Introduction
In modern capitalist societies, any government that is to survive and develop must be able to guarantee stable capital accumulation and to maintain a minimum degree of political legitimacy (O'Connor, Reference O'Connor1973; Badie and Birnbaum, Reference Badie, Birnbaum and Goldhammer1983: 34). Legitimacy here refers to the people's belief that their country, political system, officials, or policies are morally correct or at least morally acceptable (Rodee et al., Reference Rodee, Anderson, Christol and Greene1983: 21; Wong and Wan, Reference Wong and Wan2008a: 124). The importance of legitimacy lies in the fact that it is the foundation of capital accumulation: a government that is not morally acceptable to its people will not only find it difficult to maintain stable and effective capital accumulation, but will inevitably face numerous socio-political challenges and crises. Thus, legitimacy can be regarded as the cornerstone of regime stability.
Legitimacy is generally a reflection of the level of trust that the public has in its government. It represents the amount of space that a government can stretch its administration. Thus, there is a positive relationship between legitimacy and political trust (Gamson, Reference Gamson1968: 44; Wang, Reference Wang2005). Political trust has, in consequence, become an important issue in political and academic circles. Academic research in Europe and the US has largely focused on the level of trust that people have in their government and in the democratic system. Some scholars have found that, in recent years, advanced democracies such as the US are suffering a long-term crisis of a low level of political trust, and worry that the democratic system could eventually collapse (Miller, Reference Miller1974; Hetherington, Reference Hetherington1998: 803–4; Newton, Reference Newton2006). There are also scholars who view a low level of trust as a normal democratic phenomenon. For example, Norris (Reference Norris and Norris1999) has proposed the notion of ‘critical citizenship’, arguing that modernization weakens people's willingness to submit to the ruler and political authority, and in turn gives rise to a sizable group of critical citizens who challenge their government through various courses of action in the hope of finding a better way to consolidate democracy. Asian scholars, on the other hand, place more emphasis on the difference in political trust between democratic and non-democratic regimes. They have discovered that political trust is in fact high in authoritarian countries such as China, and that such factors as improvements in standards of living and socio-economic conditions, as well as manipulations of the authoritarian political culture, all help to explain this phenomenon (Chen et al., Reference Chen, Zhong, Hillard and Scheb1997; Shi, Reference Shi2001; Wang, Reference Wang2005).
Using the data from a cross-national social survey, this article attempts to examine and compare the level of political trust in Hong Kong and Taiwan, to analyse the major determinants, and to discuss their related implications. Both Hong Kong and Taiwan are Chinese societies that began to take off economically in the 1960s and have become relatively wealthy and socially stable since the 1990s. However, the political system of Hong Kong today remains non-democratic or partially democratic (Kuan and Lau, Reference Kuan and Lau1995; Wong and Wan, Reference Wong and Wan2007, Reference Wong and Wan2008a), while Taiwan has practised full democracy for many years. Comparing political trust in the two societies and the determining factors behind such trust should allow us to see more clearly, under similar socio-economic conditions, the possible differences in political trust and related complex constraints between different political regimes.
Conceptualization of political trust
The concept of political trust is filled with ambiguity and complexity, and to date the academic world has not formed any important consensus over its meaning (Newton, Reference Newton2002). Stokes (Reference Stokes, Cleveland and Lasswell1962) defined political trust as some kind of evaluative orientation of the people towards their government. Miller (Reference Miller1974) put forward a similar view, arguing that political trust is the people's normative or value expectation towards the ruling administration. However, Gamson (Reference Gamson1968: 54) believed that political systems will produce structural preferences for their people, who rarely pay attention to or question such systemic-structural effects on their political attitudes. Easton (Reference Easton1975), on the other hand, regarded political trust as a situation in which the authorities have not been questioned and the people feel that their interests will be looked after. Hetherington (Reference Hetherington2005: 9–10) also saw political trust as the people's assessment of whether or not the results of government policies are consistent with their expectations. Similarly, Rudolph and Evans (Reference Rudolph and Evans2005: 661) synthesized political trust into the overall feeling that people have towards their government. It is a reflection of the people's degree of satisfaction with the government's policies. The higher the level of trust, the more effective the government's polices are, and the more consistent the results of the government's policies are with the expectations of the people. Shi (Reference Shi2001: 401) further pointed out that political trust is a kind of faith, and that it is an essential factor in political support and of great importance because it can give a regime room to maneuver when difficulties arise.
Hence, Easton (Reference Easton1975) directly used the concept of political support to replace political trust. According to him, political support can be classified into ‘diffuse support’ and ‘specific support’. Specific support focuses on the concrete level of satisfaction that the people have towards the results of the government's policies and the performance of the related political authorities, and such an evaluation is relatively short term. Diffuse support, on the other hand, focuses on the faith that people have in the larger political system (e.g., democratic institutions), and is a result of long-term political socialization. Christensen and Lægreid (Reference Christensen and Lægreid2005: 489–90) compared the two concepts of diffuse support and specific support, and proposed four types of political trust: (1) The level of political trust (both in the political system and government performance) is high, indicating high levels of diffuse and specific support, with both mutually strengthening each other. (2) The level of trust in the political system is higher than the level of trust in the government's performance, indicating a high level of diffuse support and a low level of specific support. This means that the people approve of the existing political system, but are not satisfied with the competence of the government administration. (3) The level of trust in the government's performance is higher than the level of trust in the political system, indicating a high level of specific support and a low level of diffuse support. This means that the people approve of the competence of the government administration, but question the current political system. (4) The level of trust in both the political system and the government's performance is rather low, indicating low levels of both diffuse and specific support, with the two mutually reinforcing each other to cause a widespread crisis of legitimacy.
Among these four types of political trust, the first and fourth are self-explanatory, while the second and third types require some explanation. Logically speaking, if the people are consistently dissatisfied with the competence of the government administration, this may eventually lead them to question the political system itself. Conversely, if the people are consistently satisfied with the competence of the government administration, in the end their support and trust in the political system may be strengthened accordingly. In other words, the second and third types of political trust are relatively unstable. The development and changes in the people's attitude towards the competence of the government administration will affect their rational judgement of the political system, thus causing a shift towards the first or fourth types of political trust.
