[EAI Opinion Review] Analysis of Seoul Citizens' Perception Landscape on Free School Meals and Prospects for the Referendum

Can Universalism and Selectivism in Free School Meals Be Reconciled?

1. Introduction: Is Welfare Selectivism vs. Universal Welfare Irreconcilable?

Following the Seoul Metropolitan Council's passage of the 'Eco-Friendly Free School Meal Support Ordinance' on December 1st of last year, which mandates free meal costs for all elementary school students starting in 2011, Mayor Oh Se-hoon opposed it as welfare populism. The free school meal debate is expected to be decided by the results of a referendum, which is scheduled to be held between August 23rd and 25th. On July 20th, the Seoul Metropolitan Government's Referendum Review Committee reviewed the objection filed by the 'National Movement to Eradicate Welfare Populism' (hereinafter referred to as the 'National Movement') regarding the signatures for the anti-free school meal referendum. The committee confirmed over 520,000 valid signatures, exceeding the required 418,000, and plans to officially announce the referendum's proposal around the 25th.

Furthermore, on the same day, the Referendum Review Committee finalized the referendum question, which had been under debate, to present two options: 'A phased free school meal implementation by 2014 targeting students from the bottom 50% income bracket (hereinafter referred to as the 'National Movement's Proposal')' and 'Full implementation of free school meals for all students regardless of income, for elementary schools (2011) and middle schools (2012) (hereinafter referred to as the 'Current Seoul Metropolitan Office of Education Proposal').'

The confrontation between the Seoul Metropolitan Government and the Seoul Metropolitan Office of Education/Metropolitan Council over the free school meal plan stems from the fact that this issue is not merely a policy debate but a direct clash of welfare philosophies and methodologies underpinning the current ruling and opposition parties. While the free school meal plan implemented by the Seoul Metropolitan Council, which holds a majority of seats, and the current progressive-leaning metropolitan offices of education aligns with the principle of universal welfare, providing benefits as a realization of the universal rights (human rights) of all students, Mayor Oh Se-hoon and the Grand National Party's proposed free school meal plan aligns with the principle of selective welfare, which provides support by selecting beneficiaries.

Given the lack of substantial discussion and practice regarding specific welfare policies, let alone a distinct welfare philosophy or methodology, in South Korean society despite the emphasis on the importance of welfare, the current referendum controversy cannot be dismissed as meaningless. However, the debate between the Seoul Metropolitan Government and the Seoul Metropolitan Council, and between the ruling and opposition parties, appears to be devolving into an unproductive political dispute, failing to develop beyond a dichotomous framework that emerged after the free school meal issue gained prominence in the 2010 local elections: 'Universal Welfare = Progressive = Democratic Party's Proposal' versus 'Selective Welfare = Conservative = Grand National Party's Proposal.'

The fundamental reason why the two positions are irreconcilable and running on parallel tracks is the rigid approach taken by both political parties. The Grand National Party dismisses universal welfare itself as welfare populism and approaches all issues from a selective welfare perspective. Conversely, the Democratic Party ignores the necessity of selective welfare in certain situations and views all issues from a universal welfare standpoint. A more significant problem is that the voters, who should be at the center of these debates, are relegated to passive observers of the political discourse. Historically, political parties have mobilized their logic and macro-statistical indicators while lacking systematic research into the interests and preferences of voters.

This study will analyze the public opinion of 500 Seoul citizens regarding the free school meal referendum, where welfare philosophies clash between the ruling and opposition parties. It will examine how Seoul citizens view free school meals and how their attitudes toward the referendum differ based on these perspectives. Based on this, the study will forecast the future direction of the referendum and identify issues that the ruling and opposition parties are overlooking in their current debates.

In conclusion, contrary to the political sphere's perception of the free school meal issue as a dichotomy between universal and selective welfare, a majority of voters hold ambivalent attitudes, sharing concerns from both perspectives. In the eyes of most voters, universal and selective welfare are not mutually exclusive choices. Rather, they empathize with the core concerns of each proposal while simultaneously harboring concerns about the potential problems each might create. This reflects a multidimensional and complex attitude, not a one-sided one (Martinez et al. 2005; Kim Jang-soo 2005; Yoo Sung-jin 2008; Jeong Han-wool 2011).

