SOCIO-ECONOMIC FACTORS AFFECTING TRUST IN THE MILITARY: COMPARATIVES ON PERSPECTIVES ON CHINA, JAPAN, AND SOUTH KOREA

Sung Man Yoon

Seoul National University of Science and Technology, South Korea.

Keywords:Trust in the military, Pro-military tendency, Culture theory, Institutional theory, Socioeconomic factors, Comparative approach.

JEL Classification: B22; C12; H26.

ARTICLE HISTORY: Received: 2 October 2019, Revised: 4 November 2019, Accepted: 6 December 2019, Published: 29 January 2020

Contribution/Originality: The paper provides policy implications that South Koreans in high age groups trust the government more than the military. In contrast, Japanese people in high age groups trust the military more than the government, and as it presents factors for enhancing the pro-military tendency of people concerning China, which is a socialist state system, and Japan that has imperialism and experience in militarism.

1. INTRODUCTION

Interest in social capital has continuously emerged along with human and material capital; those are essential factors in the growth and development of nations, societies, and organizations. Since the late 1980s, prior studies had been actively conducted on social capital. This factor promotes cooperation and trading among members, thereby enabling the efficient pursuit of common goals. In particular, trust among various components of social capital has been regarded as the most important factor (Kim, 2016).

As research interest on trust has increased worldwide, dispensation has long been an important research topic in a variety of disciplines, including administration, economics, political science, sociology and psychology. Thus, research on the concept of trust, its importance, and its influencing factors has been continuously published (Lee and Cho, 2016). Besides, research on government trust has been actively pursued in the public sector; however, only a few researches have been carried out on the concept -- level diagnosis and factors of trust in the military. In addition, trust in the military can be a value or ability that exists in the military organization itself and can be too broad as a specific activity, such as the overall operation of a military organization, a specific defense policy, or a member of a military organization (Park, 2015).

The following are the important aspects of gaining public trust in the government. First, trust in the government is an important factor in reducing transaction costs and promoting cooperation across all sectors of the country and society. Second, a high degree of trust in the government has positive direct or indirect influences on the public’s compliance, favorable support, and attitude toward the introduction of government policies (Kim, 2016; Lee and Cho, 2016). Third, trust at the organizational level is another essential factor that promotes members’ commitment and cooperation. Likewise, gaining people’s trust in the military is distinguished from various points of view. First, trust in the military can facilitate the introduction and enforcement of defense policies and enable the input of relevant personnel and budgets, and directly influences the development of national defense organizations, policies, and institutions (Kim, 2016). If the trust of military organization members is high, then the morale of the military can be increased through positive commitment and cooperation within the organization, which in turn strengthens the spiritual defense force (Lee and Cho, 2016). Even though the importance of trust in the military has been highlighted, research on this field has not been actively conducted. To improve trust in the military, analysis of factors influencing the level of trust is necessary. Considering this, the present study analyzes the socio-economic factors affecting trust in the military. In particular, this study analyzes these factors from the perspectives of South Korea, China, and Japan, which are major countries in Northeast Asia. Socio-economic factors are classified into political interest factors, economic factors, occupational characteristics, and demographic factors. This study also examines whether these factors affect trust in the military. South Korea, China, and Japan are the major countries in Northeast Asia and have a common point in Confucian culture. China follows a socialist system; South Korea is confronted with North Korea, which follows a socialist system; and Japan is a former imperialist and militarist country. In terms of military service, South Korea adopts a conscription system, whereas China and Japan adopt a military recruitment system. A comparative analysis of the levels of trust in the military among countries of different national and military service systems in the past and present will provide implications in several perspectives.  The remaining part of the paper is organized from Sections 2 to 5. Section 2 elaborates a theoretical background and review of literature. Section 3 exhibits an empirical research design, such as research methodology and data. Section 4 analyses the results of descriptive statistics, univariate analysis and multivariate analysis, and finally, Section 5 summarizes and concludes the empirical results and their implications.

2. THEORETICAL BACKGROUND AND LITERATURE REVIEW

2.1. Theoretical Background

Culture theory emphasizes the role of interpersonal trust in institutions. This theory highlights that the trust of the people is the source of trust in the system and the support of the government, whereas institutional theory states that political and institutional achievements determine people’s trust (Almond and Verba, 1963; Putnam, 1993; Fukuyama, 1995; Inglehart, 1997; Edwards and Foley, 1998). Moreover, the culture theory stresses the process of trust formation on the assumption that trust is gradually increased through socialization; whereas the institutional theory utters that public trust can be raised in a short period with government performance. Institutional trust and public trust focus on the hierarchy of trust, which is the superior concept of interpersonal trusts like general and private. Following this perception, Putnam (1993) who conceptualized culture theory, argues a one-sided relationship, wherein interpersonal trust brings institutional trust. The study of Fukuyama (1995) also emphasizes the direct relationship between interpersonal trust and institutional trust. This study includes various organizations, such as churches, sports clubs, bowling organizations, community improvement organizations, and political parties, to promote interpersonal trust. In other words, people can learn how to work with and trust others through participation in various organizations. It also argues that the existence of a strong civil society as a result of these efforts increases trust in social institutions and countries (Khodyakov, 2007).