In any case, it can be said that political trust is a kind of spiritual endorsement of the people, developed from their expectations of the present and future actions of the ruling authority. Since the ruling authority consists of different institutions, the concept can be operationalized to embrace various domains of institutional trust. Conceptually, when the people trust the ruling authority as a whole, they will naturally presuppose that the government administration seeks the well-being of the people, and that the people will benefit from such a government. As a result, they should display a high level of support and tolerance towards it. In this regard, political trust can be seen as the foundation of political support and political tolerance. By contrast, if the people do not trust the government, they tend to think that all of the government's policies are formulated in the interests of private parties instead of with the good of the public in mind, and that they themselves and society as a whole will be the losers. Therefore, they tend to withdraw their support or tolerance and oppose the government in everything to protect their own rights and interests (Gamson, Reference Gamson1968: 44). It is worth noting that if the ruling authority is unable to obtain the trust of the people over the long term, sooner or later this is bound to harm the legitimacy of the entire political system (Hetherington, Reference Hetherington1998: 791–2).
Determinants of political trust
There are two broad theoretical approaches that compete to explain political trust: institutional and cultural (Mishler and Rose, Reference Mishler and Rose2001). Proponents of the institutional approach hypothesize that political trust is endogenous, i.e. arising from rational responses by individuals to the performance of political institutions (March, Reference March1988; North, Reference North1990). Institutionalists therefore typically emphasize the importance of policy outcomes. For them, trust in political institutions is a function of the extent to which these institutions produce desired outcomes (Przeworski et al., Reference Przeworski, Alvarez, Antonio Cheibub and Limongi1996). On the other hand, those advocating the cultural approach view political trust as exogenous, originating outside the political sphere, in cultural values and beliefs about people that are learned through early life socialization. For culturalists, political trust is basically an extension of interpersonal trust. It is heavily shaped by various cultural values and beliefs learned in the process of socialization early in life and later projected onto political institutions. In other words, the effects of people's evaluations of institutional performance on political trust are not independent but conditioned by cultural values and beliefs, and vary across cultures and societies (Almond and Verba, Reference Almond and Verba1965; Putnam, Reference Putnam1995; Inglehart, Reference Inglehart1997).
Empirically, both the institutional and cultural approaches have received different degrees of support. With regard to the institutional approach, studies in the US have revealed that the competence of the President, Congress, and government, and the overall performance of the economy can all affect the level of political trust among the public (Hetherington, Reference Hetherington2005). Yet, apart from conventional institutional legitimacy, other studies have indicated that the judiciary, partisanship, rule of law, people's simple instrumentalism,Footnote 1 the level of participation in policy making, and ideological orientations have all also been found to be related to the public's acceptance of the government's policies and to satisfaction with the government's performance (Hibbing and Theiss-Morse, Reference Hibbing and Theiss-Morse2001; Gibson et al., Reference Gibson, Caldeira and Spence2005). In support of the cultural approach, Christensen and Lægreid (Reference Christensen and Lægreid2005) showed that political culture and such socio-demographic variables as age, education, and occupation also have a role to play in shaping political trust. Furthermore, Chanley (Reference Chanley2002) pointed out that when a country encounters an external threat, such as the ‘9/11’ terrorist attack, the feeling that they have a shared enemy will raise the level of political trust among the people. However, in a study of ten post-Communist countries in Eastern and Central Europe and the former Soviet Union, Mishler and Rose (Reference Mishler and Rose2001) found that the results strongly pointed to the superiority of the institutional approach to political trust, while providing little support for the cultural approach.
In recent years, research taking both approaches has also been conducted in Asia. For example, Chen et al. (Reference Chen, Zhong, Hillard and Scheb1997) divided political support into affective support and instrumental support and then used the two concepts to analyse China's experience. They discovered that the Chinese government enjoys a high level of affective support and a somewhat lower level of instrumental support, with the two types of support being mutually related. In line with the institutional argument, they further found that evaluations of policy performance are important determinants of political support in China.Footnote 2 On the other hand, the cultural approach has also received empirical backing. Shi (Reference Shi2001: 419) compared China and Taiwan from the angle of cultural values, examining the influence of Confucian culture on political trust. He found that political trust in Taiwan is affected more by the performance of its democratic government. In contrast, political trust in China depends more on traditional cultural values, particularly on the Chinese preference for hierarchical order and on their collective spirit, which is characterized by a willingness to sacrifice personal interests for social harmony. Similarly, in a comparative study of eight Asian countries or regions (China, Taiwan, Hong Kong, Japan, South Korea, the Philippines, Thailand, and Mongolia), Ma (Reference Ma2007) discovered that in those countries or regions where authoritarianism is strong, political trust is also high. He thus argued that political leaders and governments under Confucian culture occupy an important symbolic authoritative status in society, and that the worship, sensitivity, and dependence of the people towards authority could easily have a profound effect on their political trust. Moreover, other scholars observed that the transition from materialism to post-materialism seen in many advanced countries of the West is also appearing in such Asian societies as Taiwan, Hong Kong, and China and beginning to have an impact on political trust in these places (Hsiao and Wan, Reference Hsiao, Wan, Lau, Wong and Wan2004; Wang, Reference Wang2005; Wong and Wan, Reference Wong and Wan2008b). Such findings show that improvements in socio-economic conditions may not necessarily strengthen people's trust in their government and political system. The emergence of post-materialist values could also have an influence on the development of political trust.
The above literature review suggests that political trust is shaped by both institutional and cultural factors, but which approach – institutional or cultural – is more powerful is so far not totally clear. One of the main objectives of this study is to compare the two approaches in explaining the experiences of political trust in Hong Kong and Taiwan. Our guiding question is whether government performance and cultural orientations, along with life satisfaction and satisfaction with the conditions of democratic rights and freedoms, in Hong Kong and Taiwan have affected the political trust of the people of these two societies and to what extent? We particularly add the two items ‘life satisfaction’ and ‘satisfaction with the condition of democratic rights and freedoms’ to our observation mainly out of the following consideration. In modern societies, quality of life is the principal concern of common people and democratic rights and freedoms are increasingly becoming universal values. Hence, whether or not the people are satisfied with the state of these two items in their country could have an impact on their trust in their government. This is because the government represents the people in governing and has the responsibility to maintain its people's quality of life and safeguard their democratic rights and freedoms. In this regard, these two items should be seen as an extension of the institutional argument and can be rationally evaluated by citizens.