If a referendum is held, ambivalent voters will also choose one side, and one proposal will be selected. However, it would be paramount for the political parties to consider how to preserve the values and strengths of the rejected proposal and, simultaneously, how to address the concerns voters have about the chosen proposal. The data is based on a special opinion poll conducted on July 23rd by the East Asia Institute, YTN, JoongAng Ilbo, and Korea Research, surveying 500 Seoul citizens.

2. Public Opinion on the Free School Meal Referendum Controversy

Referendum Interest: 77.1% Interested, 22.3% Not Interested, 0.5% Don't Know/No Answer

The referendum is not only a point of contention in the sharp welfare debate between the ruling and opposition parties but also the first referendum for the Seoul Metropolitan Government. Consequently, Seoul citizens' interest in and intention to participate in the referendum far exceed the majority. According to a public opinion survey of 500 Seoul citizens, 77.1% responded that they were interested in the referendum (29.3% very interested, 47.8% somewhat interested), while only 22.3% stated they were not interested (15.9% not very interested, 6.4% not at all interested) [Figure 1].

[Figure 1] Interest in Free School Lunch Referendum: 500 Seoul Citizens

As shown in [Table 1], a high level of interest exceeding the majority is observed across all demographics. However, differences in referendum interest are found based on age group and political party affiliation. Notably, no significant difference was observed in ideological orientation. By age group, the 20s showed the lowest interest at 65.8%, while the 30s (83.0%) and 40s (81.2%), who are the direct beneficiaries of free school lunches, showed the highest interest. Among those in their 50s, 78.2% expressed interest, and among those aged 60 and above, 77.0% were interested. Regarding party affiliation, 82.5% of Grand National Party supporters expressed interest, compared to 74.0% of Democratic Party supporters, 74.3% of minor party supporters, and 73.7% of unaffiliated individuals, showing relatively lower interest. Meanwhile, there were no significant differences across ideological groups, indicating that this is a major issue between progressive and conservative camps. 74.9% of progressives, 77.7% of centrists, and 78.0% of conservatives expressed high interest in the referendum.

[Table 1] Differences in Referendum Attitudes by Demographic Group (%)

60.9% Favor Referendum, 63.3% Will Participate

Although the opposition parties are questioning the legitimacy of holding the referendum based on procedural flaws (invalid signatures) and election costs of approximately 18 billion KRW, over 60% of Seoul citizens (60.9%) are in favor of holding the referendum [Figure 2]. Opposition was 27.7%, and 11.4% were unsure.

By age group, support for the referendum was 67.2% among those in their 20s, 59.0% among those in their 30s, 55.1% among those in their 40s, 60.9% among those in their 50s, and 62.5% among those aged 60 and above. Notably, support was higher among those in their 20s. Regarding party affiliation, 71.1% of Grand National Party supporters favored the referendum, while support was weaker among Democratic Party supporters (56.4%), minor party supporters (55.0%), and unaffiliated individuals (53.2%). Differences in attitudes toward holding the referendum were also observed based on ideological orientation. 67.3% of conservatives, 57.9% of centrists, and 55.0% of progressives were in favor [Table 1].

[Figure 2] Support for Referendum (%) : 500 Seoul Citizens

When asked about their intention to participate in the referendum, 63.3% expressed their willingness to participate, which is higher than the 25.3% who stated they would not participate [Figure 3]. 11.3% were undecided. While it is difficult to predict voter turnout based on a survey conducted a month before the referendum, scheduled around August 24th, and intention surveys often elicit socially desirable responses, the figures suggest a non-negligible possibility of the referendum proceeding.

In terms of intention to participate in the referendum, participation was lowest among those in their 20s and highest among those aged 60 and above, mirroring general voter turnout trends. Among those in their 20s, participation intention was 54.8%, 59.7% for those in their 30s, and 53.8% for those in their 40s. However, 70.0% of those in their 50s and 70.1% of those aged 60 and above expressed their intention to participate. Regarding party affiliation, 73.8% of Grand National Party supporters indicated they would participate, compared to 52.3% of Democratic Party supporters, 58.5% of minor party supporters, and 60.3% of unaffiliated individuals. Among conservatives, 69.2% stated they would participate, while 61.2% of centrists and 56.7% of progressives indicated their intention to participate, showing relatively lower participation compared to conservatives.

