A. Data

Thus far, there has been nearly no in-depth statistics suitable for an analysis on South Korean perceptions towards inter-Korean economic cooperation. In this research, we employ a dataset constructed by KDI based on a survey conducted at the end of 2019 (Dec. 13 to Dec. 15). The total number of the respondents is 1,000, 49.6% of which are males and 50.4% females. The regional, age, and level of education compositions of the respondents are similar to those in national statistics and thus suitably represent the overall population.

The survey questionnaire is consisted of ten sections, with each of which asking for respondents’ opinions on various social and economic issues, namely, ‘Measures on the Aging Population’, ‘Jobs’, ‘Education’, ‘Cultural Life’, ‘Public Expenditures’, ‘Government Regulations’, ‘Local Government Policies’, ‘Foreign Trade’, ‘Public Policy Directions and Evaluations’, and most importantly ‘Inter-Korean Economic Cooperation’. As the focus of this research is perceptions on economic cooperation, we mainly utilize the questions under the section ‘Inter-Korean Economic Cooperation’. However, we also employ several questions from other sections as our control variables for the empirical analysis, which will be covered in detail in the next section. Although this survey includes comprehensive questions pertaining to inter-Korean cooperation, the dataset bears limitations similar to those in ordinary cross-section data in that it only reflects public opinions at the specific time of the survey. In particular, at the end of 2019 when the survey was conducted, one can argue that public sentiment with regard to this issue was likely to be more skeptical towards the economic cooperation due to the depressing result of the Hanoi summit of February of 2019 and that this time-specific factor may have resulted in some degree of negative-sentiment-leaning bias in the data. However, it is reasonable to believe that the overall positive sentiment on North Korea-related issues held by the public in 2018 was exceptional, as the number and the intensity of the interactions between the two Koreas and between the US and North Korea as observed by the public reached levels for which comparable precedents are difficult to find since the division of the peninsula. Given that the negative shift in public sentiment about economic cooperation in 2019 after the Hanoi summit arose during such exceptional times, we consider that the aforementioned concerns over negative-sentiment-bias in the data are not overly severe and therefore reasonably appropriate for the empirical analysis here. A survey conducted in February of 2019 during the Hanoi Summit by Realmeter, a Korean public opinions research firm, shows that 68.9% of the respondents agreed that the Kaesong Industrial Complex (KIC) project or Mt. Kumgang tourism should resume; this outcome was 19.5% higher than the finding of an identical survey conducted in 2017 (49.4%). However, public opinion reverted to its ordinary level after the Hanoi summit in 2020, where 43.6% of the respondents signaled their support for the reimplementation of the previous cooperation projects, thus confirming our earlier conjecture.

TABLE 1

DATA COMPARISON

Source: Statistics Korea (http://kostat.go.kr/portal/eng/index.action, Search Date: 2021. 7. 16).

B. Descriptive Statistics

In this subsection, we investigate overall perceptions held by South Koreans towards inter-Korean economic cooperation. We introduce the general statistics of the questions in the ‘Inter-Korean Economic Cooperation’ section of the KDI survey dataset. The main questions are, ‘Do you approve or disapprove of Inter-Korean Economic Cooperation?”, “How urgent do you feel about Inter-Korean Economic Cooperation?”, “Which of the following types of the economic cooperation do you think should be prioritized?”, “Which of the following types of funding do you prefer for economic cooperation?”, “Which of the following states do you think the final goal of the economic cooperation should aspire to?” The questionnaire also includes subsample questions that ask about specific reasons for approval or disapproval and about the urgency of economic cooperation. Summary statistics for all of the questions in the section can be found in the Table 2.