Data and measures
Source of data
The data used in this study came from the 2006 AsiaBarometer survey project led by Professor Takashi Inoguchi. The survey project covered seven countries or regions within the Chinese character-using zone: namely, Hong Kong, Taiwan, mainland China, Japan, South Korea, Singapore, and Vietnam. The participants in the Hong Kong and Taiwan surveys consisted of adults aged 20–69. The sample was a probability sample. The survey employed a structured questionnaire, and data were collected through face-to-face household interviews. The Hong Kong survey was conducted from 11 July to 3 August 2006, and a total of 1,000 people were successfully interviewed. The Taiwan survey was conducted from 3 July to 11 August 2006, with a successful sample of 1,006 respondents.Footnote 3
Measurement of variables
(1) Political trust: Since the power of modern governments normally operates through the executive, legislative, and judicial branches, we also divided political trust into three indicators: trust in the government, trust in the judiciary, and trust in the legislature. The response categories were coded on a four-point scale ranging from 1 = greatly distrust, 2 = distrust, 3 = trust, and 4 = greatly trust. The higher the score is, the higher the level of political trust.
(2) Evaluation of government performance: Government administration covers different domains. Therefore, we divided government performance into eight domains for respondents to evaluate: dealing with the economy, dealing with problems of political corruption, dealing with human rights problems, dealing with the problem of unemployment, dealing with the problem of crime, dealing with public services problems, dealing with ethnic conflicts, and dealing with environmental problems. The response categories were coded on a four-point scale, ranging from 1 = very poor, 2 = poor, 3 = good, and 4 = very good. The higher the score is, the more favourable the evaluation.
(3) Life satisfaction: Since life satisfaction encompasses different facets, we chose seven domains of life by which to measure the respondents' satisfaction: housing, household income, health, education, public security, social welfare, and the democratic system. The response categories were measured on a five-point scale, ranging from 1 = very dissatisfied, 2 = dissatisfied, 3 = average, 4 = satisfied, and 5 = very satisfied. The higher the score is, the greater the degree of satisfaction.
(4) Satisfaction with the condition of democratic rights and freedoms: A total of six indicators were used to measure the respondents' satisfaction with the conditions of democratic rights and freedoms: the right to vote, the right to participate in any kind of organization, the right to gather and demonstrate, the right to be informed about the work and functions of the government, freedom of speech, and the right to criticize the government. The response categories were coded on a four-point scale, ranging from 1 = very dissatisfied, 2 = dissatisfied, 3 = satisfied, and 4 = very satisfied. The higher the score is, the greater the degree of satisfaction.
(5) Cultural orientations: This mainly measures the possible influence of three cultural factors on political trust, namely, interpersonal trust, post-materialism, and traditionalism. First of all, we measured interpersonal trust by directly asking the respondents the following question: ‘Generally do you think most people can be trusted or do you think you can't be too careful in dealing with people?’ The response categories were coded on a two-point scale, ranging from 1 = you can't be too careful in dealing with people, and 2 = most people can be trusted. As to the measurement of post-materialism, Inglehart's four-choice measure was used to gauge an individual's value priorities.Footnote 4 The respondents were asked to answer the following question: ‘If you have to choose, which one of the following would you say is the most important? And which would be the second most important? (1) maintaining order in the nation; (2) giving people more say in important government decisions; (3) fighting rising prices; and (4) protecting freedom of speech.’ This question yielded relative rankings for four societal goals that reflect materialist and post-materialist needs. Two of these goals measure materialist needs. Specifically, ‘fighting rising prices’ involves the need for sustenance, and ‘maintaining order in the nation’ involves the need for safety. The remaining two goals assess post-materialist needs. Specifically, ‘having more say in government’ involves the need for belonging and esteem, and ‘protecting free speech’ involves intellectual needs. Based on the results of these value priorities, the respondents were then divided into the following three value types: 1 = the materialist, 2 = the mixed, and 3 = the post-materialist.Footnote 5 As for measuring traditionalism, there were two indicators: the respondents were asked for their views on the following statements: ‘The traditional culture of our country is superior to that of other countries’, and ‘Greater respect for traditional authority is a good thing or a bad thing’. The feeling of superiority with regard to traditional culture was coded on a five-point scale, ranging from 1 = very much disagree, 2 = disagree, 3 = average, 4 = agree, and 5 = very much agree. The higher the score is, the stronger the feeling of superiority. Respect for traditional authority was coded on a three-point scale, ranging from 1 = a bad thing, 2 = don't mind, and 3 = a good thing. The higher the score is, the greater the respect for traditional authority.
Level of political trust
Table 1 lists the level of trust that the Hong Kong and Taiwan people have in their government, judiciary, and legislature. First, the people of Hong Kong have a good level of trust in their government and judiciary, with those expressing trust exceeding 55% and those expressing a lack of trust amounting to less than 40%. However, they regard their legislature with less trust, with 45.7% expressing trust – slightly less than the 49.7% who expressed a lack of trust. As for Taiwan, the proportion of the people that expressed trust in their government, judiciary, and legislature was 35.0%, 32.4%, and 21.4%, respectively. Not only are these percentages very much lower than the respective percentages of those who expressed distrust, at 62.9%, 63.7%, and 73.9%, but the Taiwan figures on trust in the three power bodies are more than 20 percentage points lower than those for Hong Kong. This more or less reflects the existence of serious problems in political trust in Taiwan brought about by the collective feeling of setback following the establishment of democracy there, the most serious of which concerns the legislature. With regard to the scale means of the indicators for political trust for the two places, for Hong Kong not only are two indicators (trust in the government and trust in the judiciary) higher than the scale mid-point (2.5), all of the mean values of the three indicators are also 0.5 point higher than the related mean values in Taiwan.Footnote 6 In other words, Hong Kong's system of partial democracy has not particularly posed a challenge to the Hong Kong people's political trust, while Taiwan's system of full democracy has not caused the Taiwan people to have greater trust in their executive, judicial, and legislative bodies. Clearly, an explanation for the huge difference in political trust between the two societies must be sought elsewhere.
