Introduction
According to Sztompka (1996), “trust is a bet on the future contingent action of others”
(p. 39). Sometimes trust may be treated as “a psychological trait” at the individual
level. However, without doubt, trust can be “shared by a number of individuals” in
a certain society (Sztompka, 1998, p. 20). In this perspective, trust is considered
at the societal level, which includes, but reaches beyond, the attitudes of individuals.
And it is believed that “when there is trust there are increased possibilities for
experience and action” (Luhmann, 2017, p. 8); “a nation's well-being, as well as its
ability to compete,” is conditioned on “a single, pervasive cultural characteristic:
the level of trust inherent in a society” (Fukuyama, 1995, p. 7).
It has been found that the democratic order has a significant trust-generating force
because establishing a set of universal criteria that regulates the institutions of
both government and civil society contributes to generating social trust among citizens
(Sztompka, 1996; Bielefeld, 2006). However, little research has been done to explore
the process and mechanism of trust building in countries that have not set up a democratic
order. To fill the research gap, this study aims to explore the trust building process
among nonprofit organizations (NPOs), the government and service users in a changing
China.
Moral Resources, Political Capital, and Trust-Building
The Chinese government has cautiously welcomed NPOs participating in the area of social
service delivery since the market-oriented reforms launched in 1978 (Xu and Ngai,
2011; Xu, 2016). On the one hand, a centralist cultural heritage champions authorities
and collective values. According to the 2020 Edelman Trust Barometer, Chinese people's
trust in the government and NGOs is very high, ranking first and second among 26 global
markets, respectively (Edelman, 2020). On the other hand, misunderstandings or distrust
occur from time to time between the Chinese government and NGOs/NPOs or so-called
“civil society” (Evans, 2010; Zhou, 2011). Under such circumstances there is a need
to understand how the trust-building capacity of NPOs is culturally and politically
bound and how to improve their capacity.
The moral resources and political capital perspective provides valuable information
for grassroots NPOs on how to build trust with the government in China (Xu and Ngai,
2011; Xu, 2013). And previous studies have indicated that the NPOs that are involved
in social service delivery are repositories of moral resources and therefore likely
to advance trust (Xu, 2013). Moral resources refer to the available moral choices
that could be made by any organization. There are two types of moral resources: (1)
self-chosen moral resources-I, which are rooted in Immanuel Kant's (1998) argument
of “What ought I to do?;” and (2) societally recognized moral resources-II, which
follow Adorno's (2000) argument that moral or immoral tropes are socially determined
(Xu and Ngai, 2011; Xu, 2013).
Due to the centralized political tradition, political capital is very important for
NPOs in gaining trust from the government (Xu, 2013). Political capital means the
capital that will improve or enhance the organizations' status, assets or access in
the existing political system. There are two types of political capital: (1) ascribed
political capital-I, which refers to the political status that is conferred upon certain
organizations through historical inheritance, and (2) achieved political capital-II,
which refers to the political resources achieved by the organizations' own efforts
(Blau and Duncan, 1967; Xu and Ngai, 2011).
It was found that possessing the “societally recognized moral resources-II” is very
important for grassroots and foreign organizations which have little ascribed political
capital-I, because moral resources-II may help organizations to gain the trust of
the public and the government, which may enable them to build political capital-II.
In this sense, the NPOs aiming to provide social services—which usually focus on the
common good and therefore possess moral resource-II—have a promising future with regards
to improving trust-building between heterogeneous groups (Xu and Ngai, 2011).
It is worth noting trust relationships are vulnerable as trust can be withdrawn from
objects which have previously been trusted. For example, when “gifts” accepted by
officials or medical doctors secure favors or preferential treatment, both institutional
trust and positional trust may be destroyed by the bribe givers and bribe takers (Sztompka,
1996; Heimer, 2001). In this regard, ethics rules such as non-distribution constraint,
which allows NPOs to make profits but prevents them from distributing them to private
parties, are crucial factors in improving the trustworthiness of NPOs and convincing
an increasingly skeptical public (Hansmann, 2003; Becker, 2018; Vaceková and Plaček,
2020). Therefore, the trust-building process demands a broad and comprehensive perspective
(Becker et al., 2019) and the relationship between NPOs and stakeholders (e.g., the
government and service users) needs to be further investigated.
Revisiting the Dimensions of trust
Trust can be studied at different levels. First, at the individual level, interpersonal
trust can be described as a three-part relation: “A trusts B to do X” (Hardin, 2001,
p. 14). Yet, who is the B, what is the X, and how can a trusting relationship develop?
From a cognitive perspective, scholars believe that trust is grounded in the moral
commitments of the trusted (Messick and Kramer, 2001). In other words, “A trusts B”
because A knows that B has strong moral commitments to live up to certain trust expectations
that A places in B (Messick and Kramer, 2001). But how can A believe that B will follow
the ethical rule? Furthermore, in reality, even if A knows that B is a very honest
person, A might trust B to manage her/his money but not trust B to take care of her/his
baby. An answer to this question is that the truster's belief derives from experience,
which means that B's qualities or previous behaviors convince A to trust B.Moreover,
based on the cognitive assessment, studies have verified that if one's experience
with others (especially in one's early years) has been good and cooperative, then
one tends to trust others (Yamagishi, 2001). In this sense, trust is not only a matter
of knowledge or belief in somebody or something, but also a learning process that
can be measured through behavior (Hardin, 2001; Yamagishi, 2001). Empirical studies
of trust have divided people into those high trusters, who are more likely to trust
strangers, and low trusters, who are likely to distrust others. The concept of generalized
trust refers to the observation that “some people have a greater psychological disposition
to trust than others do,” is developed (Hardin, 2001, p. 15).
