[Public Opinion Brief 81-1] Vote Changes and Their Causes in 5 Regions during the June 2nd Local Elections

[Public Opinion Brief 81] EAI · JoongAng Ilbo · SBS · Korea Research 2010 Local Election Panel Survey

2nd Survey of the 5-Region Panel Survey for the June 2nd Local Elections

1. Vote Changes and Their Causes in 5 Regions during the June 2nd Local Elections

2. Preliminary Analysis of the 2nd 5-Region Panel Survey

EAI, JoongAng Ilbo, SBS, and Korea Research plan to conduct three panel surveys from before the election to immediately after, targeting 600 respondents in five regions—Seoul, Gyeonggi, Gyeongnam, Chungnam, and Jeonbuk—recruited in proportion to the population. This survey presents the results of the second investigation. The aim is to track how the sentiments of voters in these regions change over time.

Cheonan Incident Aftermath: Strong Winds in the Seoul Metropolitan Area, Light Breezes in Chungnam/Gyeongnam

Why Did the Consensus on Judging the Administration Fail to Translate into Votes in the Seoul Metropolitan Area?

Opposition Parties Misjudged the Nature of Local Elections, Where National and Local Winds Intersect

Jeong Han-wool (EAI Center for Public Opinion Analysis, Deputy Director)

The June 2nd local elections can be summarized by the keyword 'wind election.' The opposition parties consistently began their campaigns relying on the 'administration judgment' wind, and continued with the 'Roh wind.' Recent opinion polls predicting the Grand National Party's advantage in all regions except Chungnam and Gyeongnam, which are the respective regional strongholds of the parties, have lent credibility to the analysis that the 'North wind,' triggered by the sinking of the Cheonan, has overshadowed the 'Roh wind.' In this survey, only 35.9% of respondents stated they would consider the 'Roh wind' as a voting criterion, while 54.1% indicated they would consider the Cheonan incident.

The survey results show a clear pattern where respondents who use the Cheonan incident as a criterion for their voting choice support the Grand National Party, while those who consider the 'Roh wind' support the Democratic Party or opposition candidates. Therefore, the analysis that the 'Roh wind' was obscured by the Cheonan incident is partially valid. Indeed, a comparison between the first panel survey conducted from May 10-13 in Seoul, Gyeonggi, Gyeongnam, Chungnam, and Jeonbuk, and the second panel survey conducted from May 24-26, during the period when opinion polls were permitted, reveals a trend of rising presidential approval ratings and public opinion favoring stability over a check on power across all regions. The 'rally around the flag' effect, where support for the incumbent government and ruling party increases during external crises, is clearly evident. It is not unreasonable for the opposition parties and a considerable portion of the public to believe that the government's announcement regarding the Cheonan incident contained political intentions.

Looking at presidential approval ratings, Seoul saw an increase of 5.2 percentage points from 49.3% in the first survey conducted two weeks prior to 54.1%, and Gyeonggi saw an increase of 10.8 percentage points from 42.5% to 53.3%. Chungnam also rose by 5.0 percentage points from 41.3% to 46.3%, and Jeonbuk by 5.0 percentage points from 30.9% to 35.9%, similar to the increase in Seoul. Gyeongnam saw a relatively small increase from 50.3% to 52.4%. The rise in approval ratings for state affairs discourages the 'administration judgment' argument aimed at winning over voters who view the elections as a mid-term evaluation of the administration. Furthermore, an average of 88.7% across the five regions agreed with the call for bipartisan cooperation on the Cheonan issue, which placed constraints on the opposition parties' ability to confront the ruling party over the Cheonan issue.

However, even after the announcement regarding the Cheonan incident, public opinion viewing the elections as a mid-term evaluation of the administration persisted among voters, and there was strong antipathy towards the 'administration judgment' argument advocated by the ruling party. Regarding the statement, 'The Lee Myung-bak administration's failures must be judged in these elections,' 59.1% in Seoul, 67.3% in Gyeonggi, 63.2% in Chungnam, 72.0% in Jeonbuk, and 63.8% in Gyeongnam agreed. In contrast, only 26.0% in Seoul, 28.1% in Gyeonggi, 29.2% in Chungnam, 26.8% in Jeonbuk, and 25.3% in Gyeongnam agreed with the judgment of the Roh Moo-hyun administration. Despite the announcement regarding the Cheonan incident, the situation was not unilaterally disadvantageous for the opposition parties.

The points of interest, therefore, are why the broad consensus for a mid-term evaluation does not translate into votes, and why, among pro-Roh candidates, Han Myeong-sook and Yoo Si-min are struggling in the Seoul metropolitan area, while Ahn Hee-jung and Kim Doo-kwan are still competing for the lead and performing well in Chungnam and Gyeongnam despite the Cheonan incident. While explanations suggest that the Seoul metropolitan area is more sensitive to national politics due to its political and geographical characteristics, and that regions geographically distant may be less sensitive to North-South confrontation issues, there is a disparity in public opinion even within the Seoul metropolitan area between Seoul and Gyeonggi Province. Moreover, the impact of the Cheonan incident in local areas, particularly Gyeongnam, was not insignificant compared to the Seoul metropolitan area.

[Figure 1] Changes in Approval Ratings Based on Consideration of the Cheonan Incident for Voting

[Figure 2] Changes in Approval Ratings Based on Consideration of the 'Roh Wind' for Voting

So, what is the reason why the considerable atmosphere of government scrutiny and judgment does not translate into votes? What are the other factors that explain the struggles of opposition candidates in the Seoul metropolitan area and their strong performance in Chungnam and Gyeongnam?

