favorability (0-10 points) between Lee Jae-myung and Yoon Suk-yeol, the top two candidates and presidential nominees of the major parties, based on favorability ratings, provides insight into the emotional evaluation of the candidates by different voter groups. As expected, the favorability gap between the two candidates is smallest on average among centrist voters (4.6), while the gap in favorability ratings is larger among progressive and conservative groups.

This result reaffirms the importance of appealing to ideologically centrist voters for the two leading candidates to win this election.

[Figure 12] Generational Gap in Presidential Candidate Favorability (Lee Jae-myung - Yoon Suk-yeol)[2]

When the gap in candidate favorability ratings is analyzed by age group, a trend of increasing gap with increasing age is observed. Consequently, the favorability gap between the two candidates is smallest in the 18-29 age group (3.0) and largest in the 60+ age group (6.3).

In other words, as age increases, voter groups show more stable candidate preferences. Conversely, in the case of the 2030 generation, the difference in candidate preference is not significant, underscoring the necessity for competing candidates to secure the votes of the 2030 generation to win this election.

When the gap in candidate favorability is divided by party affiliation, the largest gap (6.4) is observed among supporters of the two major parties, particularly the Democratic Party, followed by supporters of the People Power Party (5.9), and then by supporters of the Justice Party, the People's Party, and other parties in descending order.

[Figure 13] Favorability Gap Between Presidential Candidates by Party Affiliation (Lee Jae-myung - Yoon Suk-yeol)[3]

4. Evaluation of the Moon Jae-in Administration and Prospects for Election Victory by Political Inclination

When asked to evaluate the Moon Jae-in administration's performance on a scale of 100, supporters of the Democratic Party's candidate, Lee Jae-myung, as expected, gave the highest score (72.9 points), while supporters of the People Power Party's candidate, Yoon Suk-yeol, gave the most negative evaluation (24.3 points).

Supporters of the Justice Party, who lean progressive, gave relatively high positive evaluations of the Moon Jae-in administration, following those of the Democratic Party supporters. In contrast, evaluations of the Moon Jae-in administration by voters for the People's Party, other parties, and unaffiliated voters were predominantly negative, scoring below 50 points.

[Figure 14] Evaluation of Moon Jae-in Administration's State Operations by Preferred Candidate

Furthermore, when examining the question of predicting the winner of the next presidential election by party affiliation, while the general pattern of high victory probability for candidates of one's own party is observed as expected, the high confidence in victory among Democratic Party supporters is particularly noteworthy.

In contrast to Democratic Party supporters who view Yoon Suk-yeol's chances of winning as quite low (5.0%), it is notable that among People Power Party supporters, Lee Jae-myung's chances of winning were surveyed at 14.6%. Additionally, Lee Jae-myung's chances of winning were overwhelmingly high among supporters of other parties and unaffiliated voters.

Conversely, supporters of the People's Party do not significantly differentiate between the chances of victory for third-party candidates, including Ahn Cheol-soo. However, within this context, it is also noteworthy that they perceive Yoon Suk-yeol's chances of winning as higher than those of Lee Jae-myung or Ahn Cheol-soo.

This pattern is reconfirmed when voters are categorized by their preferred candidate rather than their party affiliation. High confidence in election victory is observed among the group of voters supporting Lee Jae-myung. Furthermore, it is quite peculiar that the possibility of Lee Jae-myung winning is considered highest across all candidate categories except for Yoon Suk-yeol's supporters.

[Figure 15] Perception of Election Victory Probability by Party Affiliation

[Figure 16] Perception of Election Victory Probability by Preferred Candidate

5. Summary and Conclusion

Despite concerns about an unprecedented presidential election characterized by 'unfavorability,' voter interest in the election was found to be considerably high. However, considering the phenomenon of somewhat lower election interest centered among ideological moderates, unaffiliated voters, or supporters of other parties, it is inferred that the current presidential election race is unfolding with voters rallying around the two main axes of 'regime change theory' and 'regime perpetuation theory.'

As has been reported multiple times, generational divides in party affiliation or preferred candidates, which have been observed in previous elections, are also confirmed in this survey. While the 40-50 age group shows a high preference for Lee Jae-myung, strong support for Yoon Suk-yeol is found among the elderly (60+). Notably, and intriguingly, Yoon Suk-yeol significantly leads Lee Jae-myung in preference among the 20s generation, particularly within the MZ generation. This so-called conservative bias among the MZ generation is explained by the conservative shift in political preferences, especially among young men in their 20s.