The first factor to highlight would be the approval rating. The proportion of the respondents who selected ‘very much disapprove’ or ‘disapprove’ stands at 33.2%, while 41.2% selected ‘approve’ or ‘very much approve’, suitably representing the divided view. Approximately a quarter (25.6%) of the respondents reported that they are neutral. The ‘approve’ group appears to consider economic cooperation as an opportunity for the South Korean economy to thrive. It was also found that 67% of respondents selected either ‘it could be a breakthrough for the Korean economy’ or ‘to secure the growth of the Korean economy’ as their reason for approving of cooperation. These proportions initially may appear to be inconsistent with the opinions of experts who estimate the expected economic benefit of cooperation to be limited. However, it must be noted that the above question is only presented to the subsample of the group who approve of economic cooperation. Therefore, it does not represent the opinions of the general public. It is reasonable to believe that the approval group evaluates the economic benefit of the cooperation as substantial.

Several different characteristics related to gender are also evident in Table 2. First, the findings show that female respondents are more cautious in general. More female respondents expressed relatively moderate options, as exemplified by their choices of ‘disapprove’, ‘neutral’, and ‘approve’ for the approval question (20.6%, 26.2% and 28.0%, respectively) compared to the selections by the male respondents. Moreover, male respondents displayed stronger hostility towards the North. The proportion of male respondents who reported ‘North Korea is an enemy state to the South’ (21.7%) in response to the question asking about their disapproval of economic cooperation is significantly higher than that of the female respondents (6.9%).

TABLE 2

SUMMARY STATISTICS

The main concerns held by the ‘disapprove’ group toward pursuing economic cooperation are either political or economic. Nearly half of the ‘disapprove’ group (45.8%) expressed concerns about the profits from the economic cooperation leaking into North Korea’s nuclear and missile development programs, while 36.4% of the group doubted the benefit of economic cooperation relative to its cost. Meanwhile, a significant proportion of the ‘disapprove’ group (13.9%) shares negative sentiment towards the North, choosing ‘North Korea is an enemy state of the South’. With regard to the urgency of economic cooperation, 46% of the respondents in total consider it to be urgent, whereas 54% selected either ‘not at all urgent’ or ‘not urgent’.

The results of the survey incorporate some important facts about the overall sentiment of South Koreans towards inter-Korean economic cooperation. As shown in Table 3, respondents who approve of economic cooperation tend to think that it is urgent while respondents who disapprove of economic cooperation tend to think that it is not urgent. This pattern is also statistically verified in that the correlation coefficient of the two variables is 0.82 and is statistically significant at the 1% level. The close tie between the two variables implies that public opinion about economic cooperation is strongly polarized, motivating further analysis.

TABLE 3

CROSS TABLE OF THE APPROVAL AND URGENCY VARIABLES

Note: The correlation coefficient of the two variables is 0.8204 and statistically significant at the 1% level.

The polarizing view of inter-Korean economic cooperation is also evident in the questions that follow. First, more than 40% of the respondents (44.3%) support a high cost of economic cooperation, as represented by ‘large-scale infrastructure development’, whereas nearly half (51.9%) of the respondents prefer forms of economic cooperation that can be had a lower cost, such as trade (23.8%), tourism (10.8%) and firm-sector cooperation (17.3%).

Similarly, nearly half of the respondents prefer domestic sources of funding, such as government spending (23.2%) and investments by domestic private firms (25.7%). The other half prefers foreign source funding such as foreign private investments (26.0%) and hosting funding from international organizations (22.4%) such as the U.N.

The distribution of the respondents with regard to the answers regarding their preferred ultimate goal of economic cooperation is relatively even. The highest degree of integration represented by the option ‘political and economic complete unification’ was indicated by 23.2% of the respondents, whereas the second highest degree of integration, represented by the option ‘economic integration maintaining respective political systems’ earned the most votes by the respondents (33.7%). The two lower degrees of integration, represented by the option ‘high degree of economic interactions’ and ‘re-implementing previous economic cooperation projects under the current condition’ were selected by 22.0% and 11.4% of the respondents, respectively.

C. Inter-Variable Correlations

There could be multiple dimensions by which these perceptions on inter-Korean economic cooperation can be interpreted. In order to investigate the characteristics of the positive and negative perception groups further, we attempt to explain the approval variable and the urgency variable by age groups, along with variables which incorporate details about the economic cooperation, such as the preferred types and preferred sources of funding.