Table 1. Political trust
Note: *** p < 0.001.
Socio-cultural orientations and feelings
Evaluation of government performance
On the eight indicators of government performance – economic development, anti-political corruption, protection of human rights, anti-unemployment, public security, public services, ethnic relations, and environmental protection – the evaluations of the Hong Kong respondents are more positive than those of their Taiwan counterparts. Moreover, except for the item on dealing with environmental problems, there is a statistically significant difference in the scale mean of the seven other items. This shows that, on the whole, the Hong Kong people give the performance of their government a higher rating than the Taiwan people do theirs. In fact, of the eight indicators of government performance, the majority of Hong Kong people gave a good rating to seven (with a scale mean of between 2.4 and 2.8 points). Only the item on dealing with the problem of unemployment did not live up to the hopes of Hong Kong residents (only 35.1% gave a rating of good and the scale mean was only 2.2 points). As to Taiwan, of the eight indicators, the majority of the people gave a negative rating to six (with a scale mean of between 1.7 and 2.4 points). Of these, the following four indicators received negative ratings from over 80% of the respondents: dealing with problems of political corruption (90.7%), dealing with the economy (87.3%), dealing with the problem of crime (86.9%), and dealing with the problem of unemployment (84.9%). Apparently, there is a huge disparity between how the people of Hong Kong and Taiwan rate the performance of their government (Table 2).
Table 2. Evaluation of government performance
Note: *** p < 0.001.
Life satisfaction
The level of satisfaction of Hong Kong and Taiwan people with the seven indicators of living conditions is not too low. In Hong Kong, on all of the indicators, the total proportion of people who expressed satisfaction exceeded the total proportion of those who indicated a lack of satisfaction. The Taiwan people displayed similar sentiments on five items; only with the indicators of public security and social welfare did the proportion of those expressing dissatisfaction exceed the proportion expressing satisfaction. When the scale means are examined, all of the seven indicators for Hong Kong are higher than the scale mid-point (3.0) by 0.2 to 0.6 points, while for Taiwan, only the indicators of public security (2.5 points) and social welfare (2.8 points) are below the scale mid-point. One can say that Hong Kong and Taiwan people are about similarly satisfied with their quality of life, because among the seven indicators, there were three items for which the scale mean was significantly higher in Hong Kong than in Taiwan, and vice versa (Hong Kong exceeded Taiwan in public security, social welfare, and the democratic system, whereas Taiwan exceeded Hong Kong in housing, household income, and education). On the remaining item, health, the two places were tied (Table 3).
Table 3. Life satisfaction
Notes: * p < 0.05, ** p < 0.01, *** p < 0.001.
Satisfaction with the condition of democratic rights and freedoms
Hong Kong and Taiwan people are quite satisfied with the condition of democratic rights and freedoms, which are regarded as universal values. For all six indicators, the proportion of those in Hong Kong who felt satisfied exceeded 50% (for a scale mean of between 2.5 and 3.0 points). For five of the indicators, the figure was over 70% (namely, the right to vote, the right to participate in organizations, the right to gather and demonstrate, freedom of speech, and the right to criticize the government). In Taiwan, there were five indicators for which the proportion of those who felt satisfied exceeded 50% (for a scale mean of between 2.7 and 3.1 points). There were also three indicators for which the figure was over 70% (namely, the right to vote, the right to participate in organizations, and freedom of speech). The only item for which the figure was less than 50% was the right to government information, at 44.6% (for a scale mean of 2.4 points). If the two societies are compared with regard to their scale means of the six indicators, it is not difficult to see that there are three indicators for which the values are significantly higher in Hong Kong than Taiwan (the right to participate in organizations, the right to gather and demonstrate, and the right to government information), and two indicators for which the scale mean is similar in the two places (freedom of speech and the right to criticize the government). This shows that, with regard to the condition of democratic rights and freedoms, the overall level of satisfaction of the Hong Kong people is slightly higher than that of the Taiwan people (Table 4).
Table 4. Satisfaction with the condition of democratic rights and freedoms
Note: *** p < 0.001.
Cultural orientations
As mentioned above, our observation of cultural orientations mainly concentrated on the respondents' view of interpersonal trust and attitudes towards post-materialism and traditionalism. First of all, Table 5 shows that both the people of Hong Kong and Taiwan do not give a high rating to the level of interpersonal trust in their society. In Hong Kong, only 33.5% of respondents thought most people could be trusted. The figure in Taiwan was somewhat higher, at 39.6%, and the difference between the two places was statistically significant. As to post-materialist orientation, Hong Kong people are apparently more post-materialist than their Taiwanese counterparts. According to our classification in Table 5,Footnote 7 10.4% of Hong Kong people belonged to the post-materialist type, whereas only 2.6% of Taiwan respondents shared a similar orientation. In addition, 71.7% of Hong Kong people can be classified as belonging to the mixed type, compared to Taiwan's 59.6%. On the other hand, 37.8% of Taiwan people showed a materialist orientation, while only 17.9% of Hong Kong people fell into this category. The post-materialism scale mean of Hong Kong (1.9) is also significantly higher than that of Taiwan (1.6), though both are below the mid-point of 2. With regard to traditionalism, Taiwan people have a stronger feeling for the superiority of traditional culture and a greater respect for traditional authority (the respective scale means were 3.4 and 2.4 points) than Hong Kong people (the respective scale means were 3.0 and 2.2 points). The related differences were also statistically significant – over half of Taiwan people agreed that their traditional culture is superior to that of other countries and that traditional authority should be given greater respect. However, less than 35% of Hong Kong people held the same views. Regarding the scale means of the two indicators of traditionalism, in both societies these were higher than or equal to their scale mid-points (the scale mid-point for respect for traditional authority was 2, while the scale mid-point for the superiority of traditional culture was 3). However, Taiwan's scale means were all higher than those of Hong Kong. This shows that among Taiwan people, traditionalism is much more widespread than among Hong Kong people (Table 5).