[Figure 3] Interest in Free School Lunch Referendum and Intention to Participate: 500 Seoul Citizens

3. Referendum Framing Debate and Seoul Citizens' Preference for Free School Lunches

Referendum Framing Debate

Article 15 of Chapter 2 of the current Referendum Act, titled "Form of Referendum," stipulates that "A referendum shall be conducted in a format that expresses agreement or disagreement on a specific matter or selects one of two options." This provision aims to prevent the referendum's content from being interpreted in multiple ways by requiring a yes/no or binary-choice question format that encompasses the entirety of public opinion. Based on this regulation, the signature petition by the National Movement Headquarters to Expel Welfare Populism is, in form, a binary-choice signature, not a yes/no question, stating: "Oppose the current comprehensive free school lunch plan, and support the plan to gradually implement free school lunches for the bottom 50% of income earners by 2014." Currently, Seoul City favors the binary-choice format, while the City Council prefers the yes/no format (Hankook Ilbo, 2011/07/17).

The issue arises because, as pointed out in EAI Public Opinion Briefing No. 98 (dated May 30, 2011), the framing of the question (format) affects the distribution of responses. In a binary-choice frame between "comprehensive free school lunch plan" and "selective free school lunch theory," support for the selective limited lunch theory was high (54.7%). However, in a yes/no frame regarding the suspension of the current free school lunch plan, a framing effect favoring comprehensive free school lunches was observed due to opposition to the suspension. According to EAI's May and June survey results, when asked in a binary-choice format (selective lunch vs. universal lunch) to Seoul citizens, support for the selective lunch theory exceeded the majority at 55.0% (June survey results). However, when asked in a yes/no format (support or opposition to the Seoul City's plan to suspend free school lunches), opposition to the suspension of free school lunches exceeded the majority at 54.2% (May survey results).

However, the survey questions in these studies were arbitrarily drafted by the research team, not the final versions decided by the deliberation committee, and were not conducted on the same subjects, weakening the validity of the comparison. Therefore, in the current survey of 500 Seoul citizens, both the final version decided by the Referendum Deliberation Committee (binary-choice) and the yes/no format were posed. For the yes/no format, questions were simultaneously asked regarding the "gradual implementation plan for the bottom 50% of income earners by 2014" and the "comprehensive implementation plan for all students regardless of income, starting with elementary schools (2011) and middle schools (2012)."

The author disagrees with the current practice in political circles and the media of referring to the National Movement Headquarters' position as the "gradual implementation plan" and the plan passed by the City Council as the "comprehensive implementation plan." This is because the essential difference between the two positions lies in whether the scope of beneficiaries is selected based on income or provided to all recipients universally, not in the procedural aspect of implementation, i.e., the stage of implementation. To avoid confusion during the actual survey, the terminology designated by the deliberation committee was used, not abbreviations.

However, a variable that could become a factor is the argument that the free school lunch plan, finalized on August 17 last year, violates the Referendum Act's provisions because it differs from the plan confirmed by the deliberation committee. The plan states: 'Seoul City, along with district offices, will gradually expand free school lunches: elementary schools in 2011, elementary and first-year middle school students in 2012, elementary and first- and second-year middle school students in 2013, and elementary and all middle school students by 2014.' If this argument is accepted, the validity of the referendum itself could become uncertain due to the content difference from the plan currently finalized by the Referendum Deliberation Committee. However, this study uses the wording determined by the Referendum Petition Deliberation Committee, as a final conclusion has not yet been reached.

Binary-Choice: Selective Lunch Theory Dominates

When surveyed using the wording decided by the Seoul Metropolitan Government Referendum Petition Deliberation Committee on the 20th, which asks to choose between "gradual implementation of free school lunches by 2014 for students in the bottom 50% of income" and "comprehensive implementation of free school lunches for all students regardless of income, starting with elementary schools (2011) and middle schools (2012)," the results, similar to the East Asia Institute's May and February survey results, show that support for the selective lunch plan exceeds the majority.

Similar to the EAI June survey results, 53.2% of respondents chose the "plan to gradually implement by 2014 for the bottom 50% of income students" (hereinafter referred to as the "bottom 50% implementation plan"), while 38.1% chose the "plan to comprehensively implement free school lunches for all students regardless of income, starting with elementary schools (2011) and middle schools (2012)" (hereinafter referred to as the "implementation plan without income distinction"). 8.7% responded "don't know." When asked in a binary-choice format, support for the free school lunch plan was slightly dominant.