First, due to the rapid economic growth and social changes that the Korean society has experienced since the Korean War, there exists noticeable generational segregation in overall values and perceptions on many social, political and economic issues in South Korea. These divisions are also observed in the perception toward economic cooperation with North Korea. Figure 1 presents the relationship between the age groups and the approval variable. Respondents who approve of economic cooperation are least frequent in the group in their 20s, and from then on, the ‘approve’ group increases with age, reaching a peak at around 50s, after which it declines for those who are 60+. Conversely, the proportion of the ‘disapprove’ group decreases up to those in their 40s and then climbs back up until the oldest group. The proportion of the ‘neutral’ group continuously decreases until the oldest group. These patterns suggest that the youngest and the oldest age groups share similar negative perceptions toward economic cooperation, whereas those in their 40s and 50s have relatively positive perceptions.

FIGURE 1.

APPROVAL/DISAPPROVAL PROPORTIONS AND PREFERRED TYPE BY AGE GROUP

In terms of preferred cooperation projects, large-scale projects of ‘infrastructure development’ are most preferred by those in their 40s, in line with the observed positive perception tendencies expressed by those in their 40s and 50s. In contrast, relatively small-scale projects such as ‘trade’ and ‘tourism’ are most preferred by those in their 60s and in their 20s, respectively; recall that these are the groups expressing negative sentiment towards economic cooperation. This pattern of preferred type by age alludes to the fact that groups with positive perceptions are more likely to support large-scale projects, whereas those with negative perceptions are more likely to support small-scale projects.

Nonetheless, older and younger generations do have similar concerns when it comes to the reasons for approving or disapproving of economic cooperation. As the Figure 2 shows, all age groups selected ‘to secure the future growth of the South Korean economy’ most frequently as the reason for approving of economic cooperation. However, the proportion of respondents who selected such options decreases with age, whereas the proportion of the respondents who are mainly concerned about the North (option ‘to incentivize the North to open and reform’) increases with age. As a result, the proportional gap between the two groups narrows significantly for those who are 60+ relative to those in their 20s.

FIGURE 2.

REASONS FOR APPROVAL/DISAPPROVAL BY AGE GROUP

The ‘reason for disapproval’ pattern also shows age dependency, although it is weaker than in the ‘reason for approval’ case. The proportion of the respondents who selected ‘high cost of the economic cooperation’ as the main reason for their disapproval is higher than the proportion of the respondents who are concerned with the nuclear and missile development programs of the North for those in their 20s as opposed to those who are 60+. The 60+ age group is mostly concerned about the nuclear and missile development programs much more than the cost of economic cooperation. Both patterns of reasons for approving and disapproving of economic cooperation imply that there are growing trends of economic concerns regarding this type of cooperation, more so among the younger generations.

Secondly, we analyze the tendencies of the positive and negative groups according to their preferred types and sources of funding. The options that these survey questions offer can largely be divided into three types in terms of the scale of investment required. More specifically, they can be divided into large-scale investments represented by the option ‘infrastructure development’, mid-scale investments represented by the two options that offer the reimplementation of tourism and firm-sector cooperation, and lastly relatively small-scale investments as represented by ‘promoting trade’.