Table 5. Cultural orientations
Notes: ** p < 0.01, *** p < 0.001.
In sum, not only do the Hong Kong people show a higher level of political trust than the Taiwan people, they are also generally more positive in their evaluations of government performance and the condition of democratic rights and freedoms than the Taiwan people. As for their degree of life satisfaction, the difference between the people of the two societies is not large. However, compared to the Taiwanese, culturally Hong Kong people have a slightly higher level of interpersonal trust, a stronger post-materialist orientation, and a weaker traditionalist tendency. What is worth asking again is whether or not the obvious differences in political trust between the Hong Kong and Taiwan people are related to their above-mentioned evaluations of government performance, life satisfaction, satisfaction with the condition of democratic rights and freedoms, and cultural orientations. If they are related, what is the nature of such relationships? Which approach is superior in explaining the experiences of Hong Kong: the institutional or the cultural?
Determinants of political trust in Hong Kong and Taiwan
In order to examine the independent effect on political trust of the evaluations of government performance, life satisfaction, satisfaction with the condition of democratic rights and freedoms, and cultural orientations, we conducted ordinary least square (OLS) regressions by having the first three indicators of political trust as dependent variables, and evaluations of government performance, life satisfaction, satisfaction with the condition of democratic rights and freedoms, and cultural orientations as independent variables. Other independent variables included such personal socio-demographic attributes as gender, age, educational attainment, and household income. A few more words of explanation are required for the construction of the regression models. As mentioned earlier in our citation of the findings of the study of Gibson et al. (Reference Gibson, Caldeira and Spence2005), the people's attitudes towards the judiciary and the rule of law will also affect their trust in the government. For this reason, apart from the three regression models on the three indicators of political trust – the executive, judicial, and legislative branches – a separate model taking the people's trust in the judiciary and legislature as independent variables was also constructed to observe their possible influence on the people's trust in the government. The results of the regression analyses are listed in Tables 6 and 7. Model 2 under trust in the government is based on the inclusion of trust in the judiciary and legislature as independent variables.
Table 6. Standardized regression coefficients on political trust in Hong Kong
Notes: * p < 0.05, ** p < 0.01, *** p < 0.001.
Table 7. Standardized regression coefficients on political trust in Taiwan
Notes: * p < 0.05, ** p < 0.01, *** p < 0.001.
Hong Kong
Several observations can be made from Table 6. First, of the indicators with a significant independent effect on the level of trust in Hong Kong's judiciary, there were a total of six items and all were institutional or institutionally related: three items on government performance (dealing with the problem of crime, dealing with public services problems, and dealing with ethnic conflicts), one item on life satisfaction (housing), and two items on satisfaction with the condition of democratic rights and freedoms (the right to vote and the right to criticize the government).Footnote 8 However, all of the items on cultural orientations and personal socio-demographic attributes had no significant independent effect on the level of trust in the judiciary. In terms of importance, the evaluation of the government's performance in dealing with the problem of crime appears to be relatively outstanding (its standardized regression coefficient was the largest, at 0.160), followed by dealing with public services problems and dealing with ethnic conflicts (the respective standardized regression coefficients were 0.130 and 0.108). Since these three indicators are not only related to government performance but also touch upon important social values or principles such as the rule of law (dealing with the problem of crime), fairness and openness (dealing with public services problems), and equality and justice between the different social groups (dealing with ethnic conflicts), their relative importance as independent variables more or less reflects that, when it comes to deciding how much they trust the judiciary, Hong Kong people tend to place a greater emphasis on whether or not the government effectively upholds these values or principles.
Second, a total of five indicators had a significant effect on the level of trust in Hong Kong's legislature: one item on government performance (dealing with ethnic conflicts), two items on life satisfaction (education and public security), and two items on cultural orientations (interpersonal trust and the superiority of traditional culture). All of the items on satisfaction with the condition of democratic rights and freedoms and personal socio-demographic attributes had no significant effect on the level of trust in the legislature. Of the five items that showed a significant effect, none could be considered very important to trust in the legislature, since the largest standardized regression coefficient (satisfaction with education) only had a value of 0.113, showing that the related indicators had a mild effect on the level of trust in Hong Kong's legislature. It is noteworthy that interpersonal trust had a negative relationship with that in the legislature; this finding unfortunately goes against the cultural explanation of political trust (Putnam, Reference Putnam1995).
Third, there were six indicators with a significant effect on the level of trust in the Hong Kong government: two items on government performance (dealing with the economy and dealing with ethnic conflicts), one item on satisfaction with the condition of democratic rights and freedoms (the right to gather and demonstrate), one item on cultural orientation (post-materialism), and two personal socio-demographic attributes (gender and age).Footnote 9 However, when the levels of trust in the judiciary and the legislature are included in the regression model (Model 2), the impact of dealing with ethnic conflicts as well as with the right to gather and demonstrate are considerably attenuated and become statistically insignificant. By contrast, the right to participate in organizations and the level of trust in the judiciary showed a significant effect, among which the standardized regression coefficient of the level of trust in the judiciary was relatively large (at 0.177). This shows that apart from economic performance, maintaining a high level of trust in the judiciary is very important for strengthening the trust that Hong Kong people have in their government.
Fourth, if the values of the adjusted R2, which reflect the explanatory power of the four regression models (including Models 1 and 2 of the level of trust in the government), are compared, we can see that the explanatory power of all of the four models cannot be considered outstanding. Among these four models, the best one is for the level of trust in the judiciary, and its adjusted R2 value was only 0.168. Next is the model for the level of trust in the legislature, and its adjusted R2 value dropped to 0.112. As for the two models for the level of trust in the government, the values of their adjusted R2 were both smaller than 0.1. In short, both institutional and cultural factors show varying degrees of effect on the Hong Kong people's level of trust in the executive, legislative, and judicial branches of their government, but such effects are nevertheless mild and isolated. In comparative terms, the institutional factors appear to be more powerful than the cultural factors in explaining Hong Kong people's political trust, especially if we consider their trust in the judicial branch to be an institutional phenomenon.