This argument on cultural theory has led to many subsequent studies. On empirical perspectives, these studies have mainly focused on revealing how personal and social values, such as interpersonal and institutional trust, are related to institutions like political institutions and stable democratic governments. Interpersonal and institutional trust, interpersonal trust and political system, and institutional trust and democratic consolidation are also analyzed in various studies. Kaase (1999) has presented a positive correlation between interpersonal and government trust using European Eurobarometer data for nine European countries. The study of Dekker and Eric (2001) argues that social trust has a positive effect on political behavior in Western Europe. Newton (2007) also explains the positive causal relationship of interpersonal trust in political institutions based on the World Values Survey results, and it reveals that Denmark (trust rate: 67%) and Norway (trust rate: 65%) are more likely to have democratic institutions, governments, law enforcement, civil society, and welfare than Tanzania (trust rate: 8%) and Brazil (trust rate of 3% is relatively fair and justified). The study of Shyu (2010) argues that institutional trust in Taiwan positively affects democracy consolidation. This study explains the causal relationship between institutional trust and democratic consolidation based on the finding that high institutional trust supports democracy and has a high objection to non-democratic alternatives. This cultural theory concludes that interpersonal trust plays a positive role in institutional trust and democratic consolidation. Conversely, the study of Mishler and Rose (2010) has failed to demonstrate the positive impacts of interpersonal trust in democratic consolidation in the perspectives of Russia. Furthermore, Newton (2007) has merely inferred their relationship based on the World Values Survey data. For that reason, there is a need to analyse the relationship between interpersonal trust and democracy empirically.

Putnam (1993) and Gibson (2001) argue that trust does not always create a positive impact on democratic consolidation. Putnam (1993) argues that trust is divided into horizontal (general trust) and vertical trust (private trust), and only horizontal trust has a positive effect. In addition, Gibson (2001) tested the impacts on democratic consolidation in Russia in terms of the types of trust to prove the claim. However, contrary to the argument of Putnam (1993), general trust has no impact, and private trust has a positive impact on democratic consolidation. Gibson (2001) argues that this relationship is a desperate aspect of Russian democracy and stresses the need to change the Russian trust structure. This result reflects that the relationship between trust and institutions can vary depending on the cultural background of each country. Institutional theory puts emphasis on the role of political institutions in promoting trust. Unlike cultural theory, it asserts that interpersonal trust brings institutional trust and government performance. Institutional theory argues that the role of political institutions can increase interpersonal trust. A good performing system increases trust, whereas a poorly performing system causes meetings and distrust (Dasgupta, 1988; Hetherington, 1998). Fair and equitable political processes and institutions are the sources of trust promotion, whereas honest, uncorrupt police force, legal system, and public bureaucracy help to improve public trust (Levi, 1996; Brehm and Rahn, 1997; Rothstein and Stolle, 2002). The study of Newton (2007) argues that the performance of governments creates a positive effect. Citizens are willing to pay taxes, value public interest, play a role as citizens, increase support for government agencies, and ultimately trust other citizens.

Based on these arguments, many studies have attempted to demonstrate the relationship between political-institutional role and trust. These studies primarily focus on how institutions, such as political performance and democratic consolidation, affect social attitudes like interpersonal and institutional trust. From the studied literature, it is understood that there are some studies that have analysed Institutional performance and interpersonal trust, institutional performance and institutional trust, democratic consolidation and interpersonal trust, democratic consolidation and institutional trust, and institutional trust and interpersonal trust. Mishler and Rose (1997) analyzed bilateral causality between trust and political-institutional performance to judge the theoretical suitability of competing cultural and institutional theories. Based on the lifelong learning model, the empirical analysis of 10 communist countries in Europe (Eastern and Central Europe and the former Soviet Union) has confirmed strong institutional theory support. In other words, interpersonal trust has no effect on institutional trust, whereas governmental outcome has a positive effect on interpersonal and institutional trust. Besides, follow-up studies on Russia have not proven the arguments of supporters of culture theory, which states that interpersonal trust enhances institutional trust and democratic regime support. Instead, institutional performance, democratic consolidation, and institutional trust have a positive effect on interpersonal trust, and in particular, the influence of institutional trust is greater than the sum of the effects of all variables on interpersonal trust (Mishler and Rose, 2010). The study conducted by Chang and Chu (2006) also confirms that government performance positively affects interpersonal trust. In the analysis of data from five East Asian countries (Japan, South Korea, Thailand, the Philippines, and Taiwan), political-institutional achievements, such as democracy satisfaction, optimistic economic evaluation, and institutional justice, are found to have a positive effect on interpersonal trust. The study of Rothstein and Stolle (2002) has also found that various types of social systems, including fair and unbiased, have a positive impact on general trust and those with generalized social welfare schemes have higher general confidence than those who do not. Later, a study by Shyu (2010) also confirms that government performance has a positive impact on institutional trust and democratic consolidation in Taiwan. In addition, Newton (2007) and Kumlin (2002) have found that good governance, government effectiveness, political stability, democratic system operation period, legislation, and income balances increase trust, and citizens’ positive evaluation of government performance positively affect general trust.

Furthermore, many studies argue that corruption, which is a negative factor in the political system, negatively affects interpersonal trust. The study conducted by Della (2000) reveals that corruption is the primary driver of distrust in Italy, Germany, and France. Increase in costs due to corruption and deterioration in the quality of public services hamper government responses to citizens, leading to distrust of governmental abilities. Anderson and Tverdova (2003) have analyzed data from 16 developed countries and emerging democracies in Eastern and Western Europe and found that citizens of corrupt countries have low trust in political institutions and low role evaluations. The same has also been found in the study of Seligson (2002), and it reveals that citizens’ experience of corruption reduces confidence in the legitimacy of the regime based on studies conducted in four Latin American countries. Furthermore, Delhey and Newton (2005) argue that political-institutional distrust ultimately has a negative impact on interpersonal trust. In addition, Chang and Chu (2006) have confirmed that corruption perceptions of political institutions in five East Asian countries have negative effects on individual and general trust. Rothstein and Stolle (2002) have also found in the World Values Survey data that as corruption spreads throughout society, countries that recently suffered corruption scandals have low interpersonal trust. Mishler and Rose (2010) have also empirically demonstrated the negative effects of political corruption on interpersonal trust in Russia.

In sum, these results validate the argument of institutional theory, which states that the insufficient role of the government or system leads to distrust in society. This distrust also leads to institutional corruption (Delhey and Newton, 2005) and corruption again leads to low governmental performance (Della, 2000) which ultimately poses a vicious cycle of distrust. However, these studies were all set in Western societies, except for (Chang and Chu, 2006). In addition, the simplification of trust to general trust has limitations.