In the June survey, which used very similar wording to the current survey, among 168 respondents in the Seoul area, 37.4% chose the "plan to implement free school lunches for all students regardless of income, starting with elementary schools this year and middle schools next year," while 55.0% chose the "plan to gradually implement free school lunches by 2014 for the bottom 50% of income earners," showing similar results to the current survey. In the February survey of 168 Seoul citizens, 27.1% responded "should be implemented for all students regardless of income," 58.9% responded "should be implemented only for some students based on income," 13.5% responded "should not implement free school lunches," and 0.5% responded "don't know/no answer." Despite differences in survey wording, the consistent trend shows that when asked in a binary-choice format, support for the selective lunch plan exceeds the majority.

Yes/No Format: Responses Divided by Yes/No Subject

However, when asked in a yes/no format, choosing between the two options, different survey results are observed. When asked about support or opposition based on the opposition party's stance, "the plan to comprehensively implement free school lunches for all students regardless of income, starting with elementary schools (2011) and middle schools (2012)," support was 44.3% versus opposition at 46.5%, indicating a close contest. Although this differs from the EAI's May survey that asked about "support or opposition to Seoul City's stance on suspending comprehensive free school lunches," the finding that the plan to provide free school lunches to all students regardless of income is relatively higher than in the binary-choice format is consistent.

In the current survey, conversely, when asked about support or opposition based on the Seoul City's plan, proposed by the National Movement Headquarters to Oppose Welfare Populism, "to gradually implement by 2014 for students in the bottom 50% of income," support was 62.8% and opposition was 27.1%. This result is favorable to Mayor Oh Se-hoon and the National Movement Headquarters to Oppose Welfare Populism, compared to both the yes/no question format for the opposition party's plan and the binary-choice question format.

In summary, for Mayor Oh Se-hoon and the National Movement Headquarters, it is advantageous to ask about support or opposition to the plan to gradually implement by 2014 for the bottom 50% of income earners, whether in a binary-choice or yes/no format. When asked about support or opposition to the plan for comprehensive implementation for all students regardless of income in a yes/no format, a close contest between the ruling and opposition parties is anticipated. The reason why support for the universal free school lunch plan tends to be higher in the yes/no frame is likely because, in the absence of an alternative, selective free school lunches are not considered as a viable alternative, and it may be interpreted as a complete suspension of free school lunches itself.

[Figure 4] Survey Results on Seoul Metropolitan Government Referendum Petition Deliberation Committee's Free School Lunch Plan (%)

[Figure 5] Support or Opposition for Universal and Selective Free School Lunch Plans (%)

Reason for Dominance of Selective Lunch Theory: Fear of Tax Increases Outweighs Stigma Concerns

Overall, when surveyed using binary-choice or yes/no formats based on the current Seoul City plan, public support appears to favor Mayor Oh Se-hoon's position, which is the plan to selectively support the bottom 50% of income earners. Why is this the case?

First and foremost, the relative preference for selective welfare theory over universal welfare theory, which underlies both lunch plans, becomes a factor favoring selective welfare theory when a choice must be made between the two ([Figure 6]). 65.6% agreed with the statement, "The government should apply welfare services such as medical care, childcare, and education equally to all citizens," while 32.5% disagreed. Don't know/no answer accounted for 1.9%. However, a significant 85.6% agreed with the statement, "The government should prioritize urgent areas and target groups based on national finances when providing welfare services," while only 12.3% disagreed and 2.1% were unsure.

[Figure 6] Agreement with Universal Welfare Theory and Selective Welfare Theory (%)

Furthermore, narrowing down to the stance on free school lunches, the concern about tax increases, as argued by Seoul City, is greater than the "stigma effect" raised by the opposition party. The stigma effect refers to the potential disadvantage students receiving benefits might face in school life and peer relationships if classified as low-income. Conversely, Seoul City argues that providing benefits to all students would lead to a significant tax burden.

As shown in [Figure 7], 76.0% agreed with the statement, "Implementing free school lunches for all students will incur substantial costs, potentially leading to tax increases," while only 22.8% disagreed. In contrast, 58.0% agreed with the statement, "Students receiving free school lunches targeting the bottom 50% of income may be classified as poor and face disadvantages," while 39.0% disagreed. Although concerns about the so-called stigma effect also exceeded the majority, they were less pronounced than concerns about tax increases. This difference appears to be a factor contributing to the majority support for the Seoul City plan.