As shown in the Figure 3, the respondents who prefer relatively large-scale investments for cooperation tend to reveal positive perceptions about economic cooperation, while the respondents who prefer relatively small-scale investments for cooperation tend to have negative perceptions. The statistics indicated that 40.3% of the respondents who prefer trade are most likely to choose ‘Very much disapprove’ or ‘Disapprove’, while over half of the respondents (51.5%) who prefer infrastructure development are most likely to choose ‘Very much approve’ or ‘Approve’. Neutral respondents appear to prefer reimplementation of previous economic cooperation projects, such as tourism and firm-sector projects (i.e., Mt. Kumgang tourism and the Kaesong Industrial Complex). One pattern in Figure 3 that stands out is that tourism is the most favored type of economic cooperation among the neutral group, as economic cooperation is likely to involve large-scale projects which affect not only the prospects of the South-North relationship in the future but also the prospects of the national economy in the long run. From this perspective, the approval and disapproval groups are likely to evaluate economic cooperation based on future political and economic benefits. The neutral group however, is presumably evaluating economic cooperation based on individual preferences, as they are likely to have either no interest or have not given serious thought to the consequences of economic cooperation on a national level. As a result, we suspect that there are two possible reasons for the neutral group’s choice of tourism as their preferred type. First, it may provide them with a chance to visit North Korea, which is directly related to a possible consumption opportunity for this group. A second possible reason is because tourism is simply the most probable and easily implementable type of economic cooperation. Because the options offered for the question, which asks about their preferred type, do not include ‘none of the above’, it is likely that the neutral group selected the most probable and most easily implementable previously implemented project, i.e., tourism.

FIGURE 3.

APPROVAL/DISAPPROVAL VS. PREFERRED TYPE OF ECONOMIC COOPERATION

The approval variable can also be explained with the preferred source of funding variable. As the Figure 4 displays, the respondents who prefer domestic sources of funding (i.e., government finance or private-sector investment) tend to approve of economic cooperation (‘Very much approve’ or ‘Approve’). Specifically, 57.3% and 58.4% of the respondents who selected public (government) investment and private-sector investment, respectively, signaled their approval of economic cooperation. On the other hand, the respondents who prefer foreign sources of funding tended to disapprove of economic cooperation (‘Very much disapprove’ or ‘Disapprove’).

FIGURE 4.

APPROVAL/DISAPPROVAL VS. PREFERRED SOURCE OF FUNDING

The observed differences in perceptions of inter-Korean economic cooperation depending on age, preferred types and preferred sources of funding offer policy implications for pursuing economic cooperation, although a more rigorous analysis is required. This is covered in the next section.

A. Model Construction

In this section, we construct an empirical model for a more in-depth investigation of the characteristics of the groups expressing positive and negative perceptions of inter-Korean economic cooperation. Through the empirical model, we aim to discover which variables affect the positive and negative perceptions under a controlled statistical environment. The specifications of the models are expressed as shown below.

In this equation, y_i represents the two dependent variables of approval and urgency with regard to economic cooperation. The approval variable scales from 1) Strongly Disapprove to 5) Strong approve, and the urgency variable scales from 1) Not at all urgent to 4) Very urgent. In addition, type_i represents the preferred type of economic cooperation, which is composed of the three binary variables of ‘infrastructure development’ and ‘trade’, each taking a value of 1 if the respective type of cooperation project is preferred by the respondents and 0 otherwise, along with ‘reimplementation of existing cooperation projects’ as the reference group. funding_i involves four binary variables, each representing the options offered in the corresponding survey question. The options are ‘domestic public funding’, ‘domestic private funding’, ‘investments by foreign private firms’, and ‘funding from an international organization’. Each of the four binary variables takes a value of 1 if the respondent has chosen a particular option, and 0 otherwise. pol_i represents a set of control variables reflecting the underlying political view of the respondent. X_i represents a set of demographic characteristic variables, in this case age, level of education, level of income, and marital status. μ_i represents a region fixed effect (see Table A1 for details).

Although the approval variable and the urgency variable both attempt to distinguish between positive and negative perceptions in general, the resulting compositions of the distinguished groups show different characteristics. The approval variable captures relatively radical respondents within the resulting positive or negative perception group who have stronger positive or negative opinions compared to those from the urgency variable. There are two main reasons behind this measurement difference. The first is ascribed to the fact that there exists the option ‘neutral’ for the approval question, which those holding mild opinions can choose, leaving only the polarized proportion of the respondents for both the positive and negative perception groups. Secondly, the approval question requires the respondents to make a categorical decision. Evidently, the majority (59.7%) of the disapproval group selected the options ‘nuclear and missile development program concerns’ or ‘North Korea is an enemy state’ to explain their disapproval; these issues are difficult to resolve given that they involve the need to build up trust between the two states. Consequentially, the characteristics of the respondents expressing disapproval to the approval question would differ from those of the ‘not urgent’ respondents when they replied to the urgency question.