Therefore, investigating whether these discussions can be applied to the situation in Northeast Asia, including South Korea, and how the effect of such political system differs in terms of the type of trust is necessary. Newton (2007) points out that distrust in the political system can reform to a special trust (private trust) like trust within the family. He also argues that authoritarian political institutions reduce general trust and make citizens more dependent on special trust (private trust) (Mishler and Rose, 1997; Sztompka, 2000).

The culture and institutional theories discussed above suggest that the level of public trust is more important than the cause and effect of government support. In addition, trust in the military (or armed forces), which is part of the government, should be supported only by the trust of the people. This trust can lead to the development of a strong defense force and enhance the morale and loyalty of its members. In particular, the Korean military, which follows a conscription system, should be trusted by the people so it can enhance its defense capability.

2.2. Literature Review

While various prior studies generally agree that government trust is a kind of political attitude that the public has on the government, views on the specific concept of government trust vary (Barber, 1983; Mayer et al., 1995). The concept of government trust is generally accepted as a multi-dimensional concept. Previous studies show that government trust is a multidimensional concept that encompasses trust in particular social and cultural backgrounds, trust as a national governing system, and evaluation of government officials performing government functions (Miller, 1974; Son and Chai, 2005).

Government trust is defined as the expectation that the government will competently and adequately perform stewardship for the people, and trust in the government is heightened only if the government has the moral and technical capacity to perform its trust duties (Chanley et al., 2000). According to Fukuyama (1995), trust is the result of positive interactions among people, societies, and nations (e.g., individual and individual, individual and society, and individual and nation). As a social capital distinguished from physical and human capital, trust forms a cyclical structure and serves as a resource for social and national development. As accumulated trust decreases due to the negative change of trust factors and the occurrence of incidents, various studies have been conducted to analyze factors affecting trust. In this context, government trust will enhance the ability and effectiveness of governance by securing people’s compliance with government policies and enhancing cooperative values. Accordingly, public trust in the government gives legitimacy to the status of government and empowers policy decision-making and enforcement (Hetherington, 2005).

On the other hand, the destruction of government trust can jeopardize the existence of the government itself. Moreover, the decline in government trust can bring down the national consensus and support for what the government wants to do, which may lead to increased conflict and policy costs in a country.

Studies on the factors affecting government trust are focused on the characteristics of trustees, such as demographic factors, social capital, and participation. Various studies have attempted to identify the factors influencing government trust by analyzing the other factors of previous studies. Park et al. (2003) have analyzed the effects of factors like private trust, group participation, political participation and interest, and regional and socio-economic background on government trust to identify the characteristics of government trust and its influencing factors. According to the results of the study, females have more trust on the government than males, and the older they are, the higher their trust on the government. However, academic qualification, marital status, and income do not have a consistent impact on all government agencies and show a negative tendency to trust on some government agencies. As such, although the demographic characteristics of respondent influence government trust, some claim that consistent characteristics in the relationship are difficult to find (Jun et al., 2013). The perception of fairness to the government through government performance influences government trust, and it is influenced by civic culture and government institutions and policies. Thus, interpersonal trust affects the level of trust in government agencies (Park and Lee, 2012; Lee and Yu, 2015). Moreover, Lee and Cho (2016) argue that the government’s recognition of responsibility and performance, interest and belief in politics, and social values have an impact on trust in the parliament, public service, administration and judiciary.

The study of Kim (2008) suggests that the military should have high social relevance to gain trust from society. In particular, his study argues that minimization of human rights violations and waste of time, systematic establishment of new occupation of military role, and management of democratic troops should be sought. In addition, Lee and Moon (2013) classify trust in the military into general, specific, and member trust. Trust is also measured as general trust based on mission performance, specific trust based on tax wastage, and professional expertise. In addition, member trust is measured by the reliability of each layer. Their study reveals that the older the age and the more the male residents have higher level of trust in the military. Notably, respondents with military experience and those from bureaucratic or hierarchical organizations have high levels of trust in the military. Private and non-commissioned officers are listed in order of the level of trust. The highest level of private trust can be attributed to sympathy.

3. RESEARCH METHODOLOGIES AND DATA

3.1. Research Methodologies

In studies on people’s perceptions of a concept, discrete models have been used to pose factors affecting the perception level (Jun and Yoon, 2018). As the perception data are categorical, some studies have relied on a probit regression model, whereas others have used multinomial or nested logit models (Rifaat and Chin, 2007). In terms of respondent levels, levels of trust in the military are classified into four categories (4 – “it is a great deal of confidence,” 3 – “quite a lot of confidence,” 2 – “not very much confidence,” or 1 – “none at all”). These categories are typically considered to ordinal outcomes. Ordered discrete choice models are generally used to analyze such ordinal response data. Among these models, the ordered probit (OP) is the most commonly used approach (Kim and Yoon, 2017; Jun and Yoon, 2018). In Equation 1, yni is the level of trust in the military n by respondent i. The OP model assumes that trust level can be represented by a latent and continuous variable (yni*), which is related to Xni as given.

Where Xni is a vector of explanatory variables, β is a vector of unknown parameters to be estimated, and εni is the random error term capturing the effects of unobserved factors, which is assumed to follow a normal distribution with zero mean and unit variance.
In the respondents’ survey data presented in Equation 2, the levels of trust in the military are scaled in four levels: 1 – “it is none at all,” 2 – “it is not very much confidence,” 3 – “it is quite a lot of confidence,” and 4 – “it is a great deal of confidence.” The independent variables (democracy level, use of armed forces, level of trust in the central government, household income level, household asset, cognitive job, fulltime job, public job, school level, gender, age, marital status, and social class) are then ordered with several or binary categories.