First, candidates Han Myeong-sook and Yoo Si-min have consistently engaged in election campaigns that rely on the national political confrontation and winds, such as 'administration judgment,' rather than focusing on the achievements of their political activities based on their regional bases. In this survey, when candidates' images were assessed, both Han Myeong-sook and Yoo Si-min lagged behind Oh Se-hoon and Kim Moon-soo in terms of trustworthiness, integrity, and approachability, with a particularly large gap in the evaluation of their administrative capabilities. Both candidates succeeded in unifying the opposition to advocate for 'administration judgment,' thereby narrowing the gap in support rates to some extent before the campaign. However, they appear to have fallen short in inspiring confidence not only in their vision and capabilities as national politicians but also in their vision for regional development and their administrative abilities. This contrasts with Ahn Hee-jung, who is based on the region's biggest issue, Sejong City, and Kim Doo-kwan, who is considered to have established deep roots in the region by running for Gyeongnam Governor for the third time [See pp. 13-14].

Second, local elections are not solely determined by the national political frame; they are also a venue for evaluating the local economic situation and, most importantly, not only the assessment of the central government but also the performance of local governments that have led local autonomy for the past four years. In fact, we analyzed the evaluations of the past four years of mayoral/gubernatorial administration by those in Seoul and Gyeonggi who agreed with the judgment of the Lee Myung-bak administration. Among the 244 individuals in Seoul who agreed with the current administration judgment, nearly 50% (121 people) evaluated Mayor Oh Se-hoon positively. Among those who agreed with the MB judgment but thought Oh Se-hoon performed well, 39.7% indicated they would vote for Oh Se-hoon, which is higher than the 34.7% who said they would vote for Han Myeong-sook. In Gyeonggi, among those who agreed with the MB judgment, 45.2% who had a positive evaluation of Governor Kim's administration stated they would support Kim Moon-soo, which was higher than the 36.3% who said they would support Yoo Si-min. This suggests that the evaluation of the incumbent is prioritized over the administration judgment argument. In contrast, the fact that Lee Wan-koo and Kim Tae-ho, who received considerable support due to positive evaluations, were not candidates in Chungnam and Gyeongnam, where Ahn Hee-jung and Kim Doo-kwan are performing well, due to the Sejong City issue and internal adjustments, likely played a significant role.

Ultimately, while national political confrontation, such as the 'administration judgment' argument, has a considerable impact on local elections, a significant number of voters also weigh the inherent meaning of local elections—who will better lead the city and provincial administration—rather than engaging in blind voting regardless of candidates or policies based solely on whether they support the 'administration judgment' argument. This is the core reason why Ahn Hee-jung and Kim Doo-kwan, who have regional bases and are rooted in local issues, are performing well, while Han Myeong-sook and Yoo Si-min, who switched positions shortly before the election, are struggling.

[Figure 3] Voting Preference Differences Among MB Judgment Proponents Based on Mayor Oh Se-hoon's Administrative Evaluation

[Figure 4] Voting Preference Differences Among MB Judgment Proponents Based on Governor Kim Moon-soo's Administrative Evaluation

In Close Electoral Districts, the Votes of Undecided Voters Will Decide the Outcome

Lim Seong-hak (University of Seoul)

The second survey of the local election panel survey was conducted from May 24th to 26th in Seoul, Gyeonggi, Chungnam, Jeonbuk, and Gyeongnam. Among these, Chungnam and Gyeongnam, which are closely contested regions, were found to have a higher proportion of undecided voters compared to Seoul, Gyeonggi, and Jeonbuk. The percentage of undecided voters who have not yet decided whom to vote for is currently 15.0% in Seoul, 16.4% in Gyeonggi, and 13.0% in Jeonbuk. In contrast, it is 23.3% in Chungnam and 32.3% in Gyeongnam. Consequently, the election outcomes in Chungnam and Gyeongnam are more difficult to predict than in other elections, and the candidate who can sway the votes of these undecided individuals is expected to win.

So, who are these undecided voters? Panel surveys, which allow for tracking changes in voters' preferred candidates by surveying the same individuals multiple times, are the only method capable of revealing their characteristics and tendencies. By analyzing cases where undecided voters in the first survey decided on a specific candidate, we can identify their characteristics and trends.

Ahn Hee-jung in Chungnam and Kim Doo-kwan in Gyeongnam, who gained higher approval ratings than their competitors in the previous survey, have succeeded in converting undecided voters into their supporters relative to their rivals. In Ahn Hee-jung's case, out of 153 undecided voters in the previous survey, 50 (32.7%) were newly absorbed as supporters in this survey, whereas Park Sang-don absorbed only 31 (20.3%). In Gyeongnam, out of 186 undecided voters in the first survey, 18.3% (34 people) shifted their support to Lee Dal-gon, but 43 people (23.1%) shifted their support to Kim Doo-kwan. The movement of undecided voters over the past two weeks has worked in favor of Ahn Hee-jung and Kim Doo-kwan.

Furthermore, among the undecided voters from the first survey who expressed a strong intention to vote, Ahn Hee-jung was supported by the largest proportion (37.9%) in Chungnam, and Kim Doo-kwan received the highest support (33.3%) in Gyeongnam. If this trend continues, their current approval ratings are expected to be maintained. However, the proportion of undecided voters in Chungnam and Gyeongnam remains high at 40.5% and 54.3%, respectively, indicating that the candidate who can win the support of undecided voters in the final campaign period will emerge victorious.

[Figure 5] Analysis of Undecided Voters Between 1st and 2nd Surveys (Unit: %)

[Figure 6] Preferred Candidates by Voting Intention Among Undecided Voters in the 1st Survey (Unit: %)