Hence, when a behavioral measure is adopted to assess generalized trust, the accounts
of trust go beyond interpersonal relationships and extend into the social realm (such
as the NPOs), because both the behavior and the generalized trust are significantly
related to a particular social context with its norms, its legal environment, and
its local culture (Sztompka, 1998).
Second, at the intra-group level, evidence showed that the generalized trust that
is primarily based on membership networks may facilitate more positive interactions
within certain types of associations. However, it is doubtful that the network-based
trust can be generalized to strangers in the society concerned (Stolle, 2001). Similarly,
many scholars have regarded kin-like relationships as one of the most important social
bases of trust. For instance, Cook and Cook and Hardin (2001) argued that it is “familial,
communal, network, and other contexts” that are grounds for trust in the people we
might trust (p. 330). Particularly, Ensminger (2001) conducted a case study of East
African herders, and found that kin relationships and reputation are significant bases
for trust. In Europe, by analyzing the data of Eurobarometer surveys undertaken during
1980–1996, Mackie (2001) pointed out that Europeans are likely to regard people of
their own country as more trustworthy than the people of other countries. Moreover,
it is found that differing patterns of family formation may have been a significant
basis for the development of trust (Mackie, 2001). However, as Fukuyama (1995) argued,
for instance, although Chinese Confucianism promotes tremendous trust in the family
setting, social trust outside the family is relatively low.
Third, consideration should also be given to the institutional level. Because intra-group
trust based on kinships or association memberships can fail to develop into generalized
trust because of inter-group conflicts of interests, Fukuyama (1995) and Knight (2001)
presumed that formal institutions (such as the state and the law) should provide assurance
to improve the trust across boundaries. Moreover, it was assumed that good governance
implies a mutual trust between citizens and governors and among the fellow citizens
(Levi and Braithwaite, 1998). Thus, this kind of trust is inherently institutional
in nature because any reference to “the fellow citizens” refers to a generic category
of “everyone else” rather than one's “neighbours” (Offe, 1999).
Last but not least, “trusting an institution” means that the citizens are confident
that the institution will continue to operate according to the established rules in
the way that the citizens have known (Offe, 1999). The level of institutional trust
may be positively associated with the level of positional trust, which means the trust
of people because they hold certain positions such as lawyers, teachers, doctors,
social workers or other professionals (Giddens, 1990; Sztompka, 1996).
Conclusion
Based on the discussion above, a research framework for trust-building, which explores
personal, positional, organizational and institutional trust between the NPOs and
stakeholders in China is developed (Figure 1). The key terms and factors are operationally
defined as follows.
Figure 1
A research framework for trust-building.
First, the process of trust-building will be analyzed from a perspective of four dimensions
(Giddens, 1990; Sztompka, 1996):
1) Personal trust, which means the trust between individuals;
2) Positional trust, which means the trust of a person in those in certain positions
such as lawyers, teachers, doctors, social workers or other professionals;
3) Intra-group trust, which means network-based trust, such as kin relationships or
small community-based local organizations;
4) Institutional trust, which means the trust in the institutional systems of a society,
such as the education, medical or judicial systems, and so on.
Second, trust-building, which refers to the process of building trust, “is also a
kind of—socially objectified... cultural capital from which individuals can draw in
their actions” (Sztompka, 1998, p. 20). In contrast to organizations that mainly pursue
their own local political or economic objectives, NPOs committed to social service
projects are likely to gain support from certain communities and are thus repositories
of moral resources, and could be an important basis for building trust with the government
(Fenton et al., 1999; Halfpenny, 2000; Lee et al., 2014; Feng, 2017). In other words,
moral resources and local culture that embody the communities and civil society and
political capital, laws, and/or regulations that are associated with the state jointly
affect the trust among NPOs, the government, and service users.
Particularly in countries where laws and regulations are relatively weak, gaining
political capital would increase governments' trust in NPOs other than social service
organizations (Xu, 2013, 2016). In recent years, NPOs began to use social media (e.g.,
WeChat, QQ, Twitter, etc.) to disseminate information, build engagement, and facilitate
action (Guo and Saxton, 2014; Svensson et al., 2015). These new Information and Communication
Technologies (ICTs) have allowed the NPOs' efforts to be more easily heard and seen
by the public. They have provided new opportunities for various NPOs to get societally
recognized moral resources-II and develop their achieved political capital-II, which
may facilitate the trust-building process between them and the government (Xu, 2014).
In short, research increasingly suggests that trust is a major factor in the success
of effective collaboration among institutions. This study improves theoretical understanding
of key factors that may facilitate, and/or hinder, the building of trust. I hope that
the further research may develop practical suggestions for improving trust-building,
and thus contribute to the welfare and success of NPOs, service users and the Chinese
government.
Author Contributions
The author confirms being the sole contributor of this work and has approved it for
publication.
Conflict of Interest
The author declares that the research was conducted in the absence of any commercial
or financial relationships that could be construed as a potential conflict of interest.