For the level of trust in the military n to occur from respondent i, the observed level of trust (yni) is related to an unobserved (latent) variable (yni*), and it is expressed as follows:

Where j is the levels of trust in the military (in this case, j=4) and μ1, μ2, μ3, and μ4 are unknown threshold parameters to be estimated. The predicted probabilities of the level of trust j (j=1, 2, 3, 4) for the given Xni can be estimated, as shown in Equation 3:

P(Yni = j) = F(μj - X'ni β)-F(μj-1 - X'ni β)                                                                            (3)

Where f(·) is the standard normal cumulative distribution function. The model parameters (e.g., β and yni*) are estimated through the maximum likelihood method. In addition, the marginal effects of the OP model with respect to explanatory variable l (βl) can be estimated, as shown in Equation 4:

Where f(·) is the density function.

The empirical analysis model of this study is designed based on the theoretical model, as shown in Equation 5. The dependent variable, the level of trust in the military, is an ordered multi-nominal variable that indicates how much respondents trust the military. To analyze the effects of socio-economic factors on the trust level, socio-economic factors are defined as political interest factors (democracy perception, trust in government and use of armed forces), economic factors (income and asset), occupational characteristics (cognitive job, fulltime job and public job) and demographic factors (school, gender, age, marital status and social class). Besides, the dummy variables of the countries are included in this equation to compare the level of confidence in the selected countries -- South Korea, China, and Japan.

Where
Trustmilitary: The levels of trust in the military denotes the four-point scale for the question “Could you tell me how much confidence you have in the armed forces?”: 4 – “it is a great deal of confidence,” 3 – “quite a lot of confidence,” 2 – “not very much confidence,” or 1 – “none at all.”
Democracy: Democracy perception denotes the 10-point scale for the question “When people freely choose their leaders in an election, how essential do you think it is as a characteristic of democracy?”: 1 means “not at all an essential characteristic of democracy” and 10 means “an essential characteristic of democracy.”
Trustgovt: The levels of trust in the central government denotes the four-point scale for the question “Could you tell me how much confidence you have in the central government?”: 4 – “it is a great deal of confidence,” 3 – “quite a lot of confidence,” 2 – “not very much confidence,” or 1 – “none at all.”
Armedforce: Justification perception of the rule of armed force denotes the 10-point scale for the question “Could you tell me for ‘the army takes over when the government is incompetent’ how essential do you think it is a characteristic of democracy?”: 1 means “not at all an essential characteristic of democracy” and 10 means “an essential characteristic of democracy.”
Income: 1 indicates the lowest income group and 10 the highest income group.
Asset: “During the past year, did your family save money, just got by, spent some savings, or spent savings and borrowed money?” (1 indicates that the respondent saves money; otherwise, 0).
Cognitive: Nature of tasks (manual vs. cognitive); 1 indicates the most manual task and 10 the most cognitive task.
Fulltime: 1 indicates that the respondent holds a fulltime job; otherwise, 0.
Public: 1 indicates that the respondent works for the government or public institution; otherwise, 0.
Educ: The eight-point scale denotes the highest educational qualification for the question “What is the highest level of school you have attended?”: 1 – “no formal school,” 2 – “elementary,” 3 – “middle school,” 4 – “high school,” 5 – “college,” 6 – “university,” 7 – “graduate school to master degree,” and 8 – “graduate school doctoral degree.”
Gender: 1 indicates that the respondent is female; 0 otherwise.
Age: the respondent’s age
Married: 1 indicates that the respondent is married; 0 otherwise.
Social: 1 indicates the lowest social class and 10 the highest social class.
 Countries: dummy variable of countries.

3.2. Data

The data used in this study are obtained from the World Values Survey (http://www.worldvaluessurvey.org). This survey conducts worldwide once in every five years. This study uses longitudinal data from 1981 to 2014 as an international level data because it analyzes trust in the military and analyzes the time series trends in South Korea, China and Japan. This integrated data includes different versions of longitudinal aggregates covering from 1981 to 2014 (Inglehart et al., 2018).

4. EMPIRICAL RESULT

4.1. Descriptive Statistics

Table 1 shows descriptive statistics of major variables. Trust in the military (Trustmilitary) is 2.1063 out of 4 points and is slightly higher than the average, but it has a standard deviation of 0.8 or higher. This finding means that the level of trust in the military among respondents varies. While considering the political interest variables, the perception of democracy (Democracy) shows an average of 8.4465 out of 10, suggesting that most of the respondents are highly aware of the level of democracy. Moreover, trust in the government (Trustgovt) is 2.4047 out of 4, which is higher than 2.1063 of trust in the military. Therefore, the results reveal that respondents generally trust the government more than the military. Armedforce, which indicates whether the army can intervene when the government is unstable, has an average of 3.5635 out of 10, which indicates that most respondents are negative toward the army.  As economic variables, income level (Income) scored 4.5672 out of 10, which is slightly below the average. The average value of assets (Asset) showing the level of saving financial assets is 0.3387, with one-third of the respondents having financial assets. As an occupational characteristic variable, Cognitive represents 2.5373 out of 10, which means that most respondents are employed in physical labor. In addition, 42.99% of all the respondents hold full-time jobs (Fulltime), and 62.73% of respondents work in the government and public sector (Public). As a demographic variable, Educ is 5.3683 on average, with respondents graduating from college or university, and Gender is 56.89%, indicating that more female respondents than male respondents participated in the survey. Average age of respondents (Age) is 43 years old. Nearly half or 44.58% of the respondents are married (Married), and Social average is 3.3368 out of 5, indicating that their social class is higher than the average.

Table-1. Descriptive statistics.