We do not utilize the survey questions that ask the respondents to explain their approval and urgency selections for the following two reasons. First, the design of the survey is such that each question requiring such a reason is only applicable to a subsample of respondents. For example, the questions that require a reason for approval are only applicable to respondents who approve of the cooperation, for instance. This substantially reduces the number of samples for the empirical analysis, which in turn undermines the accuracy of the results. Secondly, as mentioned previously, the aim of this research is to explore the characteristics of those who belong to positive and negative perception groups to derive policy implications for future directions regarding economic cooperation, which is difficult to achieve by analyzing the reasons for their respective perceptions.

B. Results

In this subsection, we report the empirical results. For each of the dependent variables, we estimate three specifications (Columns 1 through 6 in Tables 4 and 5) with an identical set of control variables. The first specifications for each dependent variable regress the three cooperation types of variables, the second set applies to the four funding source variables, and third is for all types and sources of funding variables inclusively. As presented in Tables 4 and 5, the results for the two dependent variables are nearly analogous, except for the differences in the significance levels and sizes of a few of the coefficients. Therefore, we explain the results of the regressions of the two dependent variables simultaneously.

First, there are two demographic variables that are statistically significant. For both of these, the coefficients of the age variable are positive and statistically significant, with the coefficient of the age-squared variable being negative and statistically significant. This implies that the age effect on the perception of inter-Korean economic cooperation forms an inversed U-shape, where the positive perception increases with age but at a decreasing rate (its peak is found to arrive at age 50.9 on average for the urgency variable and at age 50.6 on average for the approval variable). This pattern is consistent with the age effect on the perception of unification addressed in the previous literature, in which younger age groups have more negative perceptions compared to the older age groups (Kim, 2019). According to the ‘Unification Perception Survey’ conducted by the Institute for Peace and Unification Studies (IPUS) at Seoul National University, the average proportions of the respondents who think that unification is necessary by age group for those in their 20s, 30s, 40s, 50s, and 60s are 43.5%, 44.7%, 57.2%, 64.8% and 63.1%, respectively, from 2018 to 2020, with the most positive group found to be those in their 50s. This discrepancy in unification perception among the age groups certainly requires further research. Here, we offer some possible reasons for this. First, those currently in their 50s were the main forces behind the democratization movement of the 1980s; this group is also politically known to be pro-unification. Secondly, the group has the most positive experiences and memories of flourishing inter-Korean interactions after the ‘6.15 Communal Declaration’ of 2000, which may have induced the positive unification and economic cooperation perceptions in this group. The younger generations on the other hand do not share the same experiences compared to those in their 50s. In contrast, they have ample memories of political and military friction between the two states likely to have caused relatively negative perceptions about unification and economic cooperation. Moreover, the result of the Column (1) specification showing the approval variable regression (Table 4) suggests that male respondents have more positive perceptions of economic cooperation than female respondents.

TABLE 4

COOPERATION APPROVAL ORDERED LOGIT ESTIMATION RESULTS

Note: Robust standard errors are in parentheses (*** p<0.01, ** p<0.05, * p<0.1).

TABLE 5

COOPERATION URGENCY ORDERED LOGIT ESTIMATION RESULTS

Note: Robust standard errors are in parentheses (*** p<0.01, ** p<0.05, * p<0.1).