Taiwan
Table 7 shows that the indicators affecting the political trust of Taiwanese are not quite the same as those of Hong Kong. First, there were a total of seven indicators with a significant effect on the level of trust in Taiwan's judiciary, namely: three items on government performance (dealing with the problem of unemployment, dealing with the problem of crime, and dealing with ethnic conflicts), one item on life satisfaction (public security), and three items on satisfaction with the condition of democratic rights and freedoms (the right to vote, the right to government information, and freedom of speech). However, all of the indicators of cultural orientations and the four personal socio-demographic attributes showed no significant effect on the level of trust in the judiciary. In terms of importance, four items (government performance in dealing with the problems of unemployment and crime, satisfaction with the right to vote and the right to government information) had a relatively great effect on the dependent variable, with respective standardized regression coefficients of 0.144, 0.120, 0.105, and 0.135. The standardized regression coefficients of the remaining three indicators that showed an independent effect were all lower than 0.1. In other words, when evaluating the trustworthiness of their judiciary, the Taiwan people apparently place a greater emphasis on certain aspects of institutional performance – whether or not the government has effectively upheld some basic rights and interests of the people (fighting unemployment and crime, and protecting the right to vote and the right to government information). The cultural factors of interpersonal trust, post-materialism, and traditionalism seem to be totally unrelated to the Taiwan public's trust in the judiciary.
Second, only three indicators had a significant effect on the level of trust in Taiwan's legislature: one item on government performance (dealing with problems of political corruption), and two items on satisfaction with the condition of democratic rights and freedoms (the right to participate in organizations and the right to government information). The rest of the indicators of life satisfaction, cultural orientations, and the four personal socio-demographic attributes showed no significant effect on the level of trust in the legislature. Of the three indicators that showed a significant effect, the government's performance in dealing with problems of political corruption had the greatest impact on the level of trust in the legislature, with a standardized regression coefficient of 0.187. The standardized regression coefficients of the other two indicators were both lower than 0.1. In other words, when the Taiwan people come to decide the degree to which their legislature can be trusted, the institutional factor of handling the much-denounced problem of political corruption is indeed their important reference, whereas the cultural factors of interpersonal trust, post-materialism, and traditionalism all play no significant role in the process.
Third, there are many indicators with an independent effect on the level of trust in the Taiwan government, and these indicators include three items on government performance (dealing with the economy, dealing with human rights problems, and dealing with ethnic conflicts), two items on life satisfaction (health and education), three items on satisfaction with the condition of democratic rights and freedoms (the right to vote, the right to government information, and the right to criticize the government), and two items on cultural orientations (the superiority of traditional culture and respect for traditional authority). None of the four personal socio-demographic attributes showed a significant effect on the level of trust in the government. However, when the levels of trust in the judiciary and the legislature are taken together (Model 2), the effect of the following indicators are substantially attenuated and become statistically insignificant: dealing with ethnic conflicts, satisfaction with health and education, the right to government information, and the superiority of traditional culture.Footnote 10 By contrast, both the level of trust in the judiciary and the level of trust in the legislature showed a significant effect, and their standardized regression coefficients were 0.235 and 0.220, respectively – far higher than the other indicators that were statistically significant. This to a large extent shows that promoting trust among the Taiwan people in their judiciary and legislature is of great help in raising their trust in the government. In fact, if the values of the adjusted R2 of Models 1 and 2 are compared, we can see that the former was only 0.199 while the latter was 0.309. This further proves that including the levels of trust in the judiciary and the legislature as independent variables will certainly greatly strengthen the explanatory power of the regression model on the level of trust in the government. In addition, among the other indicators in Model 2 that show a significant effect, the standardized regression coefficient of the item dealing with the economy is also quite prominent, reaching 0.178. This basically demonstrates that economic performance is similarly an important variable affecting the Taiwan people's trust in their government that we should not overlook. Besides, since the levels of trust in the judiciary and the legislature and the item dealing with the economy are institutionally related and their standardized regression coefficients are all much greater than those of the significant cultural factors in Models 1 and 2, it is not difficult to conclude that institutional performance is much more important than the cultural approach in accounting for Taiwan's people trust in the government.
Fourth, among the four regression models, the adjusted R2 value of Model 2 for the level of trust in the government is the greatest. This again shows that the levels of trust in the judiciary and the legislature have important effects on the degree of trust that the Taiwan people have in their government. If the influences of these two variables are not taken into account, and only the other three models are compared, the indicators of government performance, life satisfaction, satisfaction with the condition of democratic rights and freedoms, and cultural orientations showed a larger effect on the level of trust in the judiciary and the level of trust in the government, because the values of the adjusted R2 of these two models reached 0.182 and 0.199, respectively. In contrast, the explanatory power of these variables on the level of trust in the legislature was smaller, since its adjusted R2 value was only 0.099. Moreover, since the cultural factors play no significant role in the Taiwan people's trust in the judiciary and the legislature and only a relatively minor role in their trust in the government, the institutional approach is therefore more powerful than the cultural approach in explaining political trust in Taiwan as a whole.
Comparison of the two societies
When we go one step further to compare Hong Kong and Taiwan, we can easily see that the overall effect on the political trust of Taiwan people of government performance, life satisfaction, satisfaction with the condition of democratic rights and freedoms, and cultural orientations is generally stronger. However, with regard to the importance of individual indicators, there are differences between the two societies. First, with regard to trust in the judiciary, it appears that Hong Kong people place relatively more emphasis on the government's performance in dealing with problems of crime, while the focus of the Taiwan people is more on the government's performance in dealing with the problem of unemployment. Second, all of the related indicators had a very mild effect on the Hong Kong people's trust in their legislature, but in Taiwan the evaluation of the government's handling of the problem of political corruption turned out to be a very important factor shaping the people's trust in their legislature. Third, the government's performance in dealing with the economy had only a moderate effect on the Hong Kong people's trust in the government, whereas the impact of this issue on the Taiwan people is striking. Fourth, the levels of trust in the judiciary and the legislature had a very noticeable effect on the Taiwan people's trust in their government. However, in Hong Kong, such an effect was confined to trust in the judiciary, and the strength of the effect was not as great as in Taiwan.Footnote 11 Fifth, in both societies, the effect of institutional performance on political trust is obviously more outstanding and more pervasive than that of cultural orientations. Yet, individual cultural factors still have a minor but observable role to play in the formation of political trust in the two societies. For example, post-materialism shows a negative relationship with the trust in the government in Hong Kong, but this pattern is not found in Taiwan. On the other hand, traditionalism (especially respect for traditional authority) exercises a positive effect on Taiwan people's trust in the government, while such an effect is absent in Hong Kong.