Variables	Mean	STD	MIN	Q1st	Median	Q3rd	Max
Tustmilitary	2.1063	0.8033	1	2	2	3	4
Democracy	8.4465	1.7409	1	8	9	10	10
Trustgovt	2.4047	0.8605	1	2	2	3	4
Armedforce	3.5635	2.9611	1	1	2	5	10
Income	4.5672	2.4097	1	3	4	6	10
Asset	0.3387	0.4733	0	0	0	1	1
Cognitive	2.5373	2.6632	1	1	1	3	10
Fulltime	0.4299	0.3248	0	0	0	1	1
Public	0.6273	0.4835	0	0	1	1	1
Educ	5.3683	2.0837	1	4	6	7	8
Gender	0.5689	0.4999	0	0	1	1	1
Age	42.9906	15.0712	17	31	42	54	91
Married	0.4458	0.2091	0	0	0	1	1
Social	3.3368	0.9001	1	3	3	4	5

Source: World Values Survey (http://www.worldvaluessurvey.org).

4.2. Univariate Analysis: t-test

Panel 1 of Table 2 shows the t-test results of trust in the military. The results reveal the following scores: 2.3752 in Japan, 2.1555 in South Korea, and 1.6957 in China, at statistically significant levels. Panel 2 of Table 3 shows the analysis results of the mean difference between trust on government (TG) and trust in the military (TM). The TG mean of South Korea is 2.6381, whereas its TM mean is 2.1555, which means that Koreans trust the general government more than the military. The result from Japan is similar to that of Korea. However, in the case of China, the mean TG is 1.6731 and the TM is 1.6957, which is statistically and significantly higher than that of the general government. The results of China, which are different from those of South Korea or Japan, implies that military status is higher than that of general government, and military role in maintaining the state system is more important than that in South Korea or Japan.

Table-2 . T–test of trust on military.

<Panel 1> Trust on Military

Countries	South Korea		China		Japan
	Mean	STD	Mean	STD	Mean	t–stat
TM	2.1555	0.8423	1.6957	0.6441	2.3752	0.7476
South Korea		–0.4598***	0.7590	0.2197***	0.7945	3.56
China	0.4598***	0.7590		0.6795***	0.7046	–56.23
Japan	–0.2197***	0.7945	–0.6795***	0.7046

Note 1: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively, for two–tailed test.
2: The above mean difference is the trust on Military (column) subtracted from the trust on Military (row) average.

<Panel 2> Trust on Military and Government.

Countries	TM: Trust on military		TG: Trust on government		Difference (TM–TG)
	Mean	STD	Mean	STD	Mean	t–stat
South Korea	2.1555	0.8423	2.6381	0.7284	–0.3182***	–24.71
China	1.6957	0.6441	1.6731	0.6271	0.0331***	3.56
Japan	2.3752	0.7476	2.8634	0.7042	–0.6418***	–56.23

 Note 1: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively, for two–tailed test.
2: The above mean difference is the trust on Military (column) subtracted from the trust on Military (row) average.

4.3. Multi-Variate Analysis: Ordered Probit and OLS Regression

Table 3 shows the results of analyzing the effects of socio-economic factors on trust in the military. The pooled sample of the first column is the result of the analysis, including all respondents from South Korea, China, and Japan. Democracy, Trustgovt, and Armedforce are statistically significant and coefficients with values of 0.070 (p<0.001), 0.600 (p<0.001), and 0.020 (p<0.001), respectively. This result implies that the higher the level of democracy is the lower the level of trust in the central government. By contrast, the more people who support military intervention in an unstable government, the more likely that the people trust the military. In addition, the economic variable Asset has a statistically significant coefficient of 0.210 (p<0.001), whereas Income has a statistically insignificant coefficient. In other words, the higher the level of financial assets is the higher the level of trust in the military. Fulltime and Public are statistically insignificant as occupational characteristics, but Cognitive has a statistically significant coefficient of 0.009 (p<0.1). Thus, the more people who do cognitive labor than physical labor, and it indicates the higher the level of trust in the military. In addition, Educ, Gender, Age, and Social are statistically significant coefficients of −0.059 (p<0.001), −0.126 (p<0.001), 0.009 (p<0.001), and 0.062 (p<0.001), respectively. In addition, demographic variables, such as Educ, Gender, Age, and Social, have statistically significant coefficients of −0.059 (p<0.001), −0.126 (p<0.001), 0.009 (p<0.001), and 0.062 (p<0.001), respectively. Thus, respondents who have low educational qualification, are male and older, and have high social class are more likely to have a higher level of trust in the military.

While Comparing it with South Korea, China, and Japan in the second column, Democracy and Trustgovt among the political interest variables show a statistically significant positive (+) or negative (−) coefficient in all three countries. However, for Armedforce, only South Korea has a statistically significant coefficient of 0.031 (p<0.001), whereas China and Japan show statistically insignificant coefficients. Thus, the greater the recognition of South Koreans that the military can intervene in politics if the government is unstable that indicates the higher their level of trust in the military. Thus, a person who recognizes that an unstable government can be replaced by an army (that is, in favor of military martial law), and has a high level of trust in the military.

As economic variables, Income and Asset show a statistically insignificant coefficient, but only South Korea’s Income show a statistically significant coefficient of −0.055 (p<0.001). Unlike China and Japan, the higher the income level of South Koreans leads to the lower their level of trust in the military. In other words, in South Korea, the higher the income level is less likely interested to have pro-military tendencies.

Cognitive, Fulltime, and Public, as occupational characteristics variables, mostly show statistically insignificant coefficients, but only Japan’s Cognitive variable has a statistically significant coefficient of 0.023 (p<0.1). Thus, Japanese who have careers in cognitive work have higher levels of trust in the military. Unlike Korea and Japan, people with cognitive jobs in Japan are more likely to have pro-military tendencies. Japan’s Public also has a statistically significant coefficient of −0.180 (p<0.1), which means that Japanese people who work in the government or public sector have a lower level of trust in the military. Unlike Korea and China, the pro-military tendency toward public sector workers in Japan is low.