[Figure 7] Concerns about Tax Increases for Universal Free School Lunches and Stigma Effect for Selective Free School Lunches (%)

4. Seoul Citizens' Perception Landscape of Free School Lunches

Beyond the analysis showing a higher proportion of agreement with selective welfare theory than universal welfare theory, and greater concern about tax increases than the stigma effect, a more notable point is that agreement with both universal welfare theory (65.6%) and selective welfare theory (85.6%) significantly exceeds the majority. This suggests the potential existence of ambivalent attitudes, where many individuals agree with both seemingly contradictory stances.

Analysis of Seoul Citizens' Perception Landscape of Free School Lunches: A New Analytical Framework

To understand the welfare perception landscape of Seoul citizens, moving beyond the dichotomy of selective and universal welfare proponents and considering ambivalent welfare perceptions, this study proposes to classify welfare perception types into four categories by cross-tabulating agreement with universal welfare and agreement with selective welfare (2x2).

(1) Universal Welfare Proponents: Agree with universal welfare theory but not with selective welfare theory.

(2) Selective Welfare Proponents: Agree with selective welfare theory but not with universal welfare theory.

(3) Ambivalent Welfare Proponents: Agree with both universal and selective welfare theories.

(4) Anti-Welfare Proponents: Disagree with both universal and selective welfare theories.

Analyzing the survey data for agreement with universal and selective welfare theories among the 500 Seoul citizens, excluding the 17 respondents who answered "don't know" for either question, reveals that ambivalent welfare proponents constitute the majority, as expected. The analysis results can be confirmed in [Figure 8]. (1) Universal welfare proponents accounted for 10.8%, (2) Selective welfare proponents for 31.9%, (3) Ambivalent welfare proponents, who agree with both, exceeded the majority at 55.5%, and (4) Anti-welfare proponents, who disagree with both, accounted for only 1.9%.

In contrast to the ruling and opposition parties, which are locked in a debate framed by the dichotomy of selective versus universal welfare, over half of the voters fall into the ambivalent attitude category, where the concerns of both sides coexist. In a situation where a majority of the public holds both perspectives, it is inevitable that many citizens will experience confusion when asked to choose between one stance or the other in a binary choice. Consequently, when asked for their support or opposition to one of the two stances, the responses are likely to vary depending on the question format. This is the fundamental reason why responses to the referendum question framing regarding free school lunches differ.

Notably, ambivalent voters, who internally hold both universal and selective welfare values, are inherently more likely to have fluid attitudes toward free school lunches compared to universal or selective welfare proponents. Their positions are more susceptible to change during subsequent election campaigns through exposure to campaigning and information from respective factions.

[Figure 8] Seoul Citizens' Welfare Perception Types (%)

Who are the proponents of conflicting welfare theories?

Through [Table 2], we will examine the socioeconomic composition of each perception type to see which class exhibits which perception type.

First, by age group, the highest proportion of individuals with conflicting attitudes were found in the 20s, followed by the 40s, 50s, and those aged 60 and above. Among those in their 20s, 65.7% were proponents of conflicting welfare theories, 25.3% were proponents of selective welfare theories, and only 8.1% were proponents of universal welfare theories. In their 30s, the proportion of conflicting welfare theorists was lower at 48.2%, while selective welfare theorists accounted for 32.8% and universal welfare theorists 17.3%. For those in their 40s, conflicting welfare theorists comprised 60.8%, selective welfare theorists 25.5%, and universal welfare theorists 12.7%. Among those in their 50s, 56.3% were proponents of conflicting welfare theories, 33.3% were selective welfare theorists, and only 6.9% were universal welfare theorists. For those aged 60 and above, conflicting welfare theorists were the lowest at 45.9%, while selective welfare theorists were the highest among all age groups at 45.9%. Universal welfare theorists accounted for only 7.1%. The proportion of proponents of welfare theories that are not clearly defined was only between 1% and 3.4%.