Voter preferences for competing presidential candidates, as expected, vary according to individual ideological and political inclinations. Lee Jae-myung receives positive evaluations from progressive voters, but his favorability rating is lower than that of former President Moon Jae-in, who belongs to the same party. Yoon Suk-yeol and Ahn Cheol-soo primarily receive support from conservative voters.

When respondents are categorized by household income, a phenomenon similar to 'class-defying' preferences is observed, consistent with past trends. For instance, Lee Jae-myung garners strong support from high-income earners (household income of 6 million KRW or more), while the conservative candidate Yoon Suk-yeol receives greater support from relatively low-income earners (below 3 million KRW).

As of 2022, while a substantial moderate segment still exists among South Korean voters, there is a linkage between voters' party affiliation and ideological preferences, meaning partisan alignment allows for some differentiation in ideological preferences based on an individual's party preference.

Furthermore, when differentiating respondents' ideological inclinations based on generation and gender, a distinct gender gap in ideological preference is observed in the 18-29 age group compared to other age cohorts, reflecting the conservative tendencies of so-called '20-something males' (이대남). In contrast, the gender gap in ideological preference is smallest in the 40s age group, which can be considered a strong support base for Lee Jae-myung.

Examining voter groups based on the favorability gap between the two main presidential candidates, the Democratic Party's voter group exhibits the most stable preference for Lee Jae-myung. As expected, the smallest favorability gap between Lee Jae-myung and Yoon Suk-yeol is observed among supporters of other parties or unaffiliated voters (none/don't know/no response).

Additionally, when examining the favorability gap between the two candidates by age group, the gap is smallest in the 18-29 age group compared to other age cohorts. Based on this finding, it is possible to understand to some extent why competing candidates are making concentrated efforts to secure the support of the 2030 generation in this presidential election.

Furthermore, when examining the probability of winning the next presidential election by party affiliation, the general pattern of high victory probability for candidates of one's own party is observed as expected. However, the high confidence in victory among Democratic Party supporters is particularly evident.

Finally, voters who believe that domestic politics and welfare should be the top priorities for the next government tend to vote for Lee Jae-myung, while those who believe international politics and tax reform should be the priorities tend to vote for Yoon Suk-yeol. In contrast, issues related to job creation and economic growth, which have traditionally served as criteria for distinguishing progressive and conservative voters and have significantly influenced the determination of preferred candidates, are currently not being effectively claimed by either Lee Jae-myung or Yoon Suk-yeol. ■

[1] The favorability gap here refers to the absolute difference between the favorability rating for candidate Lee Jae-myung (0-10) and candidate Yoon Suk-yeol (0-1) for each respondent. The scores for each group shown in the graph represent the average favorability gap calculated for each ideological group.

[2] The favorability gap here refers to the absolute difference between the favorability rating for candidate Lee Jae-myung (0-10) and candidate Yoon Suk-yeol (0-1) for each respondent. The scores for each group shown in the graph represent the average favorability gap calculated for each generational (age group) group.

[3] The favorability gap here refers to the absolute difference between the favorability rating for candidate Lee Jae-myung (0-10) and candidate Yoon Suk-yeol (0-1) for each respondent. The scores for each group shown in the graph represent the average favorability gap calculated for each party affiliation group.

■ Author: Lee Jae-mook_Associate Professor, Department of Political Science and International Relations, Hankuk University of Foreign Studies; Director of Public Relations, Hankuk University of Foreign Studies; Head of the BK21 Global Democracy and Human Security Research Team, Department of Political Science and International Relations. He obtained his Ph.D. in Political Science from the University of Iowa and currently serves as the Education Director of the Korean Political Science Association and the Secretary-General of the Korean Association of Party Studies. His main research areas include political behavior, political processes, and American politics. His recent co-authored works include "Korean Parties and Democracy Diagnosed in the Political Arena" (2018, co-authored), "American Politics and East Asian Foreign Policy" (2017, co-authored), and "ROK-US Politics: Challenges and Changes" (2014, co-authored).

■ Editor:Jeon Ju-hyun_EAI Researcher

Inquiries: 02 2277 1683 (ext. 204) | jhjun@eai.or.kr