Conclusion and discussion
Trust is an important cornerstone for the survival and development of any regime. Both democratic and non-democratic systems must face up to this social-political-psychological trend. This paper used Western and Asian studies as its basis, as well as data from a cross-national social survey, to examine the level of political trust held by the Hong Kong and Taiwan people and the possible effect on political trust of a series of institutional and cultural factors – government performance, life satisfaction, satisfaction with the condition of democratic rights and freedoms, and such cultural orientations as interpersonal trust, post-materialism, and traditionalism, with a view to finding out whether the institutional approach or the cultural approach is more powerful in explaining political trust in the two societies. A review of the related literature has shown that these two approaches compete as explanations for the origins of political trust. Hong Kong and Taiwan are Chinese societies with similar socio-economic conditions but vastly different political systems. The former still has a non-democratic or partially democratic system, while the latter has implemented a full democracy for some years. A comparison of the level of political trust in these two societies and the underlying determining factors should help us understand how different political systems with similar socio-economic conditions may differ in their political trust and whether institutional factors or cultural factors are more important in accounting for the political trust in these two Chinese societies.
With regard to political trust, we found that Hong Kong people have a high degree of trust in their government and judiciary but a relatively low level of trust in their legislature. As for Taiwan, the proportions of those trusting the government, judiciary, and legislature were all much lower than those in Hong Kong. This shows that there exists a serious problem with political trust under Taiwan's democratic system, and its legislature is the branch that is the least trusted. Clearly, Hong Kong's system of partial democracy has not particularly shaken the political trust of Hong Kong people, while Taiwan's democratic political system has not caused the Taiwan people to have greater trust in their executive, judicial, and legislative bodies. The huge disparity in political trust between the two societies likely has other roots. In fact, our study shows that Hong Kong appears to possess a relatively large number of positive institutional conditions for the development of political trust: the Hong Kong people generally give a more favourable evaluation of government performance and of the condition of democratic rights and freedoms than Taiwan people do. However, culturally Hong Kong people have a stronger post-materialist orientation but a weaker traditionalist tendency compared to the Taiwanese, both orientations of which are found to be unfavourable to the development of political trust.
Our regression analyses reveal that evaluation of government performance, life satisfaction, satisfaction with the condition of democratic rights and freedoms, and cultural orientations such as interpersonal trust, post-materialism, and traditionalism do have varying degrees of effect on the political trust of the Hong Kong and Taiwan people. In comparative terms, the overall effect of these variables on the political trust of Taiwan people is generally stronger than on the Hong Kong people, but regarding the importance of individual indicators, there are differences between the two places. Such differences include: (1) With regard to trust in the judiciary, the Hong Kong people emphasize the government's performance in dealing with the problem of crime, but the Taiwan people place greater importance on the government's performance in dealing with the problem of unemployment. (2) All of the indicators had a very mild effect on the Hong Kong people's trust in the legislature, but the government's performance in the handling of the problem of political corruption had a very clear impact on the Taiwan people's trust in the legislature. (3) The government's performance in handling economic problems had only an average effect on the Hong Kong people's trust in their government, but the influence on the Taiwan people was striking. (4) The levels of trust in the judiciary and the legislature had a very noticeable effect on the Taiwan people's trust in the government, but in Hong Kong, the effect mainly came from the trust in the judiciary, and the strength of the effect was not as great as in Taiwan. (5) The effect of institutional performance on political trust was obviously more pronounced than that of cultural orientations in both Hong Kong and Taiwan, but individual cultural factors still had a role to play in the formation of political trust in the two societies. In Hong Kong, post-materialism appeared to constrain people's trust in the government, while in Taiwan, traditionalism showed a positive effect on the same item.
A number of implications can be derived from the above findings. First, political trust has complex connotations and different aspects. Even if it is simplified into the categories of trust in the executive, judicial, and legislative bodies of government, the people's trust in these three bodies could possibly rise and fall, differ greatly, and even lead to divergent developments. In this regard, analysing them separately should yield more policy and theoretical insights.
Second, as previous studies have shown, the political system is not necessarily the only determining factor of political trust. Hong Kong, which has a system of only partial democracy, enjoys a level of political trust that is not low. Taiwan, meanwhile, practises full democracy, but political trust there is low, a fact that may provide a new footnote for this thesis. This phenomenon, of course, may be explained from the angle of ‘critical citizenship’, which could be the cause of the Taiwan people's increasingly strident demands; or from the angle of the collective mood of pragmatism in Hong Kong that followed the 1997 handover. However, before we do this, we still can look to other factors to find clues to understanding this phenomenon, both institutional and cultural.
Third, the experiences of Hong Kong and Taiwan tell us that government performance, life satisfaction, satisfaction with the condition of democratic rights and freedoms, and cultural orientations such as interpersonal trust, post-materialism, and traditionalism indeed have differing degrees of effect on the political trust of the Hong Kong and Taiwan people. But what must be emphasized is that the experiences of Hong Kong and Taiwan also indicate that different factors or indicators can be of various degrees of importance to the level of trust in the political bodies of different societies. Hence, in order to be able to clearly grasp the significance of these factors or indicators, one cannot speak in general terms but must penetrate into the political and economic circumstances of each society.