By educational attainment, those with a college degree showed the highest proportion of conflicting welfare theorists at 58.5%, followed by high school graduates at 51.4%, and those with middle school education or less at 50.0%. Conversely, proponents of selective welfare were most prevalent among those with middle school education or less at 44.7%, followed by high school graduates at 31.2% and college graduates at 30.6%. On the other hand, proponents of universal welfare were the lowest among those with middle school education or less at 2.6%, highest among high school graduates at 15.9%, and around 9.0% among college graduates.

By ideological orientation, while universal welfare theories are often associated with progressivism, the majority of progressives in Seoul actually held conflicting attitudes (58.5%), with selective welfare theorists at 25.5% and universal welfare theorists at 14.9%. Among moderates, conflicting attitudes were highest at 57.2%, followed by selective welfare theorists at 28.3% and universal welfare theorists at 11.8%. Among conservatives, conflicting welfare theorists were also the most numerous at 54.4%, and selective welfare theorists were relatively high compared to other ideological groups at 37.8%. Universal welfare theorists accounted for only 6.7%.

By party support, among supporters of the Democratic Party, which advocates for universal welfare, conflicting welfare theorists were the most numerous at 64.0%. Among supporters of other parties, this figure was 59.3%, for Grand National Party supporters it was 51.9%, and for independents it was 51.2%. Conversely, selective welfare theorists were highest among Grand National Party supporters at 40.3%, and also high among independents at 38.0%. Among Democratic Party supporters, this figure was 21.9%, and among supporters of other parties, it was 11.9%. Proponents of universal welfare were highest among supporters of other parties, which include many progressive parties, at 27.1%. Among Democratic Party supporters, this was 13.2%, among independents 8.5%, and among Grand National Party supporters, it was only 5.5%.

[Table 2] Analysis of Welfare Perception Landscape by Socioeconomic Class (%)

5. Attitudes of Conflicting Welfare Theorists Toward the Referendum

Conflicting Welfare Theorists: High Interest and Intention to Vote

Examining attitudes toward the current referendum by welfare perception type, as expected, selective welfare theorists showed the highest interest in the referendum (81.0%), the highest approval rate (64.9%), and the highest intention to vote (65.9%). Conversely, universal welfare theorists were the most passive. Their interest was 68.6%, their approval rate for the referendum was 46.2%, falling short of a majority, and their intention to vote was the lowest at 52.9%. Conflicting welfare theorists, while not as high as selective welfare theorists, showed considerable engagement and favorability toward the referendum compared to universal welfare theorists. From the perspective of advocating for selective welfare, the high proportion of conflicting welfare theorists among Seoul residents could be a favorable factor ([Figure 9]).

[Figure 9] Referendum Attitudes by Welfare Perception Type (%)

Conflicting Welfare Theorists: Concerns over the stigmatizing effects of selective meal provision are as significant as concerns over tax increases for universal meal provision.

However, examining their attitudes toward the specific referendum proposals reveals a potentially fluid factor. For conflicting welfare theorists, concerns about tax increases (81.7%) are greater than concerns about stigmatization (40.6%), unlike selective welfare theorists. Nor is the concern about stigmatization (76.9%) significantly higher than the concern about tax increases (52.9%), as seen with universal welfare theorists. Instead, concerns about tax increases (76.9%) and concerns about stigmatization (64.2%) coexist to a considerable extent. Currently, concerns about tax increases appear to be more dominant, but this suggests that their position could be more susceptible to change than any other type, depending on which aspect is emphasized during the election campaign ([Figure 10]).

[Figure 10] Concerns Regarding Each Proposal by Welfare Perception Type (%)

Conflicting Welfare Theorists' Preference for Welfare Proposals: 50.4% for Welfare for the Bottom 50% Income Bracket vs. 40.0% for Welfare Without Income Distinction

Examining voting behavior by welfare perception type reveals distinct preference patterns for each position. 72.4% of selective welfare theorists preferred the proposal for the bottom 50% income bracket to be implemented in 2014, while only 21.1% supported the universal welfare proposal that provides support without income distinction. Conversely, 83.3% of universal welfare theorists expressed support for the universal welfare proposal, with only 13.3% supporting the selective welfare proposal for the bottom 50% income bracket. However, among conflicting welfare theorists, preferences for both proposals were nearly equal. 50.4% supported the proposal for the bottom 50% income bracket, and 40.0% supported the universal welfare proposal without income distinction. 9.6% were undecided or did not respond. Considering the response distribution of conflicting welfare theorists, who constitute over half of all Seoul residents, and factoring in the margin of error and the month-long election campaign period, it is difficult to easily predict the election outcome.