Among the demographic variables, Educ shows a statistically significant coefficient of −0.042 (p<0.001) only in the Japanese sample. Gender in South Korea and China shows statistically significant coefficients of −0.107 (p<0.05) and −0.106 (p<0.1), respectively. Thus, the more highly educated Chinese people have the lower friendship tendency. Males in South Korea and China also have higher trust in the military level. The samples related to Age in South Korea and Japan show statistically significant coefficients of 0.017 (p<0.001) and 0.006 (p<0.001), respectively, and only South Korea’s Social has a statistically significant coefficient of −0.096 (p<0.05). This result implies that South Koreans and Japanese who are older and have a high social class. They have a high level of trust in the military.

Table-3 . Results of ordered probit regression: Effect of Socioeconomic factors on trusts in the military.

Variables	Pooled Sample		South Korea		China		Japan
	Estimate	Wald–Χ2	Estimate	Wald–Χ2	Estimate	Wald–Χ2	Estimate		Wald–Χ2
Intercept1	–0.337***	4.46	–0.627***	4.06	0.510	2.70	–0.675**		5.98
Intercept2	1.747***	4969.11	1.503***	1377.95	2.107***	1499.99	1.921***		2015.58
Intercept3	2.979***	6656.06	2.891***	2803.30	3.052***	1088.70	3.162***		2310.49
Democracy	0.070***	68.26	0.078***	30.04	0.072***	19.66	0.076***		27.16
Trustgovt	–0.600***	1018.63	–0.536***	257.29	–0.924***	442.24	–0.395***		115.52
Armedforce	0.020***	14.23	0.031***	11.79	0.002	0.03	0.018		1.84
Income	–0.011	1.95	–0.055***	9.58	–0.021	1.26	–0.012		1.38
Asset	0.210**	5.66	–0.479	2.01	0.149	1.19	0.063		0.23
Cognitive	0.009*	2.83	0.007	0.57	0.023*	3.90	–0.006		0.52
Fulltime	0.109	1.20	0.963	1.23	–0.067	0.17	0.011		0.01
Public	0.064	1.90	0.111	1.41	0.106	2.07	–0.180*		3.27
Educ	–0.059***	44.58	–0.023	1.60	–0.012	0.66	–0.042***		6.89
Gender	–0.126***	20.03	–0.107**	5.35	–0.106*	3.89	–0.070		1.99
Age	0.009***	84.37	0.017***	76.33	0.003	2.32	0.006***		9.72
Married	0.071	0.11	0.394	0.95	–0.524	0.81	–0.059		0.04
Social	0.062***	9.61	–0.096**	6.31	0.010	0.06	0.033		0.96
∑Countries	Included		Excluded
Log-Likelihood	–6461.87		–2535.37		–1587.67			–2067.50
Observations	23,031		7,070		7,791			8,170

Note 1: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively, for two-tailed test.

Although the central government and military can be regarded as the same government in a broad framework, separately, this study measures the level of trust in the two. In particular, it is also relevant to analyze the trust levels between the central government and the military because of the characteristics of Northeast Asia, where capitalist and socialist state systems are mixed. Therefore, this additional analysis examines the effects of socio-economic factors on the difference between trust in the military and government. As in the first column of Table 4 analyzed with pooled samples, Armedforce shows a statistically significant coefficient of 0.036 (p<0.001), whereas Democracy shows statistically insignificant results. Thus, more people who recognize that the military can intervene when the government is unstable -- the higher the level of trust in the military. From the second to the fourth column analyzed for each country, Democracy of South Korea and China shows statistically significant coefficients of −0.021 (p<0.1) and −0.022 (p<0.05), respectively. Thus, the lower the perception of democracy in South Korea and China, citizens do not trust the military more than the central government.

The samples related to Income of the pooled and South Korea show statistically significant coefficients of 0.020 (p<0.05) and 0.051 (p<0.001), respectively. This finding suggests that South Koreans with high-income levels are more likely to trust in the military than the central government. Asset of the pooled and China samples have statistically significant coefficients of −0.162 (p<0.05) and −0.173 (p<0.05), respectively. Thus, Chinese people with high financial assets may not have more confidence in the military than in the central government.

Among the occupational characteristic variables -- Cognitive and Public show statistically significant coefficients only in the pooled and China sample and pooled and Japan sample, respectively. Chinese people with cognitive jobs are more likely to trust the central government than the military. Japanese people who work in the government or the public sector are more likely to trust the military than the central government.

Table-4. Results of ordinary least regression: effect of socioeconomic factors on the difference between trusts on military and government.

Variables	Pooled Sample		South Korea		China			Japan
	Coeff.	t–stat.	Coeff.	t–stat.	Coeff.	t–stat.		Coeff.		t–stat.
Intercept	–0.105	–0.93	–0.236	–1.04	0.292*		1.75	–0.803***		–4.43
Democracy	0.003	0.48	–0.021*	–1.88	–0.022**		–2.39	–0.004		–0.40
Armedforce	0.036***	9.59	–0.011	–1.64	0.000		–0.01	0.008		0.85
Income	0.020***	3.54	0.051***	3.66	0.009		0.90	–0.005		–0.67
Asset	–0.162**	–2.47	0.166	0.64	–0.173**		–2.27	0.023		0.24
Cognitive	–0.010**	–2.46	–0.007	–0.91	–0.014**		–2.21	0.006		0.89
Fulltime	–0.134*	–1.82	–0.115	–0.18	–0.099		–1.12	–0.025		–0.25
Public	0.100***	2.94	–0.030	–0.42	–0.056		–1.39	0.126*		1.72
Educ	–0.015**	–2.36	0.006	0.39	–0.004		–0.47	0.024*		2.09
Gender	0.091***	4.33	0.163***	4.52	0.029		0.96	0.108***		2.97
Age	–0.007***	–9.65	–0.008***	–4.99	0.000		0.17	0.002*		1.81
Married	–0.125	–0.79	–0.439	–1.40	0.199		0.60	0.115		0.55
Social	–0.028*	–1.89	0.037	1.26	–0.014		–0.63	–0.096***		–3.99
∑Countries	Included		Excluded
Adj.R2	0.046		0.033		0.005				0.014
F–stat.	27.16***		7.67***		1.75***				3.65***
# of Observations	23,031		7,070		7,791				8,170

Note 1: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively, for two-tailed test.