Fourth, with regard to Hong Kong, government performance, life satisfaction, satisfaction with the condition of democratic rights and interest, and cultural orientations had only a moderate overall effect on the people's trust in the government. Thus, given the relatively high level of political trust in Hong Kong, particularly in the executive and judicial bodies of the government, we should further ponder the deeper implications of the determining factors revealed in our regression analyses.Footnote 12 We should also add to our understanding by looking at other aspects, for example, at the highly regarded civil service system and the system of rule of law handed down from the British colonial period (Lau, Reference Lau, Lau, Lee, Wan and Wong2003), as well as at the long-term political and economic support that the government of the Hong Kong Special Administrative Region is receiving from Beijing under the unique arrangement of ‘one country, two systems’ (Wong and Wan, Reference Wong and Wan2007). On the other hand, in view of the fact that the trust of the Hong Kong people is relatively high at the government level, in addition to their relatively high level of satisfaction with the condition of democratic rights and freedoms, which touches on the larger political system, the overall political trust in Hong Kong properly belongs to the first mode of political trust under the aforementioned classification of Christensen and Lægreid (Reference Christensen and Lægreid2005): a high level of trust in both the political system and the government's performance. Since Hong Kong's current political system is still one of partial democracy, we are compelled to ask: What kind of political phenomenon is reflected by such a high level of trust in both the political system and the government? What kind of public opinion is involved? Will this phenomenon prove to be an obstacle to the further democratization of Hong Kong's political system? All of these questions are important and worth further study.
Fifth, with regard to Taiwan, the fact that political trust is relatively low under full democracy would seem to provide new support for the ‘critical citizenship’ thesis of Norris (Reference Norris and Norris1999), who has contended that democratic citizens are increasingly critical of the existing authorities. However, before embracing Norris's argument, some important findings from our study should not be treated lightly. These include the impact of the problem of political corruption on trust in the legislature, and the heavy pressures on trust in the government arising from distrust in the judiciary and legislature and the government's poor economic performance. These problems obviously point to the importance of institutional performance as an explanation of political trust and therefore need to be faced up to squarely and dealt with. Otherwise, collective disaffection and a feeling of setback will spontaneously arise. What is gratifying is that our study also shows that satisfaction with the right to vote, a cornerstone of modern democracy, has a positive effect on the people's trust in both the judiciary and the government. This is essentially in line with the findings of some recent studies, which showed that democratization can develop its own process for deepening regime support (Wang et al., Reference Wang, Dalton, Shin, Dalton and Chull Shin2006: 154). Given the fact that the Taiwan people are widely (83%) satisfied with their right to vote, this more or less explains why the relatively low level of political trust in Taiwan is limited to the government, and has not caused the population to fundamentally question their electoral democracy. Instead, the maintenance of electoral democracy will help to restore and even strengthen the political trust of the people. In this regard, Taiwan's political trust would seem to belong to the second mode of political trust under the classification of Christensen and Lægreid (Reference Christensen and Lægreid2005): a high level of trust in the political system but a low level of trust in the government's performance.
Sixth, both the experiences of Hong Kong and Taiwan clearly demonstrate the superiority of the institutional approach over the cultural approach in explaining political trust, and this certainly echoes the findings of some of the previous studies (Mishler and Rose, Reference Mishler and Rose2001). It also sends an important message to governments worldwide about the important potential of institutional performance for nurturing popular trust. Nevertheless, our study shows that cultural factors cannot be totally ignored. First of all, the relationship of post-materialism with political trust found in Hong Kong reminds us again of Norris' ‘critical citizenship’, but such ‘critical citizenship’ is not limited to established democracies but may also exist in undemocratic or only partially democratic regimes such as Hong Kong. Second, the significant effect of traditional values on political trust shown in Taiwan is basically consistent with the findings of many Asian studies (Shi, Reference Shi2001; Ma, Reference Ma2007). This tells us that even though political actors' rational calculations of material interests are important to the development of political trust, such calculations are not necessarily fully rational but may be mediated by different cultural values and orientations (Shi, Reference Shi2001: 401). Cultural changes usually lag behind institutional changes, and therefore there is always a place for studying political trust from a cultural perspective. Finally, it is worth noting that in our study interpersonal trust is either negatively related to political trust (trust in the legislature in Hong Kong) or totally unrelated to it (in Taiwan). This should add new evidence to support the increasing scepticism about the wisdom of extending the argument of interpersonal trust to political trust (Mishler and Rose, Reference Mishler and Rose2001: 35).
In sum, the experiences of Hong Kong and Taiwan offer a new reference for studying political trust, not only in Chinese societies but also around the world. The above findings and implications should not be regarded as the conclusion of our study, but the starting point for the next stage of our research.
About the authors
Timothy Ka-ying Wong received his Ph.D. in sociology at the University of Manitoba, Canada. He is currently Associate Director of the Hong Kong Institute of Asia-Pacific Studies, The Chinese University of Hong Kong. His major research interests include social and political development in Taiwan and Hong Kong and cross-Taiwan-Strait relations. He has published a number of books and has contributed to such journals as Nations and Nationalism, Journal of Contemporary China, Democratization, Asian Perspective, International Journal of Social Welfare, Social Policy and Administration, and Social Indicators Research.
Hsin-Huang Michael Hsiao is the Executive Director of the Center for Asia-Pacific Area Studies (CAPAS) and Research fellow at the Institute of Sociology, both at Academia Sinica, and Professor of Sociology at National Taiwan University, Taipei, Taiwan. He served as a national Policy Advisor to the President of Taiwan between 1996 and 2006. His areas of specialization include civil society and new democracies, the middle class in the Asia-Pacific, sustainable development, and NGO studies. His most recent publications include: Rise of China: Beijing's Strategies and Implications for the Asia-Pacific (co-editor, 2009), Non-Profit Organization: Organization and Practice (co-editor, 2009), Deepening Local Sustainable Development in Taiwan (co-editor, 2008), Asia-Pacific Peace Watch (co-editor, 2008).
Po-san Wan is a research officer at the Hong Kong Institute of Asia-Pacific Studies, The Chinese University of Hong Kong. She has published a number of books and has contributed to such journals as International Journal of Social Welfare, Japanese Journal of Political Science, Journal of Contemporary China, Public Administration and Development, and Social Indicators Research.
Acknowledgements
The authors are grateful to Professor Takashi Inoguchi for granting us the permission to use the 2006 AsiaBarometer Survey dataset. We also thank Wai-wok Yau, Lok-yin Lai, and Tse-wai Wong of the Hong Kong Institute of Asia-Pacific Studies for their assistance in collecting materials and in conducting statistical analyses.