Secondly, the preferred type of economic cooperation also has an effect on perceptions. The coefficient of the ‘infrastructure development’ variable is positive and statistically significant on both of the dependent variables, implying that the respondents who prefer infrastructure development over the reference reply ‘re-implementing existing cooperation projects’ are more likely to approve of economic cooperation and consider it to be highly urgent. In contrast, the respondents who selected ‘promoting trade’ as their preferred type of economic cooperation tend to disapprove and feel that economic cooperation is not an urgent issue. The empirical results thus far suggest that the South Korean public tends to perceive infrastructure development projects as a type of economic assistance (or long-term investments) which incurs a short-term financial burden but with the possibility of long-term economic benefits. On the other hand, the public seems to consider trade with the North as a project that involves commercial transactions which may be mutually beneficial. In sum, willingness to tolerate a short-term burden appears to be the decisive factor in forming these perceptions.

With regard to the preferred source funding for economic cooperation, there exist two main narratives according to the results. First, the preference for domestic sources of funding, such as domestic government spending and domestic firm investments, over the foreign private firm investments (the reference group) is likely to result in positive perceptions toward economic cooperation. Moreover, the preference for domestic public funding induces a stronger positive perception than the preference for domestic private funding, as suggested by the coefficients of each explanatory variable. On the other hand, the preference for funding from international organizations such as the UN over foreign private firm investments is likely to result in negative perceptions with regard to economic cooperation. This result is in line with the previous results pertaining to the preferred type variables, as types of economic cooperation that incur high short-term costs are most likely to require domestic and public financing. Those in the negative perception group, on the other hand, are intolerant towards large-scale economic cooperation projects which are, in their opinion, not very beneficial. They therefore oppose the use of domestic sources of funding, especially the government financing or, in other words, the taxpayers’ money. These results provoke discussions about preferred initiating entities of economic cooperation. Those who actively support cooperation are likely to prefer cooperation mainly led by the government or domestic entities, which historically have been conventional types of entities. However, the group of people expressing skepticism prefers cooperation models that instead involve international entities.

In addition, the respondents who consider complete unification as the ultimate goal of economic cooperation tend to be those expressing positive perceptions. This result is intuitive, as economic cooperation is widely conceived as a necessary process for unification.

There are several interesting findings that incorporate the effects of political views on economic cooperation perception. Most of the control variables which reflect the political views of the respondent, such as ‘lay-off of low-performing employee’, ‘government spending cuts’ and ‘easing corporate regulations’, are statistically significant with regard to both dependent variables. The negative signs of the coefficients of these variables suggest that the more progressive the respondent is, the more likely they will express positive perceptions toward economic cooperation. More specifically, the respondents who oppose the ideas that firms should be able to lay-off low-performing employees easily, who oppose small government measures (as opposed to government spending cuts), and who oppose the overall easing of corporate regulations are likely to have positive perceptions toward economic cooperation, while the respondents who agree with the above arguments are likely to have negative perceptions. This result provides evidence of the fact that the inter-Korean economic cooperation perceptions are sensitive to the political views of people despite the fact that economic cooperation, in essence, is an economic issue. It also suggests that building up positive perceptions about this issue will not be easy, as one’s political views are often difficult to change.

Overall, the following characteristics of South Korean public opinions about inter-Korean economic cooperation could be inferred. On average, the group of people who are younger, who prefer lower short-term cost cooperative projects, who prefer foreign sources of funding, and who hold relatively conservative political views tend to have negative perceptions toward economic cooperation, while the group of people who are older, prefer higher short-term cost cooperative projects, prefer domestic sources of funding, and who have relatively progressive political views tend to have positive perceptions toward economic cooperation. As the statistical results and the underlying implications reveal, pursuing economic cooperation with the North is a polarizing issue in South Korea.

Considering the facts that public perceptions toward economic cooperation are inevitably tied to unification perceptions and that the political views of people affect their perceptions on economic cooperation, it is reasonable to conclude that the polarizing perceptions toward this issue are rigid in its nature. Consequentially, some degree of controversy in policy discussions as to which cooperation project and which source of funding should be prioritized could arise, meaning that the issue has to be approached delicately so that cooperation can be ensured without much social friction. According to the results of our analysis, cooperative projects should be pursued on a lower scale (or cost) and under international norms for the facilitation of foreign investments to persuade those with negative perceptions, which in turn will broaden the base of support for this type of cooperation.