[Figure 10] Preference for Referendum Proposals by Welfare Perception Type (%)

6. Remaining Variables

In summary, public opinion in Seoul is favorable towards Mayor Oh Se-hoon's attempt to discontinue universal free school meals and transition to selective free school meals. If this trend holds, will it remain solid? As examined earlier, the voting intentions of conflicting welfare theorists, who constitute 55.5% of Seoul residents, are expected to be the biggest variable if a referendum is held. In this regard, we will examine several remaining variables.

Variable 1. Potential for Change When Budgetary Information for Each Proposal is Provided; Support for Selective Welfare Proposal Decreases from 53.2% to 49.7%

Recent media reports indicate that implementing the current universal free meal plan would cost 400 billion won, while the selective free meal plan would cost 300 billion won. Depending on the perspective, the difference may not be significant. As discussed earlier, concerns about universal welfare theories are largely rooted in budgetary issues. Therefore, if the budgetary gap between the two plans is not substantial, public opinion could potentially shift.

In this survey, after the Referendum Review Committee of Seoul decided on the referendum question and provided information on the budgetary costs of each proposal, support for the plan targeting the bottom 50% of income earners decreased from 53.2% to 49.7%, while support for the plan benefiting all students without income distinction changed from 38.1% to 37.4%, thus narrowing the gap between the two positions. However, the proportion of those who responded 'don't know' increased to 12.8% ([Figure 11]).

A similar phenomenon was observed among conflicting welfare theorists. Before the information was provided, their preference for the bottom 50% income bracket plan was 50.4%, and after receiving the budgetary information, it decreased to 46.6%. The gap between the two positions narrowed from 10.4 percentage points before information provision to 6.3 percentage points ([Figure 12]). Considering the overwhelming proportion of this group, if debates regarding budget and the stigmatization effect of selective welfare intensify during the subsequent election campaign, the possibility of shifts in their positions cannot be ruled out.

[Figure 11] Change in Binary Choice When Information on Costs of 400 Billion Won for Universal Welfare and 300 Billion Won for Selective Welfare is Provided (%)

[Figure 12] Change in Binary Choice When Information on Costs of 400 Billion Won for Universal Welfare and 300 Billion Won for Selective Welfare is Provided (%)

Variable 2. The Extent of the Grand National Party's Support

Unlike Mayor Oh Se-hoon's consistent stance, the Grand National Party has so far shown lukewarm support for Seoul's free school meal policy. Although there are recent movements within the party to rally support, calling the free meal debate a 'holy war,' the mixed reactions, as evidenced by critical remarks from Governor Kim Moon-soo and Representative Hong Sa-duk, are likely to be a significant variable in the upcoming referendum process. Furthermore, the reactions of other presidential candidates are also a variable. For instance, the leading Grand National Party presidential candidate, former representative Park Geun-hye, is actively responding to welfare agendas by proposing a Korean-style welfare model. Representative Yoo Seong-min, who represented the pro-Park faction and came in second in the July 4th party leadership election, has also promised a bold welfare agenda. Therefore, it is difficult for the Grand National Party to provide full support, given that public sentiment against the current government for the next general and presidential elections exceeds a majority, necessitating a reform that aligns more closely with public demands and interests (EAI Policy Brief No. 100).

Currently, regarding the approach to welfare agendas, which are crucial for mitigating polarization, the selective welfare theory favored by the Grand National Party is indeed more prevalent than the universal welfare theory. However, the existence of voters with conflicting attitudes, where both positions coexist, and the need to meet the 33.3% voter turnout threshold mean that Mayor Oh Se-hoon and those advocating for selective welfare theory require the full support of the Grand National Party. Nevertheless, at present, the Grand National Party's priority is to convey a message of commitment to mitigating polarization. Engaging in a political conflict over welfare debates with the opposition party is not an easy task. There is a factual difference in political positions between the Grand National Party and its key presidential candidates, who must demonstrate sincerity in addressing welfare issues and polarization, and Mayor Oh Se-hoon, who is directly confronting the opposition party based on public support for his welfare philosophy and methodology. The actions of the Grand National Party's current leadership and prominent presidential candidates during the referendum campaign will be a significant variable.