Among the demographic variables, Educ shows statistically significant coefficients of −0.015 (p<0.05) and 0.024 (p<0.1) in the pooled and Japan sample, respectively. This result means that the higher the education level is the greater the trust in the central government than the military. However, highly educated Japanese people are more likely to trust the military than the central government. Gender has a statistically significant positive coefficient except for China, suggesting that females have more trust in the military than in the central government. Age in South Korea and Japan samples shows statistically significant coefficients of −0.008 (p<0.001) and 0.002 (p<0.1), respectively. This result shows that older South Koreans trust the central government more than the military. Moreover, older Japanese people trust more in the military than in the central government. Social of pooled and Japan samples shows statistically significant coefficients of −0.028 (p<0.1) and −0.096 (p<0.001), respectively. Thus, the more Japanese people who perceive their social class to be higher, and the more they trust the central government than the military.

5. CONCLUSIONS

The concept of trust in the military is abstract and subjective. Unlike other government agencies, considerable difficulties are encountered in enhancing trust in the military because the direct and indirect experiences of the people from South Korea’s military service system and culture are linked with other factors.

However, public trust in the military can strengthen military defense power by improving the political-economic effect and morale of the military members. Therefore, this study has analyzed various factors affecting trust in the military based on the samples of South Korea, China, and Japan. In particular, a comparative analysis of trust in the military of the three countries with different national systems and military service systems provides an opportunity to improve the level of trust in the South Korean military.

The conclusions of this study, from the results, are as follows. First, China’s trust in the military is generally lower than that of South Korea and Japan. Japan’s level of trust in the military was high until the mid-1990s, but it occurred oppositely in South Korea. Therefore, the size of trust in the military level since mid-2000 is South Korea, Japan, and China in that order. Second, China has higher trust in the military than the government, but South Korea and Japan has higher level of trust in the government. This finding indicates that the socialist system in China is more pro-military than South Korea and Japan, and the role of the military in China is relatively high.

Third, the level of democracy, which is a political interest variable. In this case, the level of trust in the military is higher when the military intervention is highly perceived by the unstable government. The higher the level of trust in the government leads to the lower the level of trust in the military. People with financial assets as an economic factor and persons with cognitive jobs as a factor of job characteristics have a high level of trust in the military. Furthermore, as demographic factors, people who have higher educational qualification, are men and older, and have high subjective social class have high trust in the military.

Fourth, in 2005, the level of confidence of South Koreans on the military was high for those who were working in the public sector with low levels of income and financial assets. However, in 2010, the impact of these factors was statistically insignificant. This result means that South Koreans’ perception of the military has improved.

Finally, South Korea and China with low democratic perception have more trust in the government than in the military. South Koreans with higher income level have shown more trust in the military than in the government. In addition, the higher the level of financial assets in China is the more stable the government than the military. Japanese people who work in the government or public sector are more likely to trust the military than the government. Furthermore, older Koreans trust the government more than the military, and older Japanese trust the military more than the government.  The results of this study explore socio-economic factors to increase the level of trust in South Korea’s military. This study provides policy implications as it presents factors for enhancing the pro-military tendency of people concerning China, which is a socialist state system, and Japan that has imperialism and experience in militarism.

Funding: This study received no specific financial support.

Competing Interests: The author declares that there are no conflicts of interests regarding the publication of this paper.

REFERENCES

Almond, G.A. and S. Verba, 1963. The civic culture: Political attitudes and democracy in five nations. Princeton: Princeton University Press.

Anderson, C.J. and Y.V. Tverdova, 2003. Corruption, political allegiances, and attitudes toward government in contemporary democracies. American Journal of Political Science, 47(1): 91-109.Available at: https://doi.org/10.1111/1540-5907.00007.

Barber, B., 1983. The logic and limits of trust. NJ: Rutgers University Press.

Brehm, J. and W. Rahn, 1997. Individual-level evidence for the causes and consequences of social capital. American Journal of Political Science, 41(3): 999-1023.Available at: https://doi.org/10.2307/2111684.

Chang, E.C. and Y.-h. Chu, 2006. Corruption and trust: Exceptionalism in Asian democracies? The Journal of Politics, 68(2): 259-271.Available at: https://doi.org/10.1111/j.1468-2508.2006.00404.x.

Chanley, V.A., T.J. Rudolph and W.M. Rahn, 2000. The origins and consequences of public trust in government: A time series analysis. Public Opinion Quarterly, 64(3): 239-256.Available at: https://doi.org/10.1086/317987.

Dasgupta, P., 1988. Trust as a commodity. In Gambetta. D. (Ed.), Trust: Making and Breaking Cooperative Relations. Oxford, UK: Basil Blackwell. pp: 49-72.

Dekker, P. and M. Eric, 2001. Uslaner. Social capital and participation in everyday life. London: Routledge.

Delhey, J. and K. Newton, 2005. Predicting cross-national levels of social trust: Global pattern or nordic exceptionalism? European Sociological Review, 21(4): 311-327.Available at: https://doi.org/10.1093/esr/jci022.

Della, P. D., 2000. Social capital, beliefs in government, and political corruption. In Pharr, S.J., Putnam, R.D. (Eds), Disaffected Democracies: What's Troubling the Trilateral Countries? Princeton, NJ: Princeton University Press.

Edwards, B. and M.W. Foley, 1998. Civil society and social capital beyond putnam. American Behavioral Scientist, 42(1): 124-139.Available at: https://doi.org/10.1177/0002764298042001010.

Fukuyama, F., 1995. Trust: The social virtues and the creation of prosperity. New York: Free Press.

Gibson, J.L., 2001. Social networks, civil society, and the prospects for consolidating Russia's democratic transition. American Journal of Political Science, 45(1): 51-68.Available at: https://doi.org/10.2307/2669359.

Hetherington, M.J., 1998. The political relevance of political trust. American Political Science Review, 92(4): 791-808.Available at: https://doi.org/10.2307/2586304.

Hetherington, M.J., 2005. Why trust matters: Declining political trust and the demise of American liberalism. Princeton, NJ: Princeton University Press.

Inglehart, R., 1997. Modernization and post modernization: Cultural, economic, and political change in 43 societies. Princeton: Princeton University Press.

Inglehart, R., C. Haerpfer, A. Moreno, C. Welzel, K. Kizilova, J. Diez-Medrano and B. Puranen, 2018. World values survey: All rounds—country-pooled datafile version. Available via world value survey. Available from http://www.worldvaluessurvey.org/WVSDocumentationWVL.jsp .

Jun, B.W. and S.M. Yoon, 2018. Taxpayer’s religiosity, religion, and the perceptions of tax equity: Case of South Korea. Religions, 9(11): 333.Available at: https://doi.org/10.3390/rel9110333.

Jun, D.S., I.W. Kwan and K.H. Jung, 2013. Analysis of trust on government: Trust on president and evaluation of governmental policies. Korean Policy Studies Review, 22(2): 181-206.

Kaase, M., 1999. Interpersonal trust, political trust and non-institutionalised political participation in Western Europe. West European Politics, 22(3): 1-21.Available at: https://doi.org/10.1080/01402389908425313.

Khodyakov, D., 2007. Trust as a process: A three-dimensional approach. Sociology, 41(1): 115-132.Available at: https://doi.org/10.1177/0038038507072285.

Kim, B.-J., 2008. Some proposals to the Korean civil-military relations for the advanced country. The Journal of Strategic Studies, 15(3): 24-57.

Kim, K. and S. Yoon, 2017. Taxpayer’s perception to tax payment in kind system in support of SMEs’ sustainability: Case of the South Korean government’s valuation of unlisted stocks. Sustainability, 9(9): 1-12.Available at: https://doi.org/10.3390/su9091523.

Kim, K.H., 2016. Review of measurement for trust on military. JoGan Kukbang Bread, 1611(16-12): 1-8.

Kumlin, S., 2002. Institutions-experiences-preferences: How welfare state design affects political trust and ideology. In Rothstein, B., and Steinmo, S. (Eds.), Restructuring the Welfare State: Political Institutions and Policy Change. New York: Palgrave/Macmillan.

Lee, C.-W. and M. Cho, 2016. Exploring the effect of government performance, political ideology, and social values on citizen’s trust on government. Korean Review of Organizational Studies, 12(4): 153-181.

Lee, S.J. and Y.S. Moon, 2013. Study on the trust on military of people. 2013 Winter Symposium of Korean Policy Association.

Lee, S.J. and H.J. Yu, 2015. A study on the factors affecting public trust in government: Focusing on mediating effects of trust in president. Journal of Korean Politics, 24(2): 53-81.

Levi, M., 1996. Social and unsocial capital: A review essay of Robert Putnam's making democracy work. Politics & Society, 24(1): 45-55.Available at: https://doi.org/10.1177/0032329296024001005.

Mayer, R.C., J.H. Davis and F.D. Schoorman, 1995. An integrative model of organizational trust. Academy of Management Review, 20(3): 709-734.Available at: https://doi.org/10.2307/258792.

Miller, A.H., 1974. Political issues and trust in government: 1964-1970. The American Political Science Review, 68(3): 951-972.Available at: https://doi.org/10.2307/1959140.

Mishler, W. and R. Rose, 1997. Trust, distrust and skepticism: Popular evaluations of civil and political institutions in post-communist societies. The Journal of Politics, 59(2): 418-451.Available at: https://doi.org/10.1017/s0022381600053512.

Mishler, W. and R. Rose, 2010. What are the political consequences of trust? A test of cultural and institutional theories in Russia. Comparative Political Studies, 38: 1050-1074.Available at: https://doi.org/10.1177/0010414005278419.

Newton, K., 2007. Social and political trust, in R. Dalton and H. Klingemann (Eds.), Oxford Handbook of Political Behavior. New York: Oxford University Press.

Park, H.-R., 2015. Tasks for South Korean military to regain people’s trust. Strategic Studies, 22(2): 139-165.

Park, H.B., H.C. Lee and Y.S. Cho, 2003. An analysis on the properties and determinants of trust in government in Korea. International Review of Public Administration, 37(3): 45-66.

Park, H.B. and Y.R. Lee, 2012. Effects of civil culture and governmental policies on trust on government. Korean Public Administration Quarterly, 24(4): 913-936.

Putnam, R., 1993. Making democracy work: Civic traditions in modern Italy. Princeton: Princeton University Press.

Rifaat, S.M. and H.C. Chin, 2007. Accident severity analysis using ordered probit model. Journal of Advanced Transportation, 41(1): 91-114.Available at: https://doi.org/10.1002/atr.5670410107.

Rothstein, B. and D. Stolle, 2002. How political institutions create and destroy social capital: An institutional theory of generalized trust. Paper Prepared for 98th Meeting of American Political Science Association in Boston, MA.

Seligson, M.A., 2002. The impact of corruption on regime legitimacy: A comparative study of four Latin American countries. The Journal of Politics, 64(2): 408-433.Available at: https://doi.org/10.1111/1468-2508.00132.

Shyu, H.-Y., 2010. Trust in institutions and democratic consolidation in Taiwan” Taiwan's politics in the 21st century: 69-100. Available at: https://doi.org/10.1142/9789814289092_0004.

Son, H.J. and W.H. Chai, 2005. The analyzing of factors influencing trust in government. Korean Public Administration Review, 39(3): 87-114.

Sztompka, P., 2000. Trust: A sociological theory. Cambridge: Cambridge University Press.

Views and opinions expressed in this article are the views and opinions of the author(s), Asian Economic and Financial Review shall not be responsible or answerable for any loss, damage or liability etc. caused in relation to/arising out of the use of the content.