Variable 3. Will the Opposition's Boycott Movement Succeed? "Do Not Agree" 59.7%

Another variable is the provision in the Referendum Act that prohibits the opening of ballot boxes if the voter turnout does not exceed one-third of the total registered voters. In other words, if the voter turnout for this referendum does not exceed 33.3%, the vote count will be invalidated. Consequently, the opposition party is raising objections based on the referendum costs exceeding 18 billion won and procedural flaws in the signature collection process, advocating for non-participation in the referendum. While public opinion indicating an intention to participate in the vote stands at 63.3%, survey responses regarding voting tend to reflect desirable responses, thus appearing higher than actual voter turnout. Therefore, it is difficult to gauge the success of the referendum based solely on this response.

To assess the potential impact of an opposition boycott, we examined opinions on non-participation in the referendum. The results show that 33.4% agree with the boycott, while 59.7% disagree, with the latter exceeding a majority. This indicates that a boycott movement launched without sufficient public consensus could face backlash.

[Figure 13] Attitude Toward the Argument for Boycotting the Referendum Due to 18 Billion Won in Costs and Signature Procedure Flaws (%)

Variable 4. Outcome of Legal Disputes

Currently, the Democratic Party in the Seoul Metropolitan Council and civic groups such as the People's Solidarity for Participatory Self-Government have filed an application for an injunction against the execution of the referendum petition approval with the Seoul Administrative Court on the 19th, citing reasons such as the signature list submitted by the National Movement to Expel Welfare Populism not conforming to the formats stipulated by the Referendum Act and ordinances.

Furthermore, the Seoul Metropolitan Office of Education has issued a statement arguing that the free school meal plan confirmed by the Office of Education on August 17th of last year differs from the text confirmed by the deliberation committee. Specifically, the plan states that 'Seoul City, together with district offices, will gradually expand free school meals: elementary schools in 2011, elementary and middle schools in 2012, elementary and middle schools grades 1 and 2 in 2013, and elementary and middle schools grades 1-3 in 2014.' This, they claim, violates the provisions of the Referendum Act. Additionally, the Office of Education is reportedly considering filing a constitutional complaint with the Constitutional Court, arguing that 'Seoul City does not have the authority to decide on free school meal matters,' and seeking a provisional injunction to halt the referendum.

In particular, if the Seoul Metropolitan Office of Education's claim that the referendum proposal differs from the plan currently envisioned by the Seoul Metropolitan Office of Education is found to be true, or if the judiciary rules that the free school meal issue does not fall under the scope of a referendum according to the Referendum Act, it could become a highly significant variable that raises questions about the validity of the actual vote.

Conclusion

In conclusion, while the political sphere is framing the referendum as a dichotomy of universal welfare = progressive and selective welfare = conservative, leading to a life-or-death political conflict, the majority of the public exhibits a conflicting attitude, holding both needs and concerns regarding both positions. Therefore, even if the selective welfare theory is passed, significant support for the universal welfare theory must be considered, and preparations for the potential stigmatization effect of selective welfare are essential. Conversely, even if the decision is for universal welfare theory, consideration must be given to alleviating concerns about the financial burden.

Currently, public opinion generally favors Mayor Oh Se-hoon due to relatively higher concerns about tax increases associated with universal welfare. However, there is still a considerable amount of time until the final vote. Moreover, the voting outcome remains fluid, depending on who can more persuasively address and propose solutions for the dual concerns and worries held by the conflicting welfare voters, who constitute over half of Seoul residents.

In this context, the author believes that regardless of victory or defeat, the first step in understanding this referendum will be for both the ruling and opposition parties, and the Seoul City government and council, to break free from the political dichotomy they have imposed upon themselves and to truly understand the conflicting and complex perception framework of the voters. To achieve this, it is hoped that the ultimate destination for the free school meal debate will be the development of a complex, secondary equation that leverages the strengths and compensates for the weaknesses of both positions, moving beyond a simple, binary, either/or equation. It is time to consider whether the welfare agenda, as demonstrated in the free school meal referendum controversy thus far, can become an agenda for mutual growth that the ruling and opposition parties can resolve collaboratively, rather than a political issue for them to fight to the bitter end. We expect both parties to demonstrate mature political leadership by developing complex solutions where both can coexist, moving beyond a zero-sum